{"id":"3d05549d-0083-4990-b780-24cd617ddba1","arxiv_id":"2507.04574","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Electrolyte wetting of fracture surfaces in cathode materials enhances lithium (de)intercalation and drives additional bulk cracking, and in simulations these wetted cracks raise first-cycle capacity and Coulombic efficiency.","lead":"A combined experimental and simulation study shows that liquid electrolyte seeping into cracks in battery cathode crystals speeds up lithium release and changes how cracks grow. The finding suggests controlled cracking could boost first-cycle capacity, though long-term damage remains a concern.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The wetting source term (Eq. 18) activates on partially damaged material (d<1) ahead of the crack tip, not only on open crack surfaces; the SI benchmark tests only d=1, so the wetting-fracture feedback may be overestimated.","rationale":"The reader's weakest assumption is that crack surfaces carry the same flux J* as external boundaries, with no transport or opening dependence. I agree that this is a key uncertainty, but I locate the most load-bearing problem one step earlier: the smeared source in Eq. (18) is active wherever d>0, including partially damaged material ahead of the crack tip. The SI benchmark only validates the regularization for a fixed, fully developed crack (d=1), so the evolving-crack case is untested. This could inflate the wetting–fracture feedback that underlies the capacity and Coulombic-efficiency predictions. The proposed Heaviside modification and length-scale sweep would settle whether the coupling is physical or a regularization effect. Since the paper is already judged CONDITIONAL, my concern does not change the verdict; it sharpens the condition that should be met before the central claim is accepted.","tokens_in":25784,"tokens_out":7226,"duration_ms":94208,"concrete_test":"Re-run the Sample-1 geometry (Figs. 2–3) and the two-grain NCM benchmark (Fig. 6) with Eq. (18) modified so that the source is multiplied by a Heaviside factor H(d − 0.9) (or, equivalently, concentrated on the d ≈ 1 iso-surface), keeping all other parameters fixed. If the wetting-induced bulk crack in Fig. 3(f) and the capacity/CE gains in Fig. 6(b) disappear or change qualitatively, the smeared source's activation at d<1 is load-bearing. As a robustness check, repeat with the phase-field length scale b halved and doubled; a strong dependence on b would indicate that the source regularization, rather than the physical wetting mechanism, controls the result.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that electrolyte infiltration on newly opened fracture surfaces enhances (de)lithiation and drives fracture. In the phase-field implementation this is represented by Eq. (18): Q = 2(G/Gi)γ(d,∇d)J* for interface cracks and Q = 2γ(d,∇d)J* for bulk cracks. The crack-surface density γ = (2d−d^2)/(πb) + b|∇d|^2/π is nonzero for every d>0, so the wetting source is active not only on open crack faces but also on partially damaged material in the diffuse zone ahead of a propagating crack tip. The only numerical validation provided (SI Fig. 9) prescribes a fully developed interface crack (d=1); it does not test a propagating crack. Because the source modifies local concentration and stress before a crack is physically open, it can artificially enhance the wetting–fracture feedback: more wetting ahead of the tip, more damage, and therefore more wetted diffuse surface. If this is the case, the predicted wetting-only bulk cracks (Fig. 3f), higher capacity, and first-cycle Coulombic efficiencies (90.4% vs 65.9%; 99.4% vs 70.7%) are overestimates. This is a structural property of Eq. (18), not merely an uncertainty in the magnitude of J*.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a multiphysics phase-field model that couples electrolyte wetting at fracture surfaces with chemo-mechanical fracture in lithium-ion battery cathode materials, and combines it with scanning electron/transmission electron microscopy and scanning transmission X-ray microscopy experiments on single-crystal α-V2O5 lamellae and polycrystalline NCM particles. The model introduces a smeared source term (Eq. 18) that adds a chemical flux J* on damaged regions representing electrolyte infiltration on bulk and interface cracks. Simulations are compared with two single-crystal experiments, and the framework is then applied to polycrystalline NCM under galvanostatic cycling, where the wetting case yields higher first-cycle capacity and Coulombic efficiency than the non-wetting case. The central claim is that electrolyte infiltration at fracture surfaces enhances (de)lithiation and compositional heterogeneity, while the enhanced (de)lithiation in turn promotes further fracture, creating a mutually reinforcing wetting–fracture coupling.","tokens_in":26038,"tokens_out":5485,"duration_ms":62477,"significance":"If the claimed mechanism is correct, the paper challenges the conventional view that crack growth is purely detrimental in liquid-electrolyte cathodes and instead suggests a possible short-term capacity benefit from controlled fracture. The model is notable for concurrently treating intergranular and transgranular fracture and for deriving the wetting source term from a thermodynamically consistent dissipation inequality. The sharp-interface benchmark in the Supporting Information provides machine-checkable support for the equivalence of the smeared source term for a fully developed crack, and the STXM maps independently show lithiation heterogeneity near cracks. These strengths make the proposed mechanism plausible and the modeling framework potentially useful for future electrode-design studies, provided the load-bearing assumptions on the wetting flux are critically tested.","major_comments":[{"comment":"The wetting source term Q = 2(G/Gi)γ(d,∇d)J* for interface cracks and Q = 2γ(d,∇d)J* for bulk cracks is not restricted to open crack faces; because γ(d,∇d) = (2d−d²)/(πb) + b|∇d|²/π is nonzero for any 0<d<1, the source also activates in the partially damaged zone ahead of a propagating crack tip. The benchmark in SI Section 6 (Fig. 9) only compares the smeared source with a sharp-interface flux for a predefined, fully cracked interface (d=1); it does not test a propagating crack. Since the pre-opening flux increases the local chemical driving force for damage, the predicted wetting-only bulk cracks (Fig. 3f) and the capacity/Coulombic-efficiency improvements (90.4% vs 65.9%; 99.4% vs 70.7%) may be systematically overestimated. I request an additional benchmark for a propagating crack, a sensitivity study with a minimum damage threshold for Q, or an explicit physical justification for why electrolyte reactions should proceed on partially damaged material ahead of the crack tip.","section":"Section 5.1, Eq. (18); SI Section 6"},{"comment":"The abstract and conclusion describe the model as 'validated' and the experiment–simulation comparisons as 'excellent agreement', but the evidence is qualitative: it rests on two single-crystal samples (Sample-1 and Sample-2) with visual matching of SEM/STEM fracture patterns and STXM concentration maps, without quantitative image metrics, error bars, or a stated protocol for distinguishing wetting contributions from baseline chemo-mechanical response. Some inputs that directly shape the fracture predictions, such as the assumed 10° b-axis misorientation and the applied flux J* (and 2J* for Sample-2, Table 1), are prescribed without measurement or uncertainty analysis. The current evidence supports a plausible mechanism but not the strong 'validated' claim; please either provide a quantitative comparison metric (e.g., crack-path overlap, concentration-profile residuals) or soften the wording throughout.","section":"Section 2.2 and Section 2.3"},{"comment":"The model assumes that every newly formed fracture surface instantaneously supplies the same chemical flux J* as the external boundary, with no dependence on crack opening, capillary resistance, electrolyte transport within the crack, or reaction kinetics. The benchmark in SI Section 6 only verifies that the phase-field representation reproduces a prescribed sharp-interface flux for a fully cracked interface; it does not measure or constrain the physical magnitude of J* on real evolving cracks. Because the quantitative capacity and Coulombic-efficiency results in Figures 6 and 7 scale directly with the total active crack area times J*, those numbers are not robust without an independent constraint on the crack-surface flux. I recommend a sensitivity analysis over J* (including J*→0 at partially open cracks) and a statement that the current model represents an upper-bound wetting scenario.","section":"Section 5.1, Eq. (24) and Eq. (18)"}],"minor_comments":[{"comment":"The heading 'Results on ploycrystalline NCM cathode materials' contains a typo; it should be 'polycrystalline'.","section":"Section 2.3 heading"},{"comment":"The in-text reference 'Figure 6 (c) presents the voltage–capacity curves' conflicts with the caption, which labels the voltage–capacity panel as (b); please correct the panel numbering in the text or the caption.","section":"Figure 6 and Section 2.3.1"},{"comment":"In the provided Supporting Information, the caption block for the benchmark appears to contain a stray 'Figure 8' caption referring to STXM/XANES spectra; this formatting error should be removed or properly placed.","section":"SI Section 6, Figure 9"},{"comment":"The entry 'Partial molar volume Ωa = 2Ωb 3.497 × 10⁻⁶ [m³/mol]' is ambiguous; please list the numerical values for Ωa and Ωb separately.","section":"Table 1"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and the central mechanism is plausible, but the load-bearing validation of the wetting source term for propagating cracks and the quantitative basis for the 'validated' claim need to be strengthened. The requested benchmarks and sensitivity analyses are feasible with the existing code and would substantially improve the reliability of the conclusions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Name],\n\nThis paper has a genuinely new experimental observation and a sensible modeling framework, but the headline numbers rest on a wetting source term that is only validated for fully formed cracks, not for propagating ones. Treat the quantitative capacity gains as upper bounds for now.\n\nWhat is new and good: the STXM mapping of lithiation heterogeneity around cracks in single-crystal alpha-V2O5 is the kind of direct evidence the field has been missing. The phase-field treatment of concurrent interface and bulk wetting cracks is a useful extension of the authors' earlier unified model, and the thermodynamic derivation in the supplement is coherent. The sharp-interface benchmark in SI Fig. 9 is a legitimate check of the smeared source for a fully cracked interface. The qualitative match with the SEM/STEM images and with Ruess et al.'s liquid-vs-solid electrolyte experiments is real.\n\nThe soft spot is the one flagged in the stress-test, and it holds up on reading. Eq. (18) activates the flux for every d>0, so the diffuse damage zone ahead of a propagating crack tip is wetted before any open surface exists there. The SI benchmark only tests d=1. That means the wetting-fracture feedback loop in the main simulations is amplified by a mechanism that has not been validated. The predicted wetting-only bulk cracks and the Coulombic efficiency gains (90.4% vs 65.9%, 99.4% vs 70.7%) are likely overestimates. This is not a fatal flaw, but it is a structural modeling assumption that needs direct testing, e.g., by thresholding the source term at some d* or by comparing against a sharp-interface model for a moving crack.\n\nOther issues are proportionate: two samples, n=1 each, and 'excellent agreement' is visual rather than quantitative. Several inputs (b-axis misorientation, J*) are chosen without sensitivity analysis. No code or data are shared. None of these by themselves sink the paper, but they keep it from being more than conditionally convincing.\n\nThe paper is for battery chemo-mechanics modelers and anyone working on using fracture to improve kinetics. It deserves a serious referee, but I would send it back for major revision with the pre-crack wetting assumption and the quantitative comparison as the priorities.","headline":"Good experimental observations and a coherent model, but the wetting source term for propagating cracks is unvalidated, so the headline capacity gains are likely overestimates.","tokens_in":26610,"tokens_out":3385,"would_cite":false,"duration_ms":37482,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Electrolyte that wets freshly opened cracks turns them into reaction surfaces, raising first-cycle capacity and Coulombic efficiency in cathode materials.","keywords":["lithium-ion batteries","cathode fracture","electrolyte wetting","phase-field fracture","chemo-mechanical coupling","alpha-V2O5 single crystal","polycrystalline NCM","Coulombic efficiency"],"falsifier":"Compare first-cycle capacity and Coulombic efficiency in cathode particles where crack surfaces are wetted versus deliberately non-wetted (for example, a high-viscosity or non-wetting electrolyte, or a crack-sealing coating) under otherwise identical current and cut-off voltages; if the wetted case does not show the predicted higher capacity and efficiency, the central claim fails. A direct calculation that reduces the crack-surface flux $J^*$ to a small fraction of the external boundary flux would show whether bulk cracking and capacity gains persist.","tokens_in":25574,"feed_emoji":"🔋","tokens_out":7828,"duration_ms":75811,"temperature":0.7,"pith_summary":"The paper argues that when liquid electrolyte wets freshly formed cracks in cathode particles, those cracks become extra electrochemical reaction surfaces, creating a mutually reinforcing loop: more cracks mean more wetted area, which speeds (de)lithiation and alters local composition, and the resulting stresses drive further crack growth. The claim matters because it challenges the default assumption that cracking in battery electrodes is purely harmful, suggesting instead that controlled fracture can improve first-cycle charging capacity and Coulombic efficiency. Evidence comes from single-crystal $\\alpha$-V$_2$O$_5$ lamellae, where simulated fracture patterns and lithium concentration maps match electron and X-ray microscopy, and from simulations of polycrystalline NCM particles under constant-current cycling, where wetting raises first-cycle Coulombic efficiency from 65.9% to 90.4% in a two-grain benchmark and from 70.7% to 99.4% in a polycrystal.","feed_headline":"Wetted cracks lift first-cycle battery capacity","feed_subtitle":"Wetting turns cracks into reaction surfaces and lifts first-cycle Coulombic efficiency in simulations.","key_machinery":"The load-bearing object is a thermodynamically consistent phase-field fracture model that treats bulk and interface cracks in a unified way and adds a smeared chemical source at crack surfaces. The wetting flux is written as $Q = 2(G/G_i)\\gamma(d,\\nabla d)J^*$ in the diffuse interface region and $Q = 2\\gamma(d,\\nabla d)J^*$ in the bulk, where $\\gamma(d,\\nabla d)$ is the crack-geometry functional, $G/G_i$ is the local-to-interface fracture energy ratio, and $J^*$ is the same prescribed chemical flux used at the external boundary. This construction converts every regularized crack surface into an active electrochemical boundary, so fracture and lithium transport feed back on each other during (de)lithiation.","core_discovery":"The central claim is that electrolyte infiltration at fracture surfaces and chemo-mechanical fracture reinforce each other in cathode materials. Wetting of newly formed interface and bulk crack surfaces adds a chemical flux at the crack, enhancing (de)lithiation and producing compositional heterogeneity near cracks; in turn, the additional lithiation strains and tensile stresses accelerate crack propagation, introduce new fracture modes such as transgranular bulk cracks, and steer crack direction relative to the fast diffusion axis. The paper reports that this coupling is captured by a unified phase-field model and validated against single-crystal experiments, and that the same mechanism, applied to polycrystalline NCM under galvanostatic cycling, yields higher specific charging capacity and higher first-cycle Coulombic efficiency in the wetting case (90.4% vs 65.9% in the two-grain benchmark, 99.4% vs 70.7% in the polycrystal). If correct, first-cycle cracks in liquid-electrolyte cathodes are not purely detrimental.","pith_inferences":["If the real wetting flux is limited by crack aperture, capillary resistance, or electrolyte transport, the predicted capacity gains would shrink; a testable extension is to make $Q$ depend on crack opening or on local electrolyte concentration.","The model implies a design window: a pore or crack network that is wetted but does not disconnect electronically could give first-cycle kinetic benefits without the long-term degradation usually blamed on cracks.","The predicted perpendicular-to-diffusion directionality of bulk cracks could be checked experimentally by varying electrolyte surface tension or wetting additives in oriented single-crystal lamellae and measuring crack orientation.","In solid-electrolyte cells, where cracks are not wetted, equivalent kinetic gains might be engineered by pre-patterning internal ion-conducting channels rather than by relying on fracture."],"forward_implications":["Wetting turns crack surfaces into additional reaction sites, increasing the electrochemically active area beyond the external particle boundary.","Under constant total current, the extra active area lowers local current density and lithium flux, reducing local charging rate and raising accessible capacity at a given cutoff voltage.","Wetting-induced bulk cracks appear when there is no wetting; the fracture mode shifts from interface-only to interface-plus-transgranular, and bulk cracks grow preferentially perpendicular to the fast diffusion axis.","First-cycle Coulombic efficiency increases from 65.9% to 90.4% (two-grain benchmark) and from 70.7% to 99.4% (polycrystal) when wetting is included.","Wetting accelerates crack propagation and increases crack density, so the same mechanism that improves first-cycle response also creates more damage that must be managed for long-term cycling."],"supporting_citations":[{"why":"Quantifies particle cracking, active surface area, and lithium diffusion in polycrystalline NCM, providing the baseline for the polycrystal simulations.","marker":"[12]"},{"why":"Provides the experimental comparison of NCM kinetics with liquid versus solid electrolyte, the key qualitative benchmark for the wetting/no-wetting capacity and cracking results.","marker":"[21]"},{"why":"Supplies the length-scale-insensitive cohesive phase-field interface formulation that lets bulk and interface cracks evolve together.","marker":"[32]"},{"why":"Introduces the phase-field treatment of electrochemical reactions at exterior and interior interfaces, the origin of the wetting source term.","marker":"[33]"},{"why":"Gives the phase-field electro-chemo-mechanical fracture framework with crack-contained electrolytes used for the bulk wetting source.","marker":"[35]"},{"why":"Contributes the phase-field formulation of chemo-mechanical fracture in battery particles that the bulk source term builds on.","marker":"[36]"},{"why":"Supplies the unified phase-field theory for damage and quasi-brittle failure underlying the cohesive fracture model.","marker":"[38]"},{"why":"Demonstrates inter- and trans-granular fracture simulations in polycrystalline NMC, the basis for extending the model to NCM particles.","marker":"[39]"}],"fun_headline_variants":["Wetting turns cracks into capacity boosters","Cracks, when wetted, lift battery capacity","Wet fractures enhance lithium-ion flow","Fractures plus wetting raise cathode efficiency","Crack-wetting synergy betters battery performance"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Every newly opened fracture surface is assumed to supply the same chemical flux $J^*$ as the outer particle boundary, with electrolyte arriving instantly and reacting without resistance from crack width, capillary forces, or transport inside the crack.","fun_headline_variants_meta":{"raw":{"variants":["Wetting turns cracks into capacity boosters","Cracks, when wetted, lift battery capacity","Wet fractures enhance lithium-ion flow","Fractures plus wetting raise cathode efficiency","Crack-wetting synergy betters battery performance"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000229,"raw_usage":{"total_tokens":1503,"prompt_tokens":992,"completion_tokens":511,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":608,"completion_tokens_details":{"reasoning_tokens":442}},"tokens_in":608,"tokens_out":511,"duration_ms":5663,"temperature":1.0,"reasoning_tokens":442,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T19:45:18.678148+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compare first-cycle capacity and Coulombic efficiency in cathode particles where crack surfaces are wetted versus deliberately non-wetted (for example, a high-viscosity or non-wetting electrolyte, or a crack-sealing coating) under otherwise identical current and cut-off voltages; if the wetted case does not show the predicted higher capacity and efficiency, the central claim fails. A direct calculation that reduces the crack-surface flux $J^*$ to a small fraction of the external boundary flux would show whether bulk cracking and capacity gains persist.","supporting_citations":[{"cited_title":"Trevisanello, R","cited_arxiv_id":null,"evidence_quote":"Quantifies particle cracking, active surface area, and lithium diffusion in polycrystalline NCM, providing the baseline for the polycrystal simulations."},{"cited_title":"Ruess, S","cited_arxiv_id":null,"evidence_quote":"Provides the experimental comparison of NCM kinetics with liquid versus solid electrolyte, the key qualitative benchmark for the wetting/no-wetting capacity and cracking results."},{"cited_title":"Chen, X.-L","cited_arxiv_id":null,"evidence_quote":"Supplies the length-scale-insensitive cohesive phase-field interface formulation that lets bulk and interface cracks evolve together."},{"cited_title":"Zhao, B.-X","cited_arxiv_id":null,"evidence_quote":"Introduces the phase-field treatment of electrochemical reactions at exterior and interior interfaces, the origin of the wetting source term."},{"cited_title":"Hageman and E","cited_arxiv_id":null,"evidence_quote":"Gives the phase-field electro-chemo-mechanical fracture framework with crack-contained electrolytes used for the bulk wetting source."},{"cited_title":"Miehe, H","cited_arxiv_id":null,"evidence_quote":"Contributes the phase-field formulation of chemo-mechanical fracture in battery particles that the bulk source term builds on."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the unified phase-field theory for damage and quasi-brittle failure underlying the cohesive fracture model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates inter- and trans-granular fracture simulations in polycrystalline NMC, the basis for extending the model to NCM particles."}],"review_version":1}