# Fileset

[Prince_Sharma_2026_d6ta03003c.pdf](https://mdr.nims.go.jp/filesets/f54c9f84-5954-41d0-a29e-0c74a39ffbca/download)

## Creator

[Prince Sharma](https://orcid.org/0000-0001-5735-5206), [Gen Hasegawa](https://orcid.org/0000-0002-9297-6902), [Kaori Sugii](https://orcid.org/0009-0005-7056-8659), [Yasuyuki Fujiwara](https://orcid.org/0000-0003-0028-0658), Hideki Iba, [Naoaki Kuwata](https://orcid.org/0000-0002-0736-6967)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals                    <i>via</i>                    isotope tracing and TOF-SIMS](https://mdr.nims.go.jp/datasets/c9908a0d-8367-4b48-84e3-8edcbd1e8286)

## Fulltext

Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS  Journal of Materials Chemistry A PAPER Published on 21 September 2026 Licensed under CC-BY 4.0Direct observatioaCenter for Green Research on Energy and En for Materials Science (NIMS), Tsukuba, Iba Naoaki@nims.go.jp bGraduate School of Chemical Sciences Sapporo, Hokkaido 060-8628, Japan cFaculty of Engineering, Shinshu University, dToyata Motor Corporation, Susono 410-119 eInstitute for Molecular Science, Nationa Okazaki, Aichi 444-8585, Japan Cite this: J. Mater. Chem. A, 2026, 14, 42453 Received 9th April 2026 Accepted 3rd September 2026 DOI: 10.1039/d6ta03003c rsc.li/materials-a This journal is © The Royal Society on and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS Prince Sharma, a Gen Hasegawa, ae Kaori Sugii, a Yasuyuki Fujiwara, c Hideki Ibaad and Naoaki Kuwata \*ab Understanding intrinsic ion transport in solid electrolytes is essential for the development of high-performance all-solid-state batteries. However, quantitative evaluation of bulk diffusion, decoupled from grain-boundary contributions, remains challenging, particularly in fast-ion-conducting materials. In this study, we establish a methodology for quantitatively extracting intrinsic lithium-ion diffusion coefficients in single-crystalline solid electrolytes using time-of-flight secondary ion mass spectrometry (TOF-SIMS) combined with 6Li isotope exchange. Perovskite-type La(1−x)/3LixNbO3 (x = 0.08, LLNbO) is employed as a model system due to its high lithium-ion conductivity and structural stability. 6Li isotope exchange is carried out using 6LiNO3 solution, followed by controlled thermal annealing to induce diffusion. High-resolution TOF-SIMS is used to obtain spatially resolved lithium concentration profiles. The diffusion profiles were initially analysed using analytical solutions of Fick's second law. The final bulk diffusion coefficients were then determined using a two- dimensional numerical diffusion model that accounts for finite-size effects and the finite sample geometry. This analytical framework enables reliable reconstruction of diffusion profiles and quantitative extraction of intrinsic bulk diffusion coefficients over a wide range of temperatures and timescales. The proposed methodology provides a general approach for quantitative microscale characterization of ion transport in solid-state materials and offers a useful framework for the design of advanced solid electrolytes.Introduction There is a great need for the development of next-generation electrochemical energy storage systems, specically all-solid-state batteries, for which the central need is to understand ion trans- port in solid electrolytes (SEs).1–4 Generally, all the key battery performance metrics including rate capability, ionic conductivity, long-term stability and interfacial charge transfer are fundamen- tally governed by the ionic transport in SEs.5,6However, elucidating intrinsic ion transport mechanisms remains challenging due to the complex interplay of microstructural features such as grain boundaries, phase heterogeneity, and defect distributions.7–11 In most polycrystalline SEs, these extrinsic factors oen dominate ionic conduction, obscuring the true bulk transport properties governed by lattice dynamics, intrinsic defect chemistry, andvironmental Materials, National Institute raki 305 0044, Japan. E-mail: KUWATA. and Engineering, Hokkaido University, Wakasato, Nagano, 380-8533, Japan 3, Japan l Institutes of Natural Science (NINS), f Chemistry 2026crystallographic anisotropy. Therefore, direct measurement of intrinsic bulk ion diffusion, free from grain boundary and inter- facial contributions, is essential for establishing reliable structure– property relationships. Despite extensive efforts, experimental techniques capable of resolving intrinsic ion diffusion with both high spatial resolution and quantitative accuracy remain limited. Conventional methods such as electrochemical impedance spec- troscopy (EIS) provide macroscopic conductivity but cannot unambiguously distinguish between bulk and interfacial contributions.12–14 Pulsed-eld gradient nuclear magnetic reso- nance (PFG-NMR) enables measurement of self-diffusion coeffi- cients but lacks spatial resolution and may require isotopic labelling depending on the system.10,15,16 Similarly, quasi-elastic neutron scattering (QENS) provides insights into short-range ion dynamics17 but is constrained by large sample requirements and limited access to neutron facilities. To overcome these limitations, isotope tracer techniques combined with spatially resolved anal- ysis offer a promising pathway for directly probing ion diffusion. In particular, time of ight secondary ionmass spectrometry (TOF- SIMS), coupled with lithium isotope exchange, enables high- resolution mapping of lithium distribution in solids.12,18–24 However, in fast-ion-conducting solid electrolytes, accurate reconstruction of diffusion proles is challenging due to artifacts induced by the analysis ion beam and sputtering processes, as well as the lack of standardized isotope-exchange protocols. As a result,J. Mater. Chem. A, 2026, 14, 42453–42463 \| 42453[http://crossmark.crossref.org/dialog/?doi=10.1039/d6ta03003c\&domain=pdf\&date\_stamp=2026-10-09](http://crossmark.crossref.org/dialog/?doi=10.1039/d6ta03003c&domain=pdf&date_stamp=2026-10-09)[http://orcid.org/0000-0001-5735-5206](http://orcid.org/0000-0001-5735-5206)[http://orcid.org/0000-0002-9297-6902](http://orcid.org/0000-0002-9297-6902)[http://orcid.org/0009-0005-7056-8659](http://orcid.org/0009-0005-7056-8659)[http://orcid.org/0000-0003-0028-0658](http://orcid.org/0000-0003-0028-0658)[http://orcid.org/0000-0002-0736-6967](http://orcid.org/0000-0002-0736-6967)[https://creativecommons.org/licenses/by/4.0/](https://creativecommons.org/licenses/by/4.0/)Journal of Materials Chemistry A Paperthe application of this approach to quantitatively extract intrinsic bulk diffusion coefficients remains limited.10,21–23,25,26 In this study, we establish a robust methodology for quanti- fying intrinsic lithium-ion diffusion in single-crystalline solid electrolytes using TOF-SIMS combined with 6Li isotope exchange. By employing well-dened single-crystal systems, this approach eliminates grain boundary effects and enables direct visualization and quantication of bulk diffusion behaviour with high spatial resolution. As for a model system, we focus on perovskite-type lanthanum–lithium niobate (LLNbO) and its substituted vari- ants, such as La(1−x)/3LixNbO3, where x represents the lithium ion concentration, which exhibit a favourable combination of high ionic conductivity and chemical stability.7,27–34 While previous studies have reported anisotropic ionic conductivity and the inuence of grain boundaries in polycrystalline samples, the intrinsic bulk diffusion behaviour in highly crystalline LLNbO has not been directly visualized.35–37 Here, 6Li isotope exchange is introduced into single-crystalline LLNbO, followed by controlled thermal annealing to induce diffusion. High-resolution TOF-SIMS was employed to obtain lithium concentration proles in LLNbO single crystals. The diffusion proles were initially analysed using analytical solutions of Fick's second law and subsequently rened using a two-dimensional numerical diffusionmodel to account for nite-size effects and the nite sample geometry, enabling quan- titative determination of intrinsic bulk lithium diffusion coeffi- cients. This methodology provides a robust framework for quantitative analysis of intrinsic lithium-ion diffusion in single- crystalline solid electrolytes. The approach presented here offers a general framework for quantitatively probing intrinsic ion diffusion in solid-state materials and provides a powerful tool for advancing the design of high-performance solid electrolytes for next-generation batteries.Experimental section Growth of single crystals Single-crystalline perovskite-type La(1−x)/3LixNbO3 (x= 0.08) was prepared via a multi-step solid-state reaction followed by a seedless directional solidication process, as reportedFig. 1 Schematic representation of the 6Li isotope exchange and diffusi 42454 \| J. Mater. Chem. A, 2026, 14, 42453–42463previously.38 First, polycrystalline LLNbO was synthesized from LiNbO3 and LaNb3O9 precursors. The resulting material was then processed into dense rods and subjected to directional solidication under a controlled temperature gradient to obtain bulk single crystals. The obtained crystals were cut into 4.7 × 4\.7 × 0.9 mm3 rectangular slabs and mechanically polished prior to diffusion measurements.Isotope exchange Lithium isotope exchange was performed using a 6Li-enriched LiNO3 aqueous solution (1 mol L−1) following a previously re- ported methodology.10 The solution was prepared by reacting 6Li2CO3 (95% 6Li, Cambridge Isotope Laboratories) with concentrated HNO3 (65 wt%, FUJIFILM Wako) in deionized water. Single-crystalline LLNbO samples (4.7 × 4.7 × 0.9 mm3) were partially immersed in the 6LiNO3 solution for different durations (47, 115, and 137 h for Samples A–C, respectively) to control the diffusion length. The thickness direction (0.9 mm) corresponds to the \[001\] axis, while the in-plane directions (4.7 mm) correspond to the \[100\]/\[010\] axes. Isotope exchange and subsequent diffusion measurements were carried out along the in-plane direction, i.e., predominantly along the \[100\] axis as shown in Fig. 1. Aer isotope exchange, the single crystals were rapidly cooled to 163 K inside the TOF-SIMS chamber to suppress further lithium diffusion and preserve the isotope distribution. Sample A was used to evaluate room-temperature diffusion and subsequent time-dependent redistribution aer storage. Samples B and C were further subjected to stepwise annealing between 300 and 523 K to investigate temperature- dependent diffusion behaviour. Aer each annealing step, the samples were cooled to 163 K prior to TOF-SIMS analysis. Depth proles of 6Li were obtained by TOF-SIMS, enabling quantita- tive analysis of lithium diffusion as a function of time and temperature. The integrated areas of the experimentally ob- tained isotope proles aer isotope exchange and aer each annealing step were quantitatively compared, as summarized in Table S1. The integrated value used as the tting parameter was kept constant for each sample. The experimentally obtained integrated values were in good agreement with theon measurement process in LLNbO. This journal is © The Royal Society of Chemistry 2026Paper Journal of Materials Chemistry Acorresponding tting parameters. This quantitative agreement supports the assumption that the total amount of 6Li in the sample is conserved during annealing and validates the use of the closed-system diffusion model.X-ray diffraction (XRD) X-ray diffraction (XRD) measurements were performed using a Rigaku SmartLab diffractometer to conrm the single- crystalline nature of LLNbO.Impedance spectroscopy Ionic conductivity was measured by impedance spectroscopy over the frequency range of 20 Hz to 100 MHz. Measurements between 200 and 470 K were conducted using an E4990A impedance analyser (Keysight, USA) equipped with a 4990EDMS-120K system (TOYO Corporation, Japan), while higher-temperature measurements (570–1080 K) were per- formed using a high-temperature system (HT-Z2-HF, TOYO Corporation, Japan). All impedance measurements were carried out on LLNbO samples with dimensions of 4.7 × 4.7 × 0.9 mm3, where thin gold electrodes were sputtered onto opposite sides, with the electric eld applied along the in-plane \[100\] direction.TOF-SIMS Prior to depth proling, TOF-SIMS imaging was performed using a TOF-SIMS 5 instrument (IONTOF GmbH, Germany) operated at 163 K under liquid nitrogen cooling to preserve the isotope distribution during analysis. A Bi3 2+ primary ion beam (30 kV) was rastered over an area of 500 × 151 mm2, providing a lateral resolution of approximately 1× 1 mm2. Before imaging, the sample surface was cleaned using an Ar gas cluster ion beam. Subsequently, depth proling was carried out using the same instrument operated at 163 K under liquid nitrogen cooling to suppress lithium diffusion during analysis. A Bi3 2+ primary ion beam (30 kV) was used for analysis, combined with an electron ood gun for charge compensation and Ar gas cluster ion beam (10 kV, 150 × 150 mm2 raster) for sputtering. The use of charge neutralization was essential due to the insulating nature of LLNbO.Fig. 2 XRD pattern showing only 00l reflections, confirming the single-crystalline nature and strong c-axis orientation of LLNbO. Inset: crystal structure.Diffusion analysis The lithium concentration proles obtained by TOF-SIMS were initially analysed using analytical solutions of Fick's second law. For diffusion lengths comparable to the sample dimensions, diffusion coefficients were subsequently rened using a two- dimensional numerical diffusion model that accounts for the nite sample geometry. The numerical model reproduces the experimentally observed concentration proles by solving the diffusion equation under the actual sample geometry. Details of the numerical procedure are provided in the SI (Fig. S1 and S2).This journal is © The Royal Society of Chemistry 2026Results and discussion Crystal structure and orientation XRD was used to characterise the crystal structure of the grown single crystal of LLNbO. The diffraction pattern shows sharp and well-dened 00l reections, which conrms its single- crystalline nature, as shown in Fig. 2. The presence of only 00l peaks (001–006) indicates a strong preferential orientation along the c-axis. The inset illustrates the XRD measurement geometry and the crystal structure of LLNbO. These results validate the phase purity and preferred orientation of the LLNbO single crystal. Previous studies have demonstrated that the lattice param- eters of LLNbO vary systematically with Li composition.39 The lattice parameter obtained in the present study (c = 7.891 Å) is consistent with previously reported LLNbO single crystals with Li compositions near x z 0.08, supporting the nominal composition used in this study.40,41 Although the exact Li concentration cannot be determined solely from XRD, previous ICP-correlated studies have suggested that the high-quality crystal regions used for the present measurements correspond approximately to x z 0.08 (typically x z 0.075–0.085). Any remaining uncertainty in the Li concentration is not expected to affect the main conclusions of this work.Ionic conductivity and diffusion coefficient from impedance spectroscopy Fig. 3a presents the frequency dependence of the real part of the AC conductivity (s0) measured at various temperatures for the LLNbO single crystal (Sample B). In contrast to polycrystalline ceramics such as LLTO, no grain boundary contribution is observed, reecting the single-crystalline nature of the sample. Accordingly, the observed response arises solely from bulk ionic conduction and the electrode interface. The high-frequencyJ. Mater. Chem. A, 2026, 14, 42453–42463 \| 42455Fig. 3 (a) Frequency dependence of the real part of the AC conductivity measured at various temperatures and (b) the corresponding conductivity diffusion coefficient of LLNbO measured for Sample B. Journal of Materials Chemistry A Paperregion (105 to 108 Hz) corresponds to the intrinsic bulk response, while the low-frequency regime (f \< 102 Hz) is asso- ciated with electrode polarization. The bulk ionic conductivity (sbulk) is obtained from the frequency-independent plateau in s0(f). Fig. 3b shows the Arrhenius plot of sbulk. Linear behaviour is observed in the temperature range 240–500 K, from which the activation energy was determined. The ionic conductivity follows the Arrhenius relation sbulkT ¼ s0e �  Ea kBT (1) where s0 is the pre-exponential factor, Ea is the activation energy, kB is the Boltzmann constant, and T is the absolute temperature. The activation energy derived from the linear region of the Arrhenius plot (240–500 K) is 0.33 eV, which is comparable to values reported for Li-ion transport in other perovskite-type single crystals.7,38 Above 500 K, a slight deviation from Arrhenius behaviour is observed. Previous studies have suggested thermally induced structural modications in LLNbO around this temperature range,7 which may contribute to this behaviour. However, the present results do not allow us to determine the underlying mechanism. The conductivity diffusion coefficient (Ds) is calculated using the Nernst–Einstein relation:Fig. 4 TOF-SIMS images of the relative 6Li/(6Li + 7Li) intensity ratio in the corresponds to the diffusion direction. (a) After 6Li isotope exchange at 3 broader 6Li distribution after annealing indicates thermally activated Li d 42456 \| J. Mater. Chem. A, 2026, 14, 42453–42463Ds ¼ kBT nz2e2 sbulk (2) where z is the valence of Li+ (z= 1), e is the elementary charge, kB is the Boltzmann constant and n is the number density of Li+ ions (1.32 × 1021 cm−3). The Li-ion concentration (n) in the Nernst–Einstein equation was calculated from the nominal Li composition (x z 0.08) and the unit-cell volume determined by XRD. Although the concentration of mobile Li ions may be smaller than the total Li concentration, it has not yet been experimentally determined for LLNbO. Therefore, the total Li concentration was used to calculate the conductivity-derived diffusion coefficient (Ds). The calculated Ds values are shown in Fig. 3b. Visualization of the isotope distributions Fig. 4 shows TOF-SIMS images of the relative 6Li/(6Li + 7Li) intensity ratio near the interface (liquid level) in a single- crystalline LLNbO sample. Fig. 4a shows the isotope distribu- tion aer 6Li exchange at 300 K for 138 h, whereas Fig. 4b shows the distribution aer subsequent annealing at 523 K for 2 h. Following isotope exchange, a concentration gradient of 6Li is observed, demonstrating the replacement of 7Li by 6Li within the LLNbO crystal. Aer annealing, the concentration gradient becomes broader, indicating thermally activated bulk diffusion.ab-plane of a single-crystalline LLNbO sample, where the vertical axis 00 K for 138 h and (b) after subsequent annealing at 523 K for 2 h. The iffusion. This journal is © The Royal Society of Chemistry 2026Fig. 5 Schematic illustration of the procedure used to construct one-dimensional isotope concentration profiles from TOF-SIMS depth profiling. Paper Journal of Materials Chemistry ANotably, no features associated with grain boundary diffusion are observed, consistent with the single-crystalline nature of LLNbO. Large-area TOF-SIMS mapping performed before depth-prole analysis further conrmed the lateral uniformity of the isotope distribution over the analysed region (Fig. S3). One-dimensional analysis of tracer diffusion proles To quantify lithium tracer diffusion, one-dimensional isotope concentration proles were constructed from TOF-SIMS depthThis journal is © The Royal Society of Chemistry 2026proling, as illustrated in Fig. 5. Pointwise measurements were performed along the diffusion direction, starting from the interface with the 6Li-enriched solution (liquid level) and extending into the bulk crystal. At each position (e.g. points 2, 4, and 6 in Fig. 5), a TOF-SIMS depth prole was acquired, and the integrated 6Li and 7Li signal intensities were used to determine the local isotope fraction. This procedure was repeated for all three samples. Depth proling was conducted over a 50 × 50 mm2 area using 150 sputtering cycles (1 s per cycle),J. Mater. Chem. A, 2026, 14, 42453–42463 \| 42457Fig. 6 One-dimensional 6Li isotope concentration profiles of (a) Sample A, (b) Sample B and (c) Sample C after isotope exchange at 300 K for 47, 115, and 137 h, respectively. Fig. 7 (a) One-dimensional 6Li isotope concentration profile of Sample A after isotope exchange at 300 K for 47 h, followed by storage under ambient conditions for two weeks. The solid line represents the fitted profile based on eqn (7). (b and c) One-dimensional 6Li isotope concentration profiles of Samples B and C after annealing at different temperatures (300–623 K). The solid lines represent the fitted profiles based on eqn (7). Journal of Materials Chemistry A Papercorresponding to a total sputtering time of 150 s. The sputtering depth was calibrated by measuring the depth of a TOF-SIMS sputtered crater using a laser microscope aer sputtering under the same experimental conditions (Fig. S4). Based on the42458 \| J. Mater. Chem. A, 2026, 14, 42453–42463measured sputtering rate, the analysis depth aer 150 s was estimated to be approximately 60 nm. Representative depth proles for all samples and conditions are provided in Fig. S5– S14. The m/z spectrum (Fig. S15) conrms the clear detection of both 6Li and 7Li isotopes.This journal is © The Royal Society of Chemistry 2026Paper Journal of Materials Chemistry AThe local 6Li isotope fraction is calculated as:42 C ¼ I6Li I6Li þ I7Li (3) where I6Li and I7Li are the respective TOF-SIMS signal intensities. To determine the local isotope fraction, the 6Li and 7Li inten- sities were averaged over the sputtering time range of 60–120 s, corresponding to the stable region of the depth prole (Fig. 5). This range was selected to avoid surface-related artifacts and transient signal variations at the beginning of sputtering. Fig. 6 presents the one-dimensional 6Li isotope concentra- tion proles obtained aer isotope exchange at 300 K for 47 h (Sample A), 115 h (Sample B), and 137 h (Sample C). In all samples, the 6Li fraction decreases monotonically with increasing distance from the original liquid level, demon- strating inward tracer diffusion into the crystal. As expected for Fickian diffusion, the diffusion proles broaden with increasing exchange time, indicating progressive penetration of 6Li into the crystal. To obtain an initial estimate of the lithium tracer diffusion coefficient, the experimental concentration proles were analyzed using the one-dimensional analyticalCðx; tÞ ¼ 1 L ðL 0 f �  x 0 �  dx 0 þ 2 L XN n¼1 exp �  �  n2p2D1t1 L2 �  cos �npx L �  ðL 0 f �  x 0 �  cos &#32; npx 0 L ! dx 0 (6)solution of Fick's second law. Because the measurements were performed on single-crystalline LLNbO, grain-boundary diffu- sion is absent, as conrmed by the TOF-SIMS images (Fig. 4). Therefore, the analytical model describes lattice diffusion only. When a sample is brought into contact with an isotope reservoir of sufficiently large capacity, the concentration prole is described by the semi-innite solution to Fick's second law 43,44: CðxÞ �  C0 CS �  C0 ¼ erfc �  x 2 ffiffiffiffiffiffiffiffiffi D0t0 p �  (4) Here, CS is the isotope concentration in the diffusion source, C0 is the initial isotope concentration in the sample, D0 is the diffusion coefficient during the isotope exchange, and t0 is the exchange time. In the present experiment, however, only part of the sample was in contact with the isotope solution,43–45 allow- ing the concentration prole to be approximated by: CðxÞ ¼ 8\>\< \>: CS 0# x\\h ðCS �  C0Þerfc �  x�  h 2 ffiffiffiffiffiffiffiffiffi D0t0 p �  þ C0 h# x (5) where h is the distance from the sample edge to the liquid level. By tting eqn (5) to the experimental concentration proles, initial estimates of the lithium tracer diffusion coefficients were obtained for Samples A–C. Overall, the analytical solution reproduced the measured concentration proles reasonably well, indicating that one-dimensional Fickian diffusionThis journal is © The Royal Society of Chemistry 2026provides an appropriate rst approximation for lithium trans- port in single-crystalline LLNbO. However, as discussed below, this analytical model does not fully account for the nite sample geometry, and the diffusion coefficients were therefore further rened using two-dimensional numerical analysis. Following the room-temperature isotope exchange, Sample A was stored under ambient conditions for an additional two weeks to evaluate the temporal evolution of the diffusion prole at room temperature without further contact with the 6Li- enriched solution. In contrast, Samples B and C were annealed inside the TOF-SIMS chamber in the temperature range 300–673 K for 2 h at each temperature. Aer each annealing step, the samples were rapidly cooled to 163 K to suppress further diffusion. Because these post-exchange diffusion processes occurred without contact with an external isotope source, the constant surface concentration boundary condition used in eqn (5) was no longer applicable. Instead, the concentration proles were analyzed using the analytical solution of Fick's second law for a nite closed system of length L. For an arbitrary initial isotope distribution f(x), the general solution is given by43here, L is the sample length, D1 is the diffusion coefficient at the annealing temperature, and t1 is the annealing time. By substituting the initial prole obtained from eqn (5), the concentration prole is expressed as: Cðx; tÞ �  C0 CS �  C0 ¼ h L þ L�  h L erfc �  L�  h 2 ffiffiffiffiffiffiffiffiffi D0t0 p �  þ2 L ffiffiffiffiffiffiffiffiffi D0t0 p r ( 1�  exp " �  ðL�  hÞ2 4D0t0 \#) þ2 p XN n¼1 An n exp �  �  n2p2D1t1 L2 �  cos �npx L �  (7) The coefficient An is dened as: An ¼ 2ffiffiffiffi p p ð L�h 2 ffiffiffiffiffiffiffi D0t0 p 0 dxexp ��x2�sin np �  2 ffiffiffiffiffiffiffiffiffi D0t0 p �  xþ h �  L &#32; (8) with L, h, D0 and t0 xed, the coefficient An is uniquely deter- mined for each value of n. Accordingly, D1 was the only adjustable tting parameter. The parameters CS, C0, h, D0 and t0, obtained by tting the post-exchange prole with eqn (5), were xed during the anal- ysis, and the sample length was set to L = 4.7 mm. The total amount of 6Li in the sample, estimated from the post-exchange proles, was conserved during annealing, supporting the use of the prole obtained from eqn (5) as the initial condition for theJ. Mater. Chem. A, 2026, 14, 42453–42463 \| 42459Fig. 8 Experimental 6Li isotope concentration profiles and corresponding two-dimensional numerical simulations for Samples A–C. (a, c and e) Profiles immediately after isotope exchange at 300 K for Samples A, B, and C, respectively. The dashed lines indicate the isotope distributions immediately after removal from the isotope source, whichwere used as the initial conditions for the one-hour diffusion simulations shown by the solid lines. The one-hour diffusion time corresponds approximately to the time between removal from the isotope source and the start of the TOF-SIMS measurement. (b, d and f) Temperature-dependent isotope concentration profiles for Samples A, B, and C, respectively. Filled circles and solid lines represent the experimental data and the corresponding two-dimensional numerical simulations, respectively. The simulations assumed that the lithium diffusion coefficient along the thickness direction (\[001\]) was one-third of that along the in-plane direction, based on the ionic-conductivity anisotropy. The diffusion coefficients were determined by fitting the simulated concentration profiles to the experimental data. Journal of Materials Chemistry A Papersubsequent diffusion analysis. This analytical approach accounts for the nite sample geometry during post-exchange annealing.43,46,47 Fig. 7a compares the isotope concentration proles of Sample A measured immediately aer isotope exchange and aer subsequent storage for two weeks at room temperature under ambient conditions. The tted proles reproduced the experimental concentration proles well and were used to determine the diffusion coefficients. The prole broadening observed aer storage indicates that lithium redistribution42460 \| J. Mater. Chem. A, 2026, 14, 42453–42463continues within the crystal under the internal concentration gradient even aer removal from the isotope solution. Fig. 7b and c show the experimental and tted 6Li proles for Samples B and C aer annealing at different temperatures. The tted proles reproduce both the concentration gradients and diffusion length observed experimentally. Furthermore, at elevated temperatures, the isotope concentration proles become atter owing to enhanced lithium diffusion, resulting in increased uncertainty in the tted diffusion coefficients. Nevertheless, the concentration proles remain sufficientlyThis journal is © The Royal Society of Chemistry 2026Table 1 Comparison of the lithium tracer diffusion coefficients in LLNbO obtained from the one-dimensional analytical fitting and the two- dimensional numerical simulation of the TOF-SIMS isotope concentration profiles. The average diffusion coefficient obtained from four measurements at 300 K is 9.3 × 10−13 m2 s−1, with a relative standard deviation of approximately 20%. This variation provides an estimate of the experimental uncertainty in the room-temperature diffusion measurements Sample Temperature (K) Diffusion coefficient from 1D analytical tting (m2 s−1) Diffusion coefficient from 2D numerical simulation (m2 s−1) Sample A 300 4.5 × 10−13 7.5 × 10−13 300 (for two weeks) 7.4 × 10−13 7.4 × 10−13 Sample B 300 1.2 × 10−12 1.2 × 10−12 373 1.5 × 10−11 1.5 × 10−11 523 3.1 × 10−10 3.1 × 10−10 Sample C 300 7.4 × 10−13 1.0 × 10−12 373 2.4 × 10−11 1.4 × 10−11 443 6.7 × 10−11 6.7 × 10−11 500 1.8 × 10−10 1.8 × 10−10 623 6.2 × 10−10 6.2 × 10−10 Paper Journal of Materials Chemistry Awell-dened for reliable diffusion analysis. These isotope concentration proles served as the basis for the subsequent two-dimensional numerical analysis described in the following section. Two-dimensional numerical diffusion analysis Although the one-dimensional analytical solution provides a useful rst approximation, nite-size effects associated with the actual sample geometry are not explicitly considered. To evaluate their inuence on the measured isotope concentration proles, a two-dimensional numerical analysis was performed. Details of the simulation model, numerical procedure, initial conditions, and boundary conditions are provided in the SI. Representative comparisons between the experimental isotope concentration proles and the corresponding two- dimensional numerical ts are shown in Fig. 8. The numer- ical model successfully reproduced both the shape and the diffusion length of the experimental isotope concentration proles. The diffusion anisotropy adopted in the numericalFig. 9 (a) Arrhenius plot of the lithium tracer diffusion coefficients (DSI isotope concentration profiles, together with the conductivity diffusio LLNbO. The solid line represents the Arrhenius fit to the low-tempera (b) Temperature dependence of the Haven ratio calculated from the refi coefficients. This journal is © The Royal Society of Chemistry 2026analysis was estimated from the ionic conductivity anisotropy measured for Sample C (Fig. S16). Because Samples A and C had immersion depths comparable to the diffusion length, nite- size effects became signicant and the diffusion coefficients obtained from the two-dimensional analysis were larger than those obtained from the one-dimensional analytical solution. In contrast, Sample B, which had a shallower immersion depth, showed essentially the same diffusion coefficient in both analyses. These results indicate that the one-dimensional analytical solution tends to underestimate the diffusion coefficient when nite-size effects become signicant, whereas it provides reli- able estimates when such effects are negligible. The diffusion coefficients rened by the two-dimensional numerical analysis are summarized in Table 1 and were used for the Arrhenius analysis discussed below. For Sample A, the same diffusion coefficient also successfully reproduced the isotope concentra- tion prole measured aer two weeks of storage, further sup- porting the validity of the two-dimensional numerical analysis.MS) obtained by two-dimensional numerical analysis of the TOF-SIMS n coefficients (Ds) derived from ionic conductivity measurements of ture diffusion coefficients, yielding an activation energy of 0.33 eV. ned lithium tracer diffusion coefficients and the conductivity diffusion J. Mater. Chem. A, 2026, 14, 42453–42463 \| 42461Journal of Materials Chemistry A PaperThe two-dimensional numerical analysis also led to a revi- sion of the diffusion coefficient for Sample C at 373 K aer removal from the isotope source. During the relatively short diffusion period aer sample removal, the isotope concentra- tion near the region previously in contact with the isotope source can decrease not only by in-plane diffusion but also by diffusion in the thickness direction. When such a prole is analysed using a one-dimensional diffusion model, this addi- tional concentration decrease can be interpreted as faster in- plane diffusion, leading to an overestimation of the in-plane diffusion coefficient. By explicitly accounting for diffusion in the thickness direction, the two-dimensional analysis therefore gives a lower diffusion coefficient at 373 K than the one- dimensional analysis. The rened lithium tracer diffusion coefficients obtained from the two-dimensional numerical analysis are compared with the conductivity diffusion coefficients derived from ionic conductivity measurements in Fig. 9a. Within the investigated temperature range, the two diffusion coefficients are in good agreement and exhibit similar temperature dependencies, although only a limited number of tracer diffusion data points were obtained at elevated temperatures. The resulting Haven ratios (DSIMS/Ds)48,49 are shown in Fig. 9b. No clear temperature dependence was observed over the investigated temperature range, although some scatter was present. The average Haven ratio was 0.95, indicating that the tracer and conductivity diffusion coefficients are generally comparable. Although this value is higher than the typical Haven ratio of∼0.5 reported for conventional oxide electrolytes, similar behaviour has been reported for perovskite-type solid electrolytes with low Li concentrations,10 where Li ions are considered to migrate in a largely uncorrelated manner. LLNbO has a double-perovskite structure, in which La and Li occupy A-sites within the same layer. For the composition x = 0.08, the occupancies of La, Li, and vacancies on the A-sites in this layer are approximately 0.61, 0.16, and 0.23, respectively. Considering only Li-accessible sites, the vacancy fraction is estimated to be approximately 59%, indicating that more than half of the Li-accessible sites are vacant. This low Li occupancy and high vacancy fraction are expected to reduce correlation effects during Li-ion migration, which is consistent with the observed Haven ratio close to unity. Within the investigated temperature range, the tracer diffu- sion coefficients exhibit a temperature dependence consistent with that of the low-temperature conductivity diffusion coeffi- cients, which yield an activation energy of 0.33 eV. The good agreement between the tracer and conductivity diffusion coef- cients further supports the validity of the present diffusion analysis. Conclusion Lithium tracer diffusion in single-crystalline LLNbO was systematically investigated by combining 6Li isotope exchange with TOF-SIMS depth proling performed at multiple positions along the diffusion direction. To improve the accuracy of the diffusion analysis, the isotope concentration proles were42462 \| J. Mater. Chem. A, 2026, 14, 42453–42463analyzed using both one-dimensional analytical tting and two- dimensional numerical simulation that accounts for nite-size effects. The rened tracer diffusion coefficients obtained from the two-dimensional numerical analysis showed good agree- ment with the conductivity diffusion coefficients derived from ionic conductivity measurements. The corresponding Haven ratio was close to unity, indicating largely uncorrelated lithium- ion migration in LLNbO. The high vacancy fraction on the Li- accessible A-sites is considered to reduce correlation effects during lithium-ion migration, providing a structural origin for the observed Haven ratio. The present methodology, combining lithium isotope exchange, TOF-SIMS proling, and two- dimensional numerical analysis, provides a reliable and quan- titative approach for determining lithium tracer diffusion coefficients, particularly in fast lithium-ion conductors where nite-size effects become signicant.Author contributions Prince Sharma performed the isotope exchange experiments, conducted TOF-SIMS measurements and analysis, and prepared the initial dra of the manuscript. Gen Hasegawa contributed to TOF-SIMS measurements, data analysis, and critical revision of the manuscript. Kaori Sugii performed the temperature-dependent ionic conductivity measurements. Yasuyuki Fujiwara prepared the single-crystal samples, contributed to compositional analysis, and participated in data interpretation and discussion. Hideki Iba contributed to project planning, data interpretation, and scientic discussions. Naoaki Kuwata conceived and supervised the project, contrib- uted to data interpretation, and revised and nalized the manuscript.Conflicts of interest The authors declare no competing nancial interest.Data availability The data supporting the ndings of this study are available within the article and its supplementary information (SI). Additional raw data, including TOF-SIMS depth proles and isotope diffusion datasets, are available from the corresponding author upon reasonable request. Supplementary information is available. See DOI: https://doi.org/10.1039/d6ta03003c.Acknowledgements This study is supported by the JST COI-NEXT “Center for Advanced Battery Collaboration” (grant number JPMJPF2016), JSPS KAKENHI (grant number 24KF0271), and JST GteX Program (grant number JPMJGX23S2). This work is partly sup- ported by JSPS KAKENHI (grant number 24H02205 (Ion Jamology)), NIMS All-Solid-State Battery Materials Open Plat- form (MOP), and SOLiD-Next (JPNP23005), subsidized by the New Energy and Industrial Technology DevelopmentThis journal is © The Royal Society of Chemistry 2026[https://doi.org/10.1039/d6ta03003c](https://doi.org/10.1039/d6ta03003c)Paper Journal of Materials Chemistry AOrganization (NEDO). The TOF-SIMS and XRD measurements are supported by the NIMS Battery Research Platform. References 1 C. Li, Z. Yu Wang, Z. Jiang He, Y. Jiao Li, J. Mao, K. Hua Dai, C. Yan and J. Chao Zheng, Sustain. Mater. Technol., 2021, 29, e00297. 2 J. Janek and W. G. Zeier, Nat. Energy, 2023, 8, 230–240. 3 W. D. Richards, L. J. Miara, Y. Wang, J. C. Kim and G. Ceder, Chem. Mater., 2016, 28, 266–273. 4 Y. Xiao, Y. Wang, S. H. Bo, J. C. Kim, L. J. Miara and G. Ceder, Nat. Rev. Mater., 2020, 5, 105–126. 5 J. Maier, Nat. Mater., 2005, 4, 805–815. 6 Springer Series in Solid-State Sciences, ed. H. Mehrer, Springer Berlin Heidelberg, Berlin, Heidelberg, 2007, vol. 155, pp. 1–23. 7 X. Hu, C. A. J. Fisher, S. Kobayashi, Y. H. Ikuhara, Y. Fujiwara, K. Hoshikawa, H. Moriwake, K. Kohama, H. Iba and Y. Ikuhara, Acta Mater., 2018, 156, 379–388. 8 H. Lian, X. Hu, D. Xiong, S. Tao, B. Zhong, B. Song, W. Deng, H. Hou, G. Zou and X. Ji, Energy Storage Mater., 2024, 71, 103593. 9 D. Wang, H. Shi, S. Wang, X. Wu, W. Jiang, S. Liang and Z. Xu, Coord. Chem. Rev., 2024, 508, 215776. 10 G. Hasegawa, N. Kuwata, K. Hashi, Y. Tanaka and K. Takada, Chem. Mater., 2023, 35, 3815–3824. 11 G. Hasegawa, N. Kuwata, T. Ohnishi and K. Takada, J. Mater. Chem. A, 2023, 12, 731–738. 12 I. O. Santos-Mendoza, J. Vázquez-Arenas, I. González, G. Ramos-Sánchez and C. O. Castillo-Araiza, Int. J. Chem. React. Eng., 2019, 17, 20180095. 13 B. Tian, J. Światowska, V. Maurice, S. Zanna, A. Seyeux, L. H. Klein and P. Marcus, J. Phys. Chem. C, 2013, 117, 21651–21661. 14 V. F. Lvovich, Impedance Spectroscopy: Applications to Electrochemical and Dielectric Phenomena, John Wiley \& Sons, Inc., Hoboken, NJ, USA, 2012. 15 X. Liu, Z. Liang, Y. Xiang, M. Lin, Q. Li, Z. Liu, G. Zhong, R. Fu and Y. Yang, Adv. Mater., 2021, 33, 2005878. 16 F. Blanc, M. Leskes and C. P. Grey, Acc. Chem. Res., 2013, 46, 1952–1963. 17 M. Matsuura, Y. Fujiwara, H. Moriwake, K. Ohara and Y. Kawakita, Phys. Rev. B, 2021, 104, 94305. 18 J. F. M. Oudenhoven, L. Baggetto and P. H. L. Notten, Adv. Energy Mater., 2011, 1, 10–33. 19 R. Baddour-Hadjean and J. P. Pereira-Ramos, AIP Conf. Proc., 2010, 1267, 1137–1138. 20 J. Zhu, L. Lu and K. Zeng, ACS Nano, 2013, 7, 1666–1675. 21 P. Sharma, G. Hasegawa, S. Xing and N. Kuwata, Sci. Technol. Adv. Mater., 2025, 26, 2580917. 22 D. Uxa, E. Hüger, K. Meyer, L. Dörrer and H. Schmidt, Chem. Mater., 2023, 35, 3307–3315. 23 J. N. Coles and J. V. P. Long, Philos. Mag., 1974, 29, 457–471. 24 K. Watanabe, I. Sakaguchi, S. Hishita, N. Ohashi and H. Haneda, Appl. Phys. Express, 2011, 4, 55801. 25 N. Molina, A. Dolocan, J. F. Curry, G. J. Rodin and F. Mangolini, Anal. Chem., 2026, 98, 5134–5140.This journal is © The Royal Society of Chemistry 202626 S. Xing, G. Hasegawa, P. Sharma, D. Ito, K. Kamiguchi, K. Takada and N. Kuwata, ACS Energy Lett., 2026, 11, 1889– 1896. 27 W. Araki, Y. Nagakura and Y. Arai, Ceram. Int., 2020, 46, 6270–6275. 28 M. S. Ali, N. Sato, I. Fukasawa, Y. Maruyama, M. Nagao, S. Watauchi and I. Tanaka, Cryst. Growth Des., 2019, 19, 6291–6295. 29 Z. Yang, R. E. Ward, N. Tanibata, H. Takeda, M. Nakayama and T. Asaka, J. Phys. Chem. C, 2020, 124, 9746–9754. 30 T. T. Choudhury, W. Araki, N. Yamada and Y. Arai, Mater. Lett., 2021, 302, 130356. 31 X. Hu, C. A. J. Fisher, S. Kobayashi, Y. H. Ikuhara, Y. Fujiwara, K. Hoshikawa, H. Moriwake, K. Kohama, H. Iba and Y. Ikuhara, Acta Mater., 2017, 127, 211–219. 32 S. Garćıa-Mart́ın, J. M. Rojo, H. Tsukamoto, E. Morán and M. A. Alario-Franco, Solid State Ionics, 1999, 116, 11–18. 33 J. Ibarra, A. Várez, C. León, J. Santamaŕıa, L. M. Torres- Mart́ınez and J. Sanz, Solid State Ionics, 2000, 134, 219–228. 34 E. Garćıa-González, R. Maŕın-Gamero, M. Kuhn-Gómez, A. Kuhn, F. Garćıa-Alvarado and S. Garćıa-Mart́ın, J. Mater. Chem. A, 2024, 12, 28247–28253. 35 Y. Fujiwara, T. Taishi, K. Hoshikawa, K. Kohama and H. Iba, Jpn. J. Appl. Phys., 2016, 55, 90306. 36 M. S. Ali, Y. Maruyama, M. Nagao, S. Watauchi and I. Tanaka, Solid State Ionics, 2020, 350, 115330. 37 S. Takai, T. Mandai, Y. Kawabata and T. Esaka, Solid State Ionics, 2005, 176, 2227–2233. 38 Y. Fujiwara, T. Taishi and H. Iba, in Interface Ionics: For All- Solid-State Batteries and Solid State Ionics Devices, Springer, 2024, pp. 151–160. 39 X. Hu, S. Kobayashi, Y. H. Ikuhara, C. A. J. Fisher, Y. Fujiwara, K. Hoshikawa, H. Moriwake, K. Kohama, H. Iba and Y. Ikuhara, Acta Mater., 2017, 123, 167–176. 40 Y. Fujiwara, K. Hoshikawa and K. Kohama, Growth of solid electrolyte La(1−x)/3LixNbO3 single crystals by the directional solidication method, J. Cryst. Growth, 2016, 433, 48–53. 41 X. Gao, C. A. J. Fisher, Y. H. Ikuhara, Y. Fujiwara, S. Kobayashi, H. Moriwake, A. Kuwabara, K. Hoshikawa, K. Kohama, H. Iba and Y. Ikuhara, J. Mater. Chem. A, 2015, 3, 3351–3359. 42 Y. Sun, G. M. Hughes, J. Bu, J. Liu, C. R. M. Grovenor and P. S. Grant, Micron, 2025, 190, 103746. 43 J. Crank, The Mathematics of Diffusion, Oxford University Press, 2nd edn, 1979. 44 H. Mehrer, Diffusion in Solids: Fundamentals, Methods, Materials, Diffusion-Controlled Processes, Springer Science \& Business Media, 2007, vol. 155. 45 N. Kuwata, G. Hasegawa, D. Maeda, N. Ishigaki, T. Miyazaki and J. Kawamura, J. Phys. Chem. C, 2020, 124, 22981–22992. 46 U. M. Gosele, Annu. Rev. Mater. Sci., 1988, 18, 257–282. 47 R. B. Fair, J. Electrochem. Soc., 1981, 128, 1360–1368. 48 A. Van Der Ven, Z. Deng, S. Banerjee and S. P. Ong, Chem. Rev., 2020, 120, 6977–7019. 49 G. E. Murch, Solid State Ionics, 1982, 7, 177–198.J. Mater. Chem. A, 2026, 14, 42453–42463 \| 42463- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS- Direct observation and quantitative analysis of lithium diffusion in LLNbO single crystals via isotope tracing and TOF-SIMS