REVIEW 4 major objections 5 minor 6 references
Observation of Order and Disorder in Solid-Electrolyte Interphases of Lithium-Metal Anodes
T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read The solid-electrolyte interphase in efficient lithium-metal anodes contains a distinct short-range-ordered amorphous phase that is absent in low-efficiency, dendrite-forming electrolytes.
desk verdict A well-executed cryo-SEND study with a likely real structural distinction that overstates its central short-range-order peak and its causal claim. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the radially averaged diffraction curve from the SEI, displayed with the intensity multiplied by the scattering vector $Q$ (the $I\times Q$ enhancement) to expose weak broad peaks. A broad first peak near 0.33 inverse angstroms is interpreted as short-range order in an amorphous phase and defines SEI Type 1; a second pattern adds anisotropic rings from defective LiF nanocrystals and defines SEI Type 2. Mapping the position of the maximum scattering peak pixel by pixel separates three populations: SEI Type 1 near 0.33 inverse angstroms, SEI Type 2 near 0.295 inverse angstroms, and the vitrified electrolyte near 0.285 inverse angstroms. That peak position is what carries the argument: it allows the authors to distinguish the two SEI types, locate them in layered maps, and correlate the presence of Type 1 short-range order with high Coulombic efficiency.
What would settle it
Prepare the same high-concentration electrolyte SEI two ways—cryo-FIB lift-out and direct electrodeposition on a TEM grid—and record radially averaged diffraction at decreasing doses. If the 0.33 inverse angstrom peak appears only after FIB thinning, or if it transforms into Li2O before a structural map can be completed, then the claimed short-range order is a preparation or dose artifact rather than the native SEI feature that the paper's conclusion rests on.
Extended reading notes
Core claim
Under cryogenic focused-ion-beam lift-out and scanning electron nanobeam diffraction at low dose, the SEI at a copper/lithium interface in a high-concentration electrolyte (4.6 m LiFSI plus 2.3 m LiTFSI in DME) separates into two structurally distinct layers: a roughly 15 nm Type 1 layer next to lithium that shows only short-range order, with broad radially averaged peaks near 0.33 and 0.6 inverse angstroms, and a roughly 35 nm Type 2 layer next to copper that shows the same short-range order plus defective LiF nanocrystals, signaled by a weak (111) reflection and peaks near 0.5 and 0.7 inverse angstroms. At the electrolyte/lithium interface, only Type 1 is found, distributed uniformly. Electron energy loss spectroscopy shows Type 1 is essentially carbon-free, bonded through Li, O, F, and S, with a diffuse Li-K edge unlike crystalline Li2O or LiF, while Type 2 has distinct fine structures consistent with LiF. In 1 M LiPF6 in EC/DEC and 1 M LiFSI in DME, which deposit dendritic lithium and cycle with lower Coulombic efficiency, the SEI lacks this strong short-range order. The paper's claim is that the short-range-ordered Type 1 phase is the key structural signature of a stable, high-performance SEI and that SEI structure directly influences deposition morphology and battery performance.
Load-bearing premise
The load-bearing premise is that the broad diffraction peak near 0.33 inverse angstroms is a genuine feature of the native SEI rather than an artifact of the measurement, the specimen preparation, or the electron beam; if that peak is not real, the Type 1/Type 2 distinction and its link to battery performance collapse.
Editorial extensions
If this is right
- In the high-concentration electrolyte, the SEI at the lithium/copper interface is layered: Type 2 forms first in the less reducing environment before lithium nucleation, and Type 1 forms preferentially in the more reducing environment at the lithium surface.
- Because Type 1 short-range order is carbon-free and appears only in the high-efficiency electrolyte, anion-derived inorganic SEIs can serve as a structural marker for electrolyte formulations that suppress dendrites.
- The short-range-ordered amorphous SEI is expected to provide homogenized, isotropic lithium-ion diffusion and a higher elastic modulus than either fully crystalline or fully disordered films, which would explain the coarse, non-dendritic lithium grains observed at 0.5 mA/cm².
- SEI layers in operating cells are thicker (about 40–60 nm) than those seen in earlier cryo-EM studies, because cryo-FIB lift-out captures the SEI at full capacity rather than the early-stage films formed on TEM grids.
- The measured dose threshold of about 3600 electrons per square angstrom for converting Type 1 short-range order into Li2O implies that many earlier phase-contrast images of SEIs may have recorded beam-modified structures, so low-dose diffraction should be the default characterization route.
Reading between the lines
- If short-range order is causal, not just correlated, then electrolytes that form a carbon-free, short-range-ordered SEI—localized high-concentration formulations, fluorinated solvents, or salt mixtures that decompose to Li–O–F–S phases—should also produce high Coulombic efficiency and coarse lithium grains; this is a testable prediction beyond the three electrolytes studied.
- The resemblance to disordered solid electrolytes suggests a design axis the paper only sketches: salt concentration and anion chemistry could tune the SEI's stiffness and ionic conductivity simultaneously, so the optimum SEI may be one with an intermediate degree of order rather than maximum crystallinity or maximum disorder.
- A systematic low-dose dose-series across several electrolytes would show how much of the historical cryo-EM SEI literature is beam artifact; the paper's own threshold for Type 1 makes that re-examination straightforward.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This manuscript reports a cryogenic scanning electron nanobeam diffraction (cryo-SEND) and cryo-EELS study of solid-electrolyte interphases (SEIs) formed on lithium-metal anodes in a high-concentration LiFSI/LiTFSI/DME electrolyte. Using cryo-FIB lift-out lamellae, the authors identify two SEI structures: Type 1, an amorphous phase with a short-range-order diffraction peak near Q ≈ 0.33 Å⁻¹, and Type 2, a mixed region containing Type 1-like short-range order plus defective LiF nanocrystals. They further map these phases as a layered structure at the Li/Cu interface and as a uniform Type 1 layer at the electrolyte/Li interface. For two lower-concentration electrolytes (1 M LiPF6 in EC/DEC and 1 M LiFSI in DME), they report dendritic Li morphology and weaker short-range order, and they correlate the presence of Type 1 short-range order with higher Coulombic efficiency. The paper concludes that SEI structure directly influences lithium deposition morphology and battery performance.
Significance. If the structural assignment is correct, the work would be significant: it demonstrates a low-dose cryo-4D-STEM methodology for beam-sensitive battery interfaces, includes beam-damage control experiments, and reports a specific amorphous SEI motif (Type 1 short-range order) that is correlated with high Coulombic efficiency. The manuscript also provides a data-availability statement with a Zenodo DOI, which is a constructive step for reproducibility. The central novelty, however, rests on the interpretation of a weak low-Q diffraction feature; until the raw-data analysis is shown, the taxonomic and causal claims remain provisional.
major comments (4)
- [Figure 2c/2d] Figure 2c and 2d: the short-range-order peak at ≈0.33 Å⁻¹, which defines Type 1, is shown only after I×Q enhancement; the unmultiplied intensity I(Q), the background subtraction, and the direct-beam exclusion are not presented. The text in the discussion of Figure 2d states that the uneven SEI/electrolyte contour caused the beam to superimpose diffraction from both moieties, and the vitrified electrolyte itself has a peak at ≈0.285 Å⁻¹. Without a raw-data reanalysis showing that the 0.33 Å⁻¹ feature survives background removal and is spatially separated from the electrolyte contribution, the existence of a distinct Type 1 short-range order is not established. Please provide the raw radially averaged profiles and a quantitative background model.
- [Figure 2c and Figure 3] The length-scale assignment of the 0.33 Å⁻¹ peak is not supported by the proposed inorganic Li-O/F/S network: d = 2π/Q ≈ 19 Å is much larger than Li-O, Li-F, or Li-S nearest-neighbor distances, and this length scale is naturally produced by anion aggregates in concentrated electrolytes. To distinguish genuine SEI short-range order from electrolyte aggregate ordering, the authors should provide a real-space pair distribution function or a simulated diffraction pattern of the proposed network, rather than relying only on peak position in the I×Q-enhanced curve.
- [Figure 4b/4c and Abstract] The causal claim in the abstract and conclusion (SEI structure "directly influences" deposition morphology and performance) is stronger than the evidence. The comparison is between three electrolytes that differ in concentration, salt, and solvent; the Coulombic-efficiency values are reported without error bars, replicate counts, or statistical tests; and the structural metric (short-range-order peak presence or intensity) is not quantitatively regressed against efficiency. The data support a correlation, not a causal link, and the conclusion should be tempered accordingly.
- [Figure S8 and beam-damage controls] The paper states that Type 1 short-range order is destroyed at ≈3600 e⁻/Ų and that EELS fine structure changes above 400 e⁻/Ų, but it does not report the accumulated dose per SEND pixel or per EELS spectrum. Without this dose information, the claim that the measured Type 1 short-range order is native rather than beam-induced cannot be fully evaluated. Please add a table or explicit statement of the electron doses used for each measurement.
minor comments (5)
- [Methods/Notation] Please define the scattering-vector convention used throughout (e.g., Q = 4π sinθ/λ) and state whether the displayed Q is in Å⁻¹ or nm⁻¹ consistently; Figure 1c uses 5 nm⁻¹ while Figures 2 and 4 use Å⁻¹.
- [Main text, convergence angle] There is a typo in the main text: "0.l5 mrad" should be "0.15 mrad."
- [Figure 2e/2f] For the Gaussian fits to the histograms, please report the bin width, the number of pixels in each region, and the fitting procedure, since the two-component interpretation depends on this histogram analysis.
- [References] Reference 24 and reference 25 are identical (Qian et al., Nat Commun 2015); please renumber or remove the duplicate.
- [Figure 3d] For the Li-K edge reference spectra, please identify the exact compounds and the acquisition conditions; the current caption does not state whether these are measured standards or literature spectra.
Circularity Check
No circular derivation: SEI-type classification and Coulombic-efficiency measurements are independent, with only minor contextual self-citations.
full rationale
The paper's central claim is an empirical correlation, not a derivation. SEI Type 1 and Type 2 are assigned from radially averaged cryo-SEND peak positions (Type 1 first peak ≈0.33 Å^-1; Type 2 first peak ≈0.3 Å^-1 with LiF reflections at 0.5/0.7 Å^-1), while Coulombic efficiencies are measured independently using standard-cycle and Aurbach methods (96.3%/98.2% for the high-concentration electrolyte versus lower values for 1 M LiFSI/DME and 1 M LiPF6 in EC/DEC). No equation fits CE from the diffraction peaks or vice versa, so the structure-performance link is not forced by construction. The two self-citations (refs 10 and 19) are used only to explain beam-induced Li2O artifacts and prior beam-damage context; they are not load-bearing for the Type 1/Type 2 taxonomy or the CE correlation, and no uniqueness theorem is imported from the authors' prior work. The I×Q enhancement is a standard display method, and the paper explicitly distinguishes the SEI Type 1 peak (0.33 Å^-1) from the vitrified-electrolyte aggregate peak (0.285 Å^-1). The admitted superposition at the uneven electrolyte/Li interface ('the electron beam to measure both components along the beam axis, superimposing the diffraction intensities from the two moieties') is a real measurement limitation for the top-interface assignment, but it is not circular: the bottom SEI Type 1 exists between Li and Cu away from bulk electrolyte, and the CE values come from independent cell cycling. The strongest claim is an inductive correlation over three electrolytes, not a self-defined prediction, so the circularity score is low.
Assumptions & free parameters
assumptions (4)
- domain assumption Cryogenic FIB lift-out and vitrification preserve the native SEI structure.
- domain assumption Multiplying the diffracted intensity by Q (I×Q, n=1) reveals or preserves true peak positions without introducing artifacts.
- domain assumption The Coulombic efficiency measurements for the three electrolytes are representative and comparable.
- domain assumption The electrolyte ordering peak at about 0.285 inverse angstroms originates from anion aggregates, as cited from prior work.
Cite this review
Pith. "Pith review of Observation of Order and Disorder in Solid-Electrolyte Interphases of Lithium-Metal Anodes." pith.science (2026). https://pith.science/paper/VVS2OWLT
@misc{pith2026250503956,
author = {Pith},
title = {Pith review of: Observation of Order and Disorder in Solid-Electrolyte Interphases of Lithium-Metal Anodes},
year = {2026},
howpublished = {\url{https://pith.science/paper/VVS2OWLT}},
note = {Machine review of arXiv:2505.03956}
}
read the original abstract
Battery interfaces critically influence lithium-metal battery performance through their role in ion diffusion and dendrite formation. However, structural characterization of these interfaces has remained challenging due to limitations in high-resolution methods and artifacts from electron irradiation. Using cryogenic conditions for both specimen preparation and scanning electron nanobeam diffraction, we can determine the structural organization at the interface between the vitrified electrolyte and adjacent layers. We identified two distinct interface types: one showing short-range order adjacent to lithium metal, and another displaying a mixed structure of short-range ordering and defective lithium fluoride nanoscale crystallites at a copper collector. Notably, short-range order appeared exclusively in electrolytes demonstrating high reversibility. Our results establish that solid-electrolyte-interphase structure directly influences lithium deposition morphology and battery performance. This methodology opens new possibilities for high-resolution characterization of interfaces in energy storage materials, advancing our understanding of their critical structural properties.
Figures
Reference graph
Works this paper leans on
-
[5]
https://doi.org/10.1038/ncomms7362. (26) Tang, W. S.; Yoshida, K.; Soloninin, A. V .; Skoryunov, R. V .; Babanova, O. A.; Skripov, A. V .; Dimitrievska, M.; Stavila, V .; Orimo, S.; Udovic, T. J. Stabilizing Superionic- Conducting Structures via Mixed-Anion Solid Solutions of Monocarba-Closo-Borate Salts. ACS Energy Lett. 2016, 1 (4), 659–664. https://doi...
-
[510]
https://doi.org/10.1126/science.aam6014. (22) Doniach, S. Changes in Biomolecular Conformation Seen by Small Angle X-Ray Scattering. Chem. Rev. 2001, 101 (6), 1763–1778. https://doi.org/10.1021/cr990071k. (23) Xie, Y .; Wang, J.; Savitzky, B. H.; Chen, Z.; Wang, Y .; Betzler, S.; Bustillo, K.; Persson, K.; Cui, Y .; Wang, L.-W.; Ophus, C.; Ercius, P.; Zhe...
-
[794]
https://doi.org/10.1039/C8EE02601G. (14) Jungjohann, K. L.; Gannon, R. N.; Goriparti, S.; Randolph, S. J.; Merrill, L. C.; Johnson, D. C.; Zavadil, K. R.; Harris, S. J.; Harrison, K. L. Cryogenic Laser Ablation Reveals Short- Circuit Mechanism in Lithium Metal Batteries. ACS Energy Lett. 2021, 6 (6), 2138–2144. https://doi.org/10.1021/acsenergylett.1c0050...
-
[1611]
https://doi.org/10.1149/1.2100722. (6) Peled, E. Advanced Model for Solid Electrolyte Interphase Electrodes in Liquid and Polymer Electrolytes. J. Electrochem. Soc. 1997, 144 (8), L208. https://doi.org/10.1149/1.1837858. (7) Zachman, M. J.; Tu, Z.; Choudhury, S.; Archer, L. A.; Kourkoutis, L. F. Cryo-STEM Mapping of Solid–Liquid Interfaces and Dendrites i...
-
[6362]
https://doi.org/10.1038/ncomms7362. (25) Qian, J.; Henderson, W. A.; Xu, W.; Bhattacharya, P.; Engelhard, M.; Borodin, O.; Zhang, J.-G. High Rate and Stable Cycling of Lithium Metal Anode. Nat Commun 2015, 6 (1),
-
[6576]
https://doi.org/10.1111/jace.19327. (32) Marple, M. A. T.; Wynn, T. A.; Cheng, D.; Shimizu, R.; Mason, H. E.; Meng, Y . S. Local Structure of Glassy Lithium Phosphorus Oxynitride Thin Films: A Combined Experimental and Ab Initio Approach. Angewandte Chemie International Edition 2020, 59 (49), 22185– 22193. https://doi.org/10.1002/anie.202009501. (33) Chou...
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.