Gas evolution in Ruddlesden–Popper-type intercalation cathodes in all-solid-state fluoride-ion-batteries: implications on battery performance and synthesis of highly oxidized oxyfluorides

Abstract

In this work, we report the gas evolution behavior of two common intercalation-based cathode materials for fluoride-ion-batteries (La₂NiO₄ and La₂CoO₄) by chemical fluorination and electrochemical fluorination, as well as the thermal stability of the electrochemically fluorinated active materials. Variable-temperature X-ray diffraction with coupled mass spectrometry shows that the fluorinated phases of both active materials release oxygen after oxidative fluorination using AgF. Differential electrochemical mass spectrometry revealed that significant CO₂, CO and NO evolution takes place during electrochemical fluorination, but O₂ evolution could not be detected. Consequently, the gas evolution observed does not compromise the long-term stability of the cell performance. Variable-temperature X-ray diffraction of the charged cathode composites reveals that the electrochemically fluorinated products decompose once heated beyond 300 °C, and then undergo similar phase evolution behavior as observed in chemical fluorination. Thus, we conclude that oxygen evolution in RP-type materials on oxidative fluorination requires thermal activation beyond the operational temperature of 170 °C, and that it can be kinetically suppressed regardless of the reduced Fermi level induced on fluorination. Our findings indicate that concepts used for the oxidative topochemical fluorination of materials need to be reconsidered: the choice of fluorination conditions must be carefully adopted to the changes of electronic structure induced and the temperature dependence of the oxygen evolution of the highly oxidized state.

---

1 Introduction

While lithium-ion-batteries (LIBs) owe their commercial success at least partly to their high voltage, cathode materials used within high voltage LIBs show side reactions on charging, where instead of the transition metal being oxidized further, the oxide ions get oxidized. This can take place either in the form of lattice oxygen redox (LOR) which is considered reversible since the oxidized oxygen species still reside in the crystal structure,^1–7 or in the form of oxygen release which is irreversible since the gaseous products leave the solid.^8–15

Anion redox has recently been reported to be a major contributor to reversible capacity in all-solid-state fluoride-ion-batteries (ASS-FIBs).^16–19 Wang et al. first reported reversible lattice oxygen redox for ASS-FIBS for the n = 2 Ruddlesden–Popper (RP) material Sr₃Fe₂O₅F₂ by soft X-ray absorption spectroscopy.¹⁷ More recent publications showed that anion redox chemistry can be enabled for other transition-metal containing RP-oxides¹⁸ or perovskites,²⁰ as well as sulfur redox in oxysulfides¹⁹ and nitrogen redox in Cu₃N¹⁶ demonstrating the formation of oxygen and nitrogen molecules using resonant inelastic X-ray scattering. While oxygen or nitrogen release was briefly mentioned in these studies, no further investigations were undertaken.

Zhang et al. reported oxide–fluoride exchange in the chemical fluorination of La₃Ni₂O₇ using CuF₂ forming a La₃Ni₂O_5.5F_3.5 fluorinated phase, though the oxygen release was not measured directly.²¹ A combination of thermogravimetric analysis with coupled mass spectrometry (TG-MS) was used along with phase quantification by X-ray diffraction to determine the final oxygen content of the fluorinated phase.

In this study, we investigate the gas evolution behavior of two common cathode active materials for ASS-FIBs (La₂NiO₄/La₂CoO₄) by chemical fluorination using AgF observed via in situ XRD with coupled mass spectrometry (XRD-MS) and differential electrochemical mass spectrometry (DEMS). The results show that chemical fluorination at 300 °C leads to significant release of oxygen, while electrochemical fluorination at 170 °C conserves the oxygen content in the material. Heating the electrochemically fluorinated compounds to 300 °C as in chemical fluorination leads to decomposition, showing that the products are metastable and oxygen release was only inhibited kinetically. Thus, our findings show that RP-type cobaltates and nickelates could in principle maintain their lattice oxygen regardless of the low Fermi level introduced on electrochemical fluorination, and that LOR with oxygen release is inhibited at the reduced operation temperature of FIBs.

2 Experimental

2.1 Synthesis

La₂CoO₄ and La₂NiO₄ samples for chemical fluorination with AgF were synthesized from precursors obtained by a citric acid-assisted combustion method. Stochiometric amounts of La₂O₃ (ChemPUR, 99.99%; dried at 900 °C for 10 h), and Ni powder (Sigma-Aldrich, 99.99%) or Co(NO₃)₃·6H₂O (Sigma-Aldrich, 98%) were dissolved in approx. 25 mL of distilled water with the addition of a few drops of concentrated HNO₃. Citric acid (Anhydrous, Grüssing, 99%; molar ratio metal ions : citric acid = 1 : 3) was added while stirring. The resulting clear solution was subsequently dried on a hot plate at 120 °C until gel formation. The gels were further heated at 350 °C until ignition. The resulting powders were heated in air at 650 °C for 10 h to ensure complete removal of the organic matrix. La₂NiO₄ was afterwards obtained from calcinating this precursor at 1050 °C for 10 h in air. La₂CoO₄ was obtained after calcinating the precursor at 1300 °C for 12 h in an atmosphere of flowing Argon with a purity of 99.999% (150 mL min⁻¹).

Before mixing with AgF, both oxides were annealed at 300 °C for 12 h in flowing N₂, subsequently put under high vacuum conditions for 12 h before storing them in an Ar filled glove box. Oxide and AgF (Sigma-Aldrich, 99.9%) mixtures in molar ratios of oxide 1 : 2 AgF were prepared by hand-grinding stochiometric amounts in the glove box.

About 10 mg of these mixtures were filled in capillaries (d_ext. = 0.8 mm) which were equipped with the self-made in situ XRD-MS adapter head (vide infra) and temporarily sealed with wax before retrieving them from the glove box.

For electrochemical fluorination, La₂NiO₄ was synthesized by heating a hand-ground mixture of La₂O₃ (Alfa Aesar, 99.9%, pre dried at 1200 °C for 12) and NiO (Thermo Fisher, 99%) to 1200 °C for 8 hours in air with an intermediate regrind.²² La₂CoO₄ was synthesized by ball-milling La₂O₃ and Co₃O₄ (Thermo Fisher, 99.7%) in isopropanol for 1 h at 600 rpm and subsequently heating to 1300 °C in argon flow.²³

La_0.9Ba_0.1F_2.9 was synthesized by ball milling LaF₃ (Strem, 99.9%) and BaF₂ (Strem, 99%) for 12 h in 50 mL ZrO₂ jars with 10 ZrO₂ balls of 10 mm diameter in argon atmosphere cathode composites of La_0.9Ba_0.1F_2.9, La₂NiO₄ or La₂CoO₄ and conductive carbon (Super P, Alfa Aesar, 99+%) were prepared by ball-milling 60 wt% La_0.9Ba_0.1F_2.9, 30 wt% active material and 10 wt% Super P for 2 h at 250 rpm in the same ball-milling jars and balls as mentioned above under argon atmosphere. A composite without active material was made by ball milling 90 wt% La_0.9Ba_0.1F_2.9 with 10 wt% Super P at the same parameters. The composites were dried at 170 °C for 15 h in vacuum after ball milling.²⁴

Electrochemical cells of 7.3 mm diameter were prepared by pressing 5 mg of the cathode composite and 10 mg of PbF₂ on either side of a 200 mg layer of the La_0.9Ba_0.1F_2.9 solid electrolyte. These layers were briefly compressed at 230 MPa before a Pb foil current collector was applied to the PbF₂ anode side. The whole cell, including the current collector, was then compacted at 460 MPa for 90 s, before being extracted from the hardened-steel die set.

2.2 Coupled in situ XRD and mass spectrometry

For analyzing the gaseous reaction products of the chemical fluorination with AgF, an in situ XRD-MS setup was used. A scheme of this setup is shown in Fig. 1. This setup consists of a Swagelok cross piece fitting to which the sample capillary can be attached via a stainless-steel ferrule which is glued to the head of the capillary. The gas transfer capillary (heated to 120 °C) of a mass spectrometer (Pfeiffer Vacuum OmniStar GSD 350) is then inserted into the sample capillary from the opposite port in such a way that it is situated approx. 15 mm above the sample surface. The carrier gas is flowing over the opening of the capillary through the two other ports, allowing control of the gas atmosphere.

Fig. 1 Cross-section rendering of the XRD-MS capillary setup used for the analysis of gaseous reaction products.

Connecting the sample capillary to the Swagelok fitting was performed under a flow of N₂ gas in order to avoid introducing O₂ and H₂O into the system. After connecting the capillary to the setup, the system was subjected to three purge evac. cycles and the purge gas (N₂) was afterwards allowed to flow through the setup for 12 h before starting the experiment.

To perform the temperature dependent XRD-MS study, the capillary reactor is introduced in a HT1 capillary furnace (STOE) which is used together with a STADI MP transmission powder diffractometer (STOE). This device works with monochromatic Mo-K_α1 radiation and is equipped with one MYTHEN2 1K Si-strip detector module (DECTRIS).

The following parameters were used for the in situ XRD experiments of this work. Diffraction patterns were recorded in the range of 2θ = 8–40° with a datapoint increment of 0.015° for 7 min total acquisition time. The sample capillary was heated stepwise from 50 to 150 °C with steps of 10 °C and 25 °C min⁻¹ and from 150 to 300 °C with steps of 5 °C and a 5 °C min⁻¹ heating rate. A diffraction pattern was recorded after every step. The samples were kept at 300 °C for a further 20 h, and diffraction patterns were recorded repeatedly until the end of the experiment.

2.3 High-temperature DEMS setup

In order to measure gas evolution from ASS-FIBs, a DEMS setup was constructed which allows for heating the cell pellet up to 200 °C. To avoid outgassing, the entire setup is constructed from stainless steel, and the heated zone is well-separated from the Viton O-rings to suppress gas permeation. The O-rings were de-gassed by heating at 100 °C in vacuum for 15 h to further reduce gas permeation through the Viton polymer. For every measurement, all O-rings which seal the sample chamber were exchanged for freshly de-gassed O-rings. Without these precautions, significant quantities of nitrogen and oxygen would outgas into the cell container over the course of one hour leading to oxygen concentrations in the order of 5–10 ppm.

The stainless-steel components which will be heated during operation were heated to ∼400 °C in air before use to establish a sufficient oxide layer to avoid further reaction of gases with the stainless-steel surface. The cell pellet was placed between two stainless steel current collector pins, and a spring was used to contact the pellet. One of the current collector pins was separated from the outer steel tube using a glass tube to prevent a short-circuit. A close-up rendering of the setup with the pellet in its measurement position is shown in Fig. 2.

Fig. 2 Close-up cross-section rendering of the cell pellet in the DEMS setup.

Leybold high vacuum valves were used to allow for closing and opening the cell container to an argon flow. Since the valves used were manual operation valves, they were equipped with 3D-printed couplings and stepper motors which were operated by an Arduino Uno microcontroller to allow automatized operation. While the cell container is closed, a bypass is opened to measure the argon (99.999%) carrier gas for a background signal while the evolved gases accumulate in the cell container. In 1 h intervals, the cell container is then opened while the bypass is closed, to flush all evolved gases towards the mass spectrometer capillary. The resulting concentration peaks from flushing the cell container reflect the gas evolution in the time between flushing cycles. They were integrated over a locally determined baseline for every concentration peak using a python script to obtain the gas evolution values.

The mass spectrometer used for the measurements in this study was a Balzers Prisma QMS200. The gases from the capillary entered a pre-evacuation chamber with a pressure of ∼25 mbar from which they pass through a leak valve into the high vacuum chamber of the mass spectrometer which is operated at an approximate pressure in the order of 10⁻⁷ mbar.

The mass spectrometer was calibrated for oxygen using different calibration gases (50 ppm, 100 ppm and 1000 ppm O₂ in N₂) as well as using air for calibration of CO₂ (420 ppm CO₂ in air). Since the carrier gas for the calibration was nitrogen or air, the ratio r(O₂) = I(O₂)/I(N₂) or r(CO₂) = I(CO₂)/(I(N₂) + I(O₂)) was the calibrated quantity, resulting in a factor of 1.36 and 0.53 to convert from r(O₂) to c(O₂) and r(CO₂) to c(CO₂) respectively. For the other gases, no calibration gas was available so a semiquantitative approach was applied, where the ratio r(gas) = I(gas)/I(Ar) was assumed to reflect its concentration directly. The concentration values of the gases which are only semiquantitative are denoted by a unit of “∼ppm” instead of “ppm” for the calibrated gases. We assume a 30–50% uncertainty for the values of CO and NO, due to the values of the calibration factors for CO₂ and O₂ as given above; regardless of the uncertainty of the absolute value, the trends discussed in this article are not affected due to the linear correlation of the ion current to the concentration.

2.4 Electrochemical measurements

Electrochemical measurements were conducted using a Biologic SP150 single channel potentiostat. A constant current of 10 µA was used for all DEMS measurements which were limited by a cut-off voltage of 2 V vs. Pb/PbF₂.

2.5 X-ray diffraction of cell pellets

X-ray diffraction of cell pellets after extraction from the DEMS setup was conducted using a RIGAKU SmartLab diffractometer using a Cu K_α X-ray tube and a HyPix-3000 detector equipped with a Ni-filter to remove the K_β radiation. Measurements were always conducted in airtight sample holders which were sealed in an argon-filled glovebox.

Variable-temperature X-ray diffraction of the cells charged in the DEMS setup were carried out on an Anton Paar DHS1100 variable temperature stage operated by an Anton Paar CCU100 temperature control system. The cells were measured in dynamic vacuum (∼10⁻² mbar). The pellets were measured between 160 °C and 300 °C in steps of 20 °C, then held at 300 °C for 7 h. A XRD pattern was recorded every 30 minutes. This setup had to be used due to the incompatibility of the cell pellets with capillaries used for the coupled XRD-MS measurements described in Section 2.2. All patterns were recorded scanning with a step size of 0.005° 2θ and a scan speed of 1.5° 2θ min⁻¹.

2.6 Thermogravimetric analysis with coupled mass-spectrometry

Thermogravimetric analysis was carried out on a Netzsch STA 449 with a heating rate of 10 °C min⁻¹. A nitrogen flow of 40 mL min⁻¹ was used to transport the exhaust gases to be analysed by a Balzers Prisma QMS200 mass spectrometer.

3 Results and discussion

3.1 In situ XRD with coupled mass spectrometry

La₂CoO₄ and La₂NiO₄ were fluorinated using two molar equivalents of AgF in the XRD-MS setup as described above. The final products formed can be described by the following general reaction schemewhile our measurements are used in order to understand the complexity of the individual reaction steps underlying this reaction equation.

The variable-temperature X-ray diffraction (VTXRD) patterns of La₂CoO₄ and La₂NiO₄ obtained during fluorination are shown in Fig. 3a and b respectively. The weight fractions of the occurring phases as well as signals for O₂ (m/z = 32) and F (m/z = 19) are additionally shown in Fig. 3c for La₂CoO₄ and Fig. 3d for La₂NiO₄, showing that F₂ release can be neglected in comparison to the evolution and release of O₂.

Fig. 3 Heatmap-plot of the VTXRD patterns of (a) La₂NiO₄ and (b) La₂CoO₄ obtained during fluorination using AgF. The dotted white/red line marks the time and temperature where the starting compound disappeared. Weight fractions obtained by Rietveld refinement of AgF, Ag₂O, Ag; La₂CoO₄ Fmmm and mono LCOF; La₂NiO₄ I4/mmm, ortho LNOF #1, mono LNOF, ortho LNOF #2 and La₂NiO₃F₂ Cccm as well as the ion current for m/z = 32 and m/z = 19 along with the temperature in the in situ XRD-MS fluorination of (c) La₂CoO₄ and (d) La₂NiO₄ using AgF. The dashed grey line marks the time of maximum oxygen release.

Upon heating the reaction mixture to 180 °C, a signal for m/z = 19 can be detected by the mass spectrometer (most likely originating from dissociation of F₂ by the ionization process), which marks the onset of the decomposition of AgF and therefore the start of the reaction. The fluorination of La₂CoO₄ follows a rather straightforward scheme, with a reduction of the weight fraction of the starting La₂CoO₄ Fmmm phase and a parallel increase of the phase fraction of a phase similar to the previously reported²³ fluorinated phase of La₂CoO₄F_1.2 with monoclinic symmetry and strongly elongated c-axis (∼15.2 Å vs. ∼12.7 Å) as described by Nowroozi et al.²³ The pattern of this phase which will be called mono LCOF was fitted by a monoclinic structure model with P2₁/c unit cell symmetry which we could derive from neutron powder diffraction analysis. These samples have the stoichiometry of La₂CoO₃F₃ as derived from chemical analysis and a full structure description will be published elsewhere. The formation of mono LCOF is correlated with oxygen release, as can be seen in the bottom graph in Fig. 3c. A dashed line marks the time of maximum oxygen release, which coincides with the turning point of the weight fractions of La₂CoO₄ and mono LCOF meaning that the rate of phase formation mirrors the oxygen release rate. This observation implies that the mono LCOF is oxygen deficient. Also, since the reaction with two equimolar amounts of AgF only results in the fluorination of ∼70% of La₂CoO₄ and the amount of fluorine lost in the atmosphere is insignificantly small, the product is most likely fluorine rich, making its stoichiometry similar to La₂CoO₃F₃ as mentioned above. Upon reacting La₂CoO₄ with three equivalents of AgF, almost complete conversion to La₂CoO₃F₃ is observed, confirming this assumption (see Fig. S1). The reaction as well as oxygen release are mostly terminated as the maximum temperature of 300 °C is reached.

For La₂NiO₄ a more complicated phase evolution is apparent:

The first phase to appear is an orthorhombically distorted version of the starting La₂NiO₄ I4/mmm oxide. Nowroozi et al.²⁵ as well as Jacobs et al.²⁶ already linked a similar small orthorhombic distortion of tetragonal La₂NiO₄ to small amounts of fluoride content which is why it will be referred to as ortho LNOF #1. This phase reaches its maximum weight fraction after 190 minutes but does not decay completely. At the end of the reaction, ∼10 wt% of ortho LNOF #1 are still present.

A second phase appears shortly after ortho LNOF #1, which is similar to previously reported²⁵ La₂NiO₄F_1.8 with an extended c-lattice parameter of c ∼ 15.2 Å as observed in electrochemical fluorination by Nowroozi et al.²⁵ However, the most prominent signal of this phase (indexed as (113) and marked with an arrow “mono LNOF” in Fig. 3b) splits into two reflections after ∼220 minutes of reaction time, indicating a change to monoclinic unit cell symmetry. Due to the similarity of lattice parameters, especially the increased c-lattice parameter of 15.25 Å, this phase was fitted using the structure of previously reported (ref. 27) La₂NiO₃F₃ wtih C2/c symmetry described by Wissel et al.²⁷ In the following, this phase is denoted with mono LNOF. The oxygen deficiency and fluorine content lower than three F per formula unit is implied from the observed oxygen release simultaneously to the monoclinic distortion and the absence of further non-fluorinated phases. This phase reaches the maximum weight fraction ∼240 minutes of reaction time and decays completely until the end of the experiment.

Signals of a third phase appear at about 200 minutes into the reaction; this phase disappears again under formation of La₂NiO₃F₂ as described by Wissel et al.²² Thus, we assume that this phase likely has a composition between La₂NiO₃F₂ and that of mono LNOF, in agreement with the structural behaviour observed (decrease of c-axis, strong distortion of the a/b-plane). It was fitted using the structure of the phase “orthorhombic #3” from ref. 27 and will be referred to as ortho LNOF #2. A small but freestanding reflection at Q = 1.85 Å⁻¹ is assigned to this phase as can be seen in Fig. S2d, e and 3a–c. Since oxygen release is detected during its formation and decay it is assumed to be even more oxygen deficient than mono LNOF. As this intermediate phase has been observed by Wissel et al.²⁷ on electrochemical fluorination of La₂NiO₃F₂, it is in agreement with our supposed overall anion content between that of La₂NiO₃F₂ and mono LNOF. The weight fraction of this phase is at its maximum between 240–400 minutes of reaction time, which is also well-reflected in the heatmap plot (Fig. 3b). Afterwards, the weight fraction decreases to zero towards the end of the reaction.

As described above, ortho LNOF #2 disappears gradually under formation of the final fourth phase La₂NiO₃F₂ Cccm, reflecting the high stability of nickel in its divalent oxidation state. In both reactions, due to the increased oxygen partial pressure above the sample, Ag₂O starts to form from Ag which is the decomposition product of AgF. However, as the oxygen is flushed out by the N₂ flow, the weight fraction of Ag₂O also decreases again, and Ag is observed as the only reaction product left after extended heating at 300 °C.

The release of oxygen upon the chemical fluorination of La₂NiO₄ and the formation of La₂NiO₃F₂ as a final product are a clear indicator that all intermediate phases are not stable at the final temperature of 300 °C and decompose under oxygen release. However, the onset of oxygen release in the fluorination of La₂NiO₄ occurs only after reaching 300 °C, contrary to La₂CoO₄ which already releases oxygen at 250 °C. Performing this fluorination reaction at a lower temperature (210 °C and 250 °C) did not result in a phase pure product either. At 210 °C, a mixture of the starting compound, ortho LNOF #1 and mono LNOF were formed, and after heating to 250 °C the decomposition into ortho LNOF #2 and La₂NiO₃F₂ started very slowly (see Fig. S4). No oxygen or fluorine could be detected in the mass spectrometer which is most likely due to very slow decomposition and formation of Ag₂O which did not decompose at the applied temperatures. It has to be noted that phases observed on AgF-based fluorination also clearly differ from what is observed applying a non-oxidative chemical fluorination with PVDF.²⁶

The lattice parameters of all RP-phases during the fluorination process are displayed in Fig. 4a and b for La₂CoO₄ and La₂NiO₄ respectively.

Fig. 4 Lattice parameters of all RP-phases present during the fluorination of (a) La₂CoO₄ and (b) La₂NiO₄ during with AgF from VTXRD. The dashed lines in the bottom graph of (a) indicate the values for b and c as 300 °C was reached to make the slight change better visible. The numerical error bars for all values have been combined into confidence intervals which are indicated by a shaded region around the respective graphs. The values of the lattice parameters are not displayed if the weight fraction is smaller than 3 wt%.

For La₂CoO₄ the observed trend is comparably simple. The starting oxide La₂CoO₄ shows an increase in the c-lattice parameter while the fluorination is taking place, indicating a partial insertion of fluorine into the lattice²³ beyond a small potential contribution from thermal expansion considering the reported thermal expansion behavior of RP-type materials.^28,29 In the a–b-plane the lattice parameters do not change drastically. For mono LCOF we observe stronger changes. Here the c-axis decreases parallel to the increase in weight fraction, followed by an additional small decrease during the 300 °C temperature hold. The a-lattice parameter on the other hand shows only a very small expansion (even after reaching the maximum temperature of 300 °C), which could indicate an increased space requirement in the octahedral plane due to reduction of the cobalt ions in the fluorinated phase. In contrast, the drop in the c-lattice parameter indicates a removal from ions on interstitial or apical octahedral sites. This is in-line with the oxygen release observed in the same timeframe showing that mono LCOF shows a flexibility towards the oxygen content, releasing oxygen as the temperature is increased which is reflected in the change of lattice parameters. We acknowledge that due to the small lattice parameter changes, the oxygen release is most likely also small.

Due to complex phase evolution behavior of the fluorination of La₂NiO₄ using AgF, the lattice parameters of all phases are affected by larger scattering and error, but general trends can still be derived. The a- and c-lattice parameters of the initial La₂NiO₄ I4/mmm phase are mostly affected by thermal expansion and become expectedly unreliable as the weight fraction drops to zero, which is why they are only shown up to ∼200 min. For ortho LNOF #1, the b- and c-lattice parameters significantly increase only during the heating stage. Therefore, we assume this to be resulting from thermal expansion. During the 300 °C temperature hold, the b-lattice parameter expands minutely while the other lattice parameters are constant.

The a-lattice parameter of mono LNOF in the space group C2/c (please note that the standard setting of this space group has a as longest axis) drops from an initial value of 15.3 Å to around 14.9 Å while a monoclinic distortion builds up and the corresponding angle increases significantly to ∼91° as the c-lattice parameter drops. The initial state is very similar to what was reported for orthorhombic La₂NiO₄F_1.8 by Nowroozi et al.,²⁵ while the final state is closer to the La₂NiO₃F₃ phase with monoclinic symmetry reported by Wissel et al.²⁷ Further, an increase of the lattice parameters in the equatorial plane is in-line with this comparison as Nowroozi et al. determined smaller equatorial lattice parameters for La₂NiO₄F_1.8 than Wissel et al. reported for La₂NiO₃F₃ (a ∼ 5.36 Å, b ∼ 5.31 Å vs. b ∼ 5.47 Å, c ∼ 5.38 Å). These similarities in the lattice parameters further support the above statement that mono LNOF transitions from orthorhombic to monoclinic symmetry while releasing oxygen from the lattice.

We observed that the a-lattice parameter (long axis) of ortho LNOF #2 is decreasing by ∼0.2 Å after a very brief initial increase while the equatorial plane is expanding in both directions by ∼0.1 Å simultaneously. These trends are like what we observed for mono LNOF and indicate that the oxygen detected by the mass spectrometer also originates from ortho LNOF #2. All lattice parameters of La₂NiO₃F₂ Cccm reflect the same trend as was observed for mono LNOF and ortho LNOF #2, decreasing of the long axis while the equatorial plane expands. The values at the end of the experiment are close to what was observed by Wissel et al. for La₂NiO₃F₂,²² even though the a-axis is slightly expanded which might be due to the higher temperature of the in situ measurement.

In summary, the fluorination of La₂CoO₄ leads to the formation of a single fluorinated phase (mono LCOF) accompanied by oxygen release. The lattice parameters of mono LCOF indicate minor release of oxygen also after phase formation. On the other hand, La₂NiO₄ shows a cascade of different phases: La₂NiO₄ I4/mmm → ortho LNOF #1 → mono LNOF → ortho LNOF #2 → La₂NiO₃F₂ Cccm. The last three phases all show lattice parameter changes indicative of oxygen release, which is reflected by the mass spectrometer signals as the total time of oxygen being detected extends to ∼500–600 minutes compared to only ∼250 minutes for La₂CoO₄

3.2 Differential electrochemical mass spectrometry

We conducted DEMS on La₂CoO₄ and La₂NiO₄ during electrochemical fluorination at 140 °C and 163 °C. This electrochemical fluorination represents a reaction equation offor which the amount of y needs to be determined. The temperature of 163 °C was chosen since it best reflects the operation temperature of FIBs from previous publications^23,25,30 (see Fig. S5). Fig. 5 shows the DEMS data for the electrochemical fluorination of composites containing La₂CoO₄ and La₂NiO₄ as well as without active material (only 90 wt% La_0.9Ba_0.1F_2.9 and 10 wt% Super P) at 163 °C including charging curve, and gas evolution values obtained by integration of the ion current converted to ppm values. The DEMS data for electrochemical fluorination at 140 °C can be seen in Fig. S6, with similar trends and further reduced overall gas release. The gases observed and their appearance is complex and thus requires detailed discussion with reference to what is known on other battery materials, e.g. in the lithium-ion battery field.

Fig. 5 DEMS measurement of La₂CoO₄, La₂NiO₄ and a composite without active material in electrochemical fluorination at 163 °C. Dotted lines in the potential graph indicate that the charging has stopped and the cell is now in relaxation. The mass spectrometer was calibrated for CO₂ by measuring air (420 ppm of CO₂). The values for CO/N₂ and NO are to be considered semi-quantitative. The capacity for the cell without active material is calculated for the same mass of active material used in the cells containing either La₂CoO₄ or La₂NiO₄. Since the mass of carbon is the same in all these cells, this way of depicting the capacity best reflects the contribution of carbon to the capacity and gas evolution for the cells containing active materials.

CO₂ release is a known phenomenon in lithium-ion batteries and lithium solid-state batteries which is attributed to surface carbonate residues which can either decompose chemically or electrochemically.^31–33 TG-MS measurements of the components of our cathode composite revealed that all components contain carbonate residues which however only decompose at temperatures >250 °C (see Fig. S8), thus ruling out a direct thermal release at the applied DEMS temperatures (see also Fig. S7).

The only reasonable assumption for the origin of NO is the release of NO_x from the decomposition of nitrates. As NO₂ is a very unstable molecule and electron ionization tends to generate mostly NO molecules and O atoms, the absence of a signal for NO₂ is well explained.³⁴ The origin of nitrate decomposition is attributed to the La_0.9Ba_0.1F_2.9 solid electrolyte via TG-MS since it shows significant mass loss accompanied by NO release when heated to 550 °C (see Fig. S8). We conclude that the precursors for the solid electrolyte contain small amounts of nitrate impurities which could not be removed by any of the pretreatment steps as described above.

No signal at m/z = 32 which would represent O₂ could be detected in the fluorination of any pellet. However, CO and CO₂ can also be formed as byproducts of evolved oxygen reacting with carbon-containing components in the cell setup.¹³ To test if this could contribute to the observed CO and CO₂, a mixture of KMnO₄ and La₂CoO₄ (as a catalyst to lower the decomposition temperature) was heated to 163 °C to intentionally release O₂ in the DEMS setup. While CO and CO₂ could be observed when the decomposition started, a signal for m/z = 32 was always present alongside (see Fig. S9), showing that if O₂ is formed in the DEMS setup, it can also be detected directly.

A composite without active material was measured in order to reflect the gas evolution upon carbon fluorination, and shows large amounts of gas evolution in the DEMS measurement at 163 °C. Gas evolution sets in directly when charging begins and increases rapidly until the evolution of CO (note that the value for CO can be affected by outgassing of adsorbed nitrogen) and CO₂ reaches stable values after ∼40 mAh g⁻¹. NO evolution peaks at this capacity and slowly decreases afterwards. While the evolution of CO₂ remains very constant until the charging process is completed, CO evolution decreases slightly after reaching a capacity of ∼250 mAh g⁻¹. Since both the solid electrolyte and the carbon additive were shown to contain impurities which release CO₂ in a TG-MS measurement (see Fig. S8), CO₂ release observed in the cells with active material is explained from this behavior. Since CO cannot originate from carbonate decomposition, other contaminations and functional groups in the carbon additive (Super P) used must lead to CO release. The potential during carbon fluorination (as was reported to take place in ASS-FIBs^24,27,30,35) shows a plateau around 1.2–1.3 V vs. Pb/PbF₂ at 163 °C or closer to 1.4 V vs. Pb/PbF₂ at 140 °C and shows a characteristic sawtooth-like shape. The sharp increase in potential always correlates exactly with the moment when the cell container is flushed, i.e. the atmosphere is exchanged for a pure argon atmosphere and all evolved gases are flushed out, indicating a sensitivity of the carbon composite to some gaseous species.

For the composite containing La₂CoO₄, the charging potential shows a plateau around 0.7–1.0 V vs. Pb/PbF₂ when charged at 163 °C, which is not affected by sawtooth-like features. Nevertheless, release of CO and CO₂ in the order of 30–40 nmol min⁻¹ g⁻¹ and 25 nmol min⁻¹ g⁻¹ respectively can be detected already. After reaching ∼70–80 mAh g⁻¹, the potential increases to around 1.2–1.3 V vs. Pb/PbF₂, very close to the potential of the cell without active material, and the sawtooth-like features start to appear. At the same time, CO₂ evolution starts to increase linearly, while CO evolution only starts increasing at ∼120 mAh g⁻¹, peaking around 175 nmol min⁻¹ g⁻¹ and 350 nmol min⁻¹ g⁻¹ for CO₂ and CO respectively between 200–250 mAh g⁻¹. This maximum in gas evolution closely matches the rate observed for the cell without active material; in combination with the absence of a signal of O₂, we conclude that the oxygen contained within CO₂ and CO thus must nearly exclusively originate from the decomposition of the carbon additive. Therefore, we assume that ∼120 mAh g⁻¹ marks the end of fluorination of La₂CoO₄ and further capacity and gas release is originating from the solid electrolyte and carbon additive. Since the theoretical capacity of La₂CoO₄ for the intercalation of two fluoride ions per formula unit is 133 mAh g⁻¹ and no O₂ release is indicated, we describe the electrochemically fluorinated product as La₂CoO₄F_2−d, implying nearly full occupation of the interstitial sites but refraining from stating a more precise fluoride content.

At 140 °C, the charging plateau of La₂CoO₄ is observed at a higher potential of ∼1.3 V vs. Pb/PbF₂, which is still below what is observed for the cell without active material. Evolution of the m/z = 28 signal is heavily affected by outgassing of surface-adsorbed N₂ at the beginning (compare Fig. S7), but as N₂ is disappearing the value is stable around 75 nmol min⁻¹ g⁻¹, which again is attributed to CO. CO₂ evolution starts out at 25 nmol min⁻¹ g⁻¹ but slowly increases in a roughly linear fashion. After reaching 100 mAh g⁻¹, CO evolution rapidly increases to 300 nmol min⁻¹ g⁻¹, while the sawtooth shape in the charging curve reaches its largest extent. The evolution of CO₂ which was steadily increasing now peaks at 150 nmol min⁻¹ g⁻¹. As the voltage increases to 2 V vs. Pb/PbF₂, the evolution of CO and CO₂ decreases again. The general trend of the gas evolution values at 140 °C is similar to the fluorination at 163 °C, only showing an increase in the gas evolution rates slightly earlier, which is in-line with reduced carbon fluorination as judged by the difference in capacity in cell without active material.

For La₂NiO₄, the CO evolution behavior at 163 °C is very similar to La₂CoO₄, only that the peak evolution of ∼300 nmol min⁻¹ g⁻¹ is already reached at ∼110 mAh g⁻¹ and the fluorination potential is higher at ∼1.2–1.3 V vs. Pb/PbF₂, very close to the potential of the cell without active material, indicating that the fluorination of La₂NiO₄ is strongly overlaid by this side reaction. Judging from the capacity where gas evolution peaked, the absence of O₂ signals as well as the fact that the amount of the CO/CO₂ formed is well explained from the side reaction of the carbon additive, we describe the electrochemically fluorinated product as La₂NiO₄F_2−d with similar reasoning as for La₂CoO₄F_2−d (the specific capacity of La₂NiO₄F₂ is also 133 mAh g⁻¹). When charging at 140 °C, the behavior of La₂NiO₄ is different, showing a charging plateau at ∼1.4 V vs. Pb/PbF₂ which is also very close to that of cell without active material. During the whole charging process, CO evolution is rather constant at 70–100 nmol min⁻¹ g⁻¹, while CO₂ evolution is very similar to that of La₂CoO₄, increasing from 0 to 75 nmol min⁻¹ g⁻¹. The CO evolution is even smaller than what was observed for cell without active material, while the CO₂ evolution is only slightly smaller. Again, no evolution of NO could be observed.

The diffraction patterns of the La₂CoO₄ and La₂NiO₄ cells after the DEMS measurement (see Fig. S12a for La₂CoO₄ at 163 °C, Fig S13a for La₂NiO₄ at 163 °C and Fig. S10a and b for both materials at 140 °C) show that the La₂CoO₄ has fully been converted to La₂CoO₄F_2−d while for La₂NiO₄, a small weight fraction of ortho LNOF #1 as also observed in chemical fluorination is present alongside La₂NiO₄F_2−d. Again, this is in agreement with the absence of oxygen evolution in the DEMS measurements, in accordance with an unchanged oxygen content of the pristine and charged cathode materials. The La₂NiO₄ cell charged at 140 °C (Fig. S10a) shows only a minor weight fraction of La₂NiO₄F_2−d while two phases of ortho LNOF #1 are present, indicative of an incomplete fluorination of the active material. Since the potential was very similar to the cell without active material, we assume that significant carbon fluorination took place, which was not the case for the La₂CoO₄ cell at 140 °C which charged at a lower potential, obtaining full conversion to La₂CoO₄F_2−d (Fig. S10c).

In summary, the DEMS measurements of La₂CoO₄ and La₂NiO₄ in electrochemical fluorination show that significant amounts of CO₂ and CO are released from contaminations in the solid electrolyte and carbon additives. When an active material is employed in the cell, those gas releases are postponed to higher states of charge, and an evolution of O₂ is not indicated, i.e. all oxygen released and contained within CO/CO₂ can be explained from the decomposition of the carbon additive. This shows that the evolution of gases from La₂CoO₄ and La₂NiO₄ in electrochemical fluorination is not accompanied by major oxygen release, contrary to what was observed in chemical fluorination as discussed in Section 3.1. Most likely, the lower temperature used in the DEMS measurements as compared to fluorination with AgF stabilizes higher oxidation states, therefore inhibiting oxygen-release.

3.3 Thermal stability of electrochemically fluorinated La₂CoO₄ and La₂NiO₄

Since the chemical fluorination experiments have shown that high oxygen contents of fluorinated materials are unstable at elevated temperatures, we also investigated the thermal stability of the charged state of the cathode materials La₂CoO₄ and La₂NiO₄. For this, the pellets fluorinated in the DEMS measurements were heated to 300 °C in dynamic vacuum while recording XRD patterns continuously, completing a scan every 30 minutes. The resulting heatmap-plots of the XRD patterns are shown in Fig. 6a for La₂CoO₄ and Fig. 6b La₂NiO₄, while the weight fractions obtained from Rietveld-refinement are displayed in Fig. 6c and d respectively.

Fig. 6 Heatmap-plot of the VTXRD patterns of electrochemically fluorinated (a) La₂CoO₄ and (b) La₂NiO₄ pellets. The weight fractions obtained by Rietveld-refinement area shown in (c) for La₂CoO₄ and (d) for La₂NiO₄. The dotted red/white line marks the time and temperature where the decomposition products appear.

We observe that electrochemically obtained La₂CoO₄F_2−d decomposes about 60 minutes after 300 °C is reached, and a new phase starts to appear which could be refined by adopting the structure of La₂NiO₃F₂ for Co instead of Ni changing the lattice parameters only minimally (see Fig. S11, S12e and f). We observed that this material La₂CoO₃F₂ was also obtained in defluorination experiments of La₂CoO₃F₃ and a full description will be published along with the structural characterization La₂CoO₃F₃ in a separate article. Prior to the decomposition reaction, the lattice parameters of La₂CoO₄F_2−d (Fig S11a) show a characteristic expansion of the equatorial plane accompanied by a contraction of the long axis, which was previously linked to oxygen release from the lattice. Obtaining a phase of the stoichiometry La₂CoO₃F₂ is in-line with the implied oxygen loss from La₂CoO₄F_2−d even though we did not detect it directly, as this VTXRD-setup is not equipped with a mass spectrometer.

The electrochemically obtained La₂NiO₄F_2−d decomposes almost completely within 30 minutes after a temperature of 300 °C is reached, resulting in a mixture of ortho LNOF #2 and La₂NiO₃F₂ as observed in chemical fluorination and an increased amount of ortho LNOF #1 compared to the start of the decomposition reaction (see Fig. S13). This implies that LNOF #1 is not only an intermediate of the fluorination of La₂NiO₄ to La₂NiO₄F_2−d but also a decomposition product. Also, the lattice parameters of La₂NiO₄F_2−d and La₂NiO₃F₂ (Fig. S11b) show very similar trends to the lattice parameters observed in chemical fluorination. As mentioned before, this characteristic expansion of the equatorial plane, contraction of the long axis and increase of the monoclinic angle implies oxygen release from both structures. The more rapid decomposition of La₂NiO₄F_2−d in comparison to La₂CoO₄F_2−d might be related to the different strength of overlap of the Ni/Co-d and O-2p energy levels. For lithium-ion batteries, it is known that the larger overlap of the Ni-d and O-2p orbitals tends to facilitate oxygen release through hole formation in the O-2p orbitals.^36,37 Similar trends in thermal stability could also be observed for LiNi_xMn_yCo_zO₂ cathode materials where higher nickel content correlated with lower decomposition temperature and more rapid decomposition reactions³⁸

We conclude that electrochemically obtained La₂CoO₄F_2−d is of a different anion composition than mono LCOF obtained from chemical fluorination. While the latter is stable towards decomposition at 300 °C, only showing minor lattice parameter changes, the prior decomposes under formation of an oxygen deficient monoclinic phase and La₂CoO₃F₂. This supports the results from DEMS, showing that electrochemical fluorination does not lead to oxygen release as was detected in chemical fluorination. On the contrary, electrochemically obtained La₂NiO₄F_2−d and its chemically fluorinated counterpart mono LNOF, both decompose at 300 °C showing similar trends in lattice parameters and the same decomposition products. Since the DEMS measurements also show no oxygen evolution, and oxygen was only detected in the chemical fluorination when mono LNOF transitioned from orthorhombic to monoclinic symmetry, we assume the electrochemically obtained La₂NiO₄F_2−d and chemically obtained mono LNOF to be of very similar anion composition.

4 Conclusions and outlook

In this study, we report the oxidative fluorination pathways of La₂CoO₄ and La₂NiO₄ using AgF which are accompanied by oxygen release as detected by in situ variable-temperature X-ray diffraction with coupled mass spectrometry. Further, we report the evolution of CO, CO₂ and NO for all-solid-state fluoride ion batteries using La₂NiO₄ or La₂CoO₄ as cathode active materials additive using a novel setup for differential electrochemical mass spectrometry which allows for measurements at elevated temperatures up to 200 °C. Oxygen release from the cathode materials could not be detected, highlighting electrochemical fluorination as a synthesis method for low-temperature fluorination to obtain compounds which are unstable at higher temperatures. Insights gained from X-ray diffraction during thermal decomposition of the charged cathode composites show that the charged cathode materials are thermally stable up to 220 °C, showing significant lattice parameter changes above 220 °C and decomposing at 300 °C.

The findings reported here further provide a completely new perspective on the mechanisms discussed for topochemical fluorination reactions^39–42 in dependence of the reaction temperature and oxidation strength of the fluorination agent used: since the electrochemical fluorination at lower temperatures can reduce the Fermi Level while at the same time kinetically blocking the oxygen release, we highlight the potential of the electrochemical approach for making new materials beyond battery application with novel electronic and/or magnetic properties. Thus, the complex interaction of charge transition levels in perovskite-related materials, their temperature-dependent accessibility and the relation to the electronic structure of the materials facilitates a novel important approach for choosing the most promising synthesis strategy for a targeted oxyfluoride compound.⁴³

As shown in an earlier publication,²⁴ La₂NiO₄ and La₂CoO₄ are able to provide high cyclic stability with excellent coulombic efficiency after initial side reactions decreased within the first few cycles. At this point, we can state that gas release from contaminations in the cathode composite in early cycles is a possible origin for the low initial coulombic efficiency, which would mark gas release from the composite as an activation process, not a degradation mechanism.

Further research could try to purify the solid electrolyte and conductive carbon to avoid gas evolution from the composites. Also, more careful attempts to obtain orthorhombic La₂NiO₄F_2−d phase pure would enable an in-depth characterization of this compound not possible before.

Author contributions

T. A.: conceptualization, data curation, formal analysis, investigation, methodology, software,^44,45 visualization, writing – original draft; J. J.: data curation, investigation, methodology, validation, visualization, writing – review & editing; F. F.: data curation; D. S.: methodology, resources; O. C.: funding acquisition, investigation, project administration, supervision, validation, writing – review & editing.

Conflicts of interest

There are no conflicts of interest to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5ta07033c.

The code for automated Rietveld refinements as well as integration of mass spectrometry data can be found at https://github.com/TommiAalto-IMW/batchXRD https://github.com/TommiAalto-IMW/DEMSIntegration with DOI: https://doi.org/10.5281/zenodo.16993766 https://doi.org/10.5281/zenodo.16993870 The version of the code employed for this study is version 1.0 and 1.2 respectively.

Acknowledgements

We thank Rotraut Merkle from the Max-Planck-Institute for Solid-State Research for letting us use the mass spectrometer and all the kind assistance with its operation. Special thanks to Florian Kaiser for the constant support in the construction and testing of the DEMS setup. Further thanks to the Institute of Chemical Process Engineering at the University of Stuttgart for access to their mass spectrometer in the earlier stages of the development of the DEMS setup. OC acknowledges funding via the collaborative research center FLAIR (Fermi level engineering applied to oxide electroceramics), which is supported by the German Research Foundation (DFG) (Project-ID No. 463184206, SFB 1548).

References

1. T.-Y. Huang, et al., Deconvolution of intermixed redox processes in Ni-based cation-disordered Li-excess cathodes, Energy Environ. Sci., 2021, 14(3), 1553–1562.
2. M. M. Rahman and F. Lin, Oxygen Redox Chemistry in Rechargeable Li-Ion and Na-Ion Batteries, Matter, 2021, 4(2), 490–527.
3. R. A. House, et al., The role of O₂ in O-redox cathodes for Li-ion batteries, Nat. Energy, 2021, 6(8), 781–789.
4. N. Guerrini, et al., Charging Mechanism of Li₂MnO₃, Chem. Mater., 2020, 32(9), 3733–3740.
5. M. Farahmandjou, et al., Oxygen redox chemistry in lithium-rich cathode materials for Li-ion batteries: Understanding from atomic structure to nano-engineering, Nano Mater. Sci., 2022, 4(4), 322–338.
6. Z. Yin, et al., Approaching a stable oxygen redox reaction in lithium-rich cathode materials: structural perspectives from mechanism to optimization, J. Mater. Chem. A, 2022, 10(37), 19387–19411.
7. Q. Li, et al., Improving the oxygen redox reversibility of Li-rich battery cathode materials via Coulombic repulsive interactions strategy, Nat. Commun., 2022, 13(1), 1123.
8. S. Sharifi-Asl, et al., Oxygen Release Degradation in Li-Ion Battery Cathode Materials: Mechanisms and Mitigating Approaches, Adv. Energy Mater., 2019, 9(22), 1900551.
9. S. Kobayashi, et al., Atomic-Scale Observations of Oxygen Release Degradation in Sulfide-Based All-Solid-State Batteries with Layered Oxide Cathodes, ACS Appl. Mater. Interfaces, 2022, 14(34), 39459–39466.
10. R. Jung, et al., Oxygen Release and Its Effect on the Cycling Stability of LiNi_xMn_yCo_zO₂ (NMC) Cathode Materials for Li-Ion Batteries, J. Electrochem. Soc., 2017, 164(7), A1361.
11. J. K. Papp, et al., A comparison of high voltage outgassing of LiCoO₂, LiNiO₂, and Li₂MnO₃ layered Li-ion cathode materials, Electrochim. Acta, 2021, 368, 137505.
12. R. Jung, et al., Temperature Dependence of Oxygen Release from LiNi_0.6Mn_0.2Co_0.2O₂ (NMC622) Cathode Materials for Li-Ion Batteries, J. Electrochem. Soc., 2018, 165(11), A2869.
13. F. Strauss, et al., Gas Evolution in Lithium-Ion Batteries: Solid versus Liquid Electrolyte, ACS Appl. Mater. Interfaces, 2020, 12(18), 20462–20468.
14. T. Bartsch, et al., Gas Evolution in All-Solid-State Battery Cells, ACS Energy Lett., 2018, 3(10), 2539–2543.
15. R. Jung, et al., Oxygen Release and Its Effect on the Cycling Stability of LiNi_xMn_yCo_zO₂ (NMC) Cathode Materials for Li-Ion Batteries, J. Electrochem. Soc., 2017, 164, A1361.
16. D. Zhang, et al., Cathode Design Based on Nitrogen Redox and Linear Coordination of Cu Center for All-Solid-State Fluoride-Ion Batteries, J. Am. Chem. Soc., 2025, 147(7), 5649–5657.
17. Y. Wang, et al., Oxyfluoride Cathode for All-Solid-State Fluoride-Ion Batteries with Small Volume Change Using Three-Dimensional Diffusion Paths, Chem. Mater., 2022, 34(23), 10631–10638.
18. H. Miki, et al., Double-Layered Perovskite Oxyfluoride Cathodes with High Capacity Involving O–O Bond Formation for Fluoride-Ion Batteries, J. Am. Chem. Soc., 2024, 146(6), 3844–3853.
19. Z. Cao, et al., Revealing the Unusual Mechanism of Mixed Cationic and Anionic Redox in Oxyfluorosulfide Cathode for All-Solid-State Fluoride-Ion Batteries, Chem. Mater., 2024, 36(4), 1928–1940.
20. Y. Wang, et al., Ultra-High-Capacity of Earth-Abundant Cathodes Enabled by Excess Fluoride-Ion Insertion/Extraction, Adv. Energy Mater., 2025, 2406131.
21. R. Zhang, M. S. Senn and M. A. Hayward, Directed Lifting of Inversion Symmetry in Ruddlesden–Popper Oxide–Fluorides: Toward Ferroelectric and Multiferroic Behavior, Chem. Mater., 2016, 28(22), 8399–8406.
22. K. Wissel, et al., Topochemical Fluorination of La₂NiO_4+d: Unprecedented Ordering of Oxide and Fluoride Ions in La₂NiO₃F₂, Inorg. Chem., 2018, 57(11), 6549–6560.
23. M. A. Nowroozi, et al., La₂CoO₄: a new intercalation based cathode material for fluoride ion batteries with improved cycling stability, J. Mater. Chem. A, 2018, 6(11), 4658–4669.
24. T. H. Aalto, et al., Toward Carbon-Free Cathodes for Fluoride Ion Batteries: Deconvoluting Effects of Active Material and Conductive Additive on Charging and Cyclic Stability, Batteries Supercaps, 2025, 2500195.
25. M. A. Nowroozi, et al., High cycle life all-solid-state fluoride ion battery with La₂NiO_4+d high voltage cathode, Commun. Mater., 2020, 1(1), 27.
26. J. Jacobs, et al., Unveiling the Fluorination Pathway of Ruddlesden–Popper Oxyfluorides: A Comprehensive In Situ X-ray and Neutron Diffraction Study, J. Am. Chem. Soc., 2025, 147(11), 9739–9751.
27. K. Wissel, et al., Electrochemical Reduction and Oxidation of Ruddlesden–Popper-Type La₂NiO₃F₂ within Fluoride-Ion Batteries, Chem. Mater., 2021, 33(2), 499–512.
28. Y. Hu, et al., La_2−xSr_xCoO_4−δ (x = 0.9, 1.0, 1.1) Ruddlesden-Popper-type layered cobaltites as cathode materials for IT-SOFC application, Int. J. Hydrogen Energy, 2013, 38(7), 3064–3072.
29. L. V. Makhnach, V. V. Pankov and P. Strobel, High-temperature oxygen non-stoichiometry, conductivity and structure in strontium-rich nickelates La_2−xSr_xNiO_4−δ (x=1 and 1.4), Mater. Chem. Phys., 2008, 111(1), 125–130.
30. M. A. Nowroozi, et al., LaSrMnO₄: Reversible Electrochemical Intercalation of Fluoride Ions in the Context of Fluoride Ion Batteries, Chem. Mater., 2017, 29(8), 3441–3453.
31. T.-Y. Huang, et al., Chemical Origin of in Situ Carbon Dioxide Outgassing from a Cation-Disordered Rock Salt Cathode, Chem. Mater., 2024, 36(13), 6535–6546.
32. S. E. Renfrew, L. A. Kaufman and B. D. McCloskey, Altering Surface Contaminants and Defects Influences the First-Cycle Outgassing and Irreversible Transformations of LiNi_0.6Mn_0.2Co_0.2O₂, ACS Appl. Mater. Interfaces, 2019, 11(38), 34913–34921.
33. S. E. Renfrew and B. D. McCloskey, Residual Lithium Carbonate Predominantly Accounts for First Cycle CO₂ and CO Outgassing of Li-Stoichiometric and Li-Rich Layered Transition-Metal Oxides, J. Am. Chem. Soc., 2017, 139(49), 17853–17860.
34. P. J. Linstrom and W. G. Mallard, The NIST Chemistry WebBook: A Chemical Data Resource on the Internet, J. Chem. Eng. Data, 2001, 46(5), 1059–1063.
35. Y. Matsuo, et al., Charge-discharge behavior of fluorine-intercalated graphite for the positive electrode of fluoride ion shuttle battery, Electrochem. Commun., 2020, 110, 106626.
36. A. Manthiram and T. Muraliganth, Lithium Intercalation Cathode Materials for Lithium-Ion Batteries, in Handbook of Battery Materials, 2011, pp. 341–375.
37. M. Dixit, et al., Origin of Structural Degradation During Cycling and Low Thermal Stability of Ni-Rich Layered Transition Metal-Based Electrode Materials, J. Phys. Chem. C, 2017, 121(41), 22628–22636.
38. S.-M. Bak, et al., Structural Changes and Thermal Stability of Charged LiNixMnyCozO2 Cathode Materials Studied by Combined In Situ Time-Resolved XRD and Mass Spectroscopy, ACS Appl. Mater. Interfaces, 2014, 6(24), 22594–22601.
39. E. E. McCabe and C. Greaves, Fluorine insertion reactions into pre-formed metal oxides, J. Fluorine Chem., 2007, 128(4), 448–458.
40. C. Greaves and M. G. Francesconi, Fluorine insertion in inorganic materials, Curr. Opin. Solid State Mater. Sci., 1998, 3(2), 132–136.
41. H. Kageyama, et al., Exploring Structures and Properties through Anion Chemistry, Bull. Chem. Soc. Jpn., 2019, 92(8), 1349–1357.
42. Y. Kobayashi, Y. Tsujimoto and H. Kageyama, Property Engineering in Perovskites via Modification of Anion Chemistry, Annu. Rev. Mater. Res., 2018, 48, 303–326.
43. A. Klein, et al., The Fermi energy as common parameter to describe charge compensation mechanisms: A path to Fermi level engineering of oxide electroceramics, J. Electroceram., 2023, 51(3), 147–177.
44. Aalto, T. DEMSIntegration. GitHub,  DOI:10.5281/zenodo.16993870.
45. Aalto, T. batchXRD. GitHub,  DOI:10.5281/zenodo.16993766.

---