{"id":"b548a754-392d-4c03-91e3-bf9a01f9b332","arxiv_id":"1908.04493","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Real-time element-resolved dissolution of the compositionally complex alloy Al1.5TiVCr shows strongly incongruent dissolution: preferential aluminium release, minimal titanium release until breakdown.","lead":"Using a flow cell connected to an ICP mass spectrometer, researchers watched a four-metal alloy dissolve in real time and found its ingredients leave at very different rates. Aluminium dissolves fastest, titanium barely at all until breakdown, and X-ray analysis shows a protective film rich in aluminium oxide with some unoxidized metal.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The fixed-valence Faraday conversion is the load-bearing assumption: the paper's own XPS data indicate higher V and Cr valences after polarisation, so the quantitative partial currents and the 'ground truth' wording are less secure than presented.","rationale":"The reader's weakest assumption is exactly the load-bearing point I would raise. The paper's central quantitative claims are the partial current densities and the 1–2 order-of-magnitude separation between Al and the other elements, all derived by assuming fixed oxidation states for the dissolving species. The manuscript itself acknowledges in the General discussion that the oxidation states cannot be determined by ICP-MS and that V and Cr in particular may change valence under polarisation. This is not an external or exotic objection; it is an internally documented limitation of the conversion step. The XPS data provide qualitative support for higher V and Cr valences in the surface film, though they do not directly measure dissolved ions. A sensitivity analysis with alternative valences is the natural test: because Al is assigned +3 and Ti +4 in all scenarios, the Al-dominance and Ti-latency conclusions are expected to survive, but the exact ratios and the residual current used to infer film formation could shift. This warrants the same CONDITIONAL verdict already given, not a harsher one, because the qualitative incongruent-dissolution picture and the XPS evidence for an Al-rich film are independent and mutually reinforcing. I therefore leave the verdict unchanged while agreeing with the reader's diagnosis.","tokens_in":9339,"tokens_out":7086,"duration_ms":79171,"concrete_test":"Recompute Figure 2c from the raw ICP-MS transients using alternative formal charges for the polarisation segment: (1) V as V4+ and Cr as Cr6+; (2) V as V5+ and Cr as Cr6+, keeping Al3+ and Ti4+ fixed. For each case, report the iAl/iV, iAl/iCr and iAl/iTi ratios in the pre-breakdown region and at +1.0 VSCE, and the sign and magnitude of ipstat minus itot. If Al remains more than one order of magnitude above V and Cr in all cases and the residual remains positive, the central qualitative claims survive and only the 'ground truth' wording needs revision; if a ratio crosses the one-order threshold, the quantitative ranking in the Conclusions is not robust.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The quantitative core of the paper is the set of partial current densities iAl3+, iTi4+, iV3+, iCr3+ in Fig. 2. These are obtained from ICP-MS element mass fluxes via Faraday's law using fixed charges (Methods, AESEC). The same alloy's XPS data (General discussion; Fig. 3) show the fraction of V4+/V5+ in the surface film rising from ~30% to ~60% and an increased Cr5+ fraction after anodic polarisation. That does not prove the dissolved ions carry these charges, but it removes the assumption that the most stable +3 state is the only relevant one. If dissolved V is partly V4+/V5+ and dissolved Cr partly Cr6+, then iV3+ and iCr3+ are systematically underestimated by up to 67% and 100%, while Al and Ti are unaffected. The reported 'ground truth dissolution rates' in Results and the quantitative claim that Al dissolution is 1–2 orders of magnitude above V, Cr and Ti therefore rest on a valence model the manuscript itself flags as uncertain in the General discussion. The direction of the error is known and bounded, so the qualitative incongruent-dissolution picture is likely robust; but the precise partial currents, and the use of itot vs ipstat to infer film formation, are not secure without a sensitivity analysis.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports the first application of atomic emission spectroelectrochemistry (AESEC, using inline ICP-MS) to a compositionally complex alloy, Al1.5TiVCr, in 0.1 M NaCl. Element-resolved dissolution rates (iAl3+, iTi4+, iV3+, iCr3+) are measured during open-circuit exposure and potentiodynamic polarization, and the results are correlated with XPS depth profiles of the native and post-polarization surface films. The paper claims that dissolution is incongruent, with Al dissolving 1–2 orders of magnitude faster than V, Cr, and Ti; that Ti does not dissolve appreciably until breakdown near +0.7 VSCE; that the sum of partial currents itot remains below the applied current ipstat, indicating surface film formation; and that XPS shows a surface film dominated by Al oxide with unoxidised metal present for all four elements. The authors conclude that this demonstrates, for the first time, element-resolved real-time dissolution behavior for a CCA.","tokens_in":9564,"tokens_out":3743,"duration_ms":36767,"significance":"If the results hold, this is a valuable methodological contribution: it extends AESEC to CCAs and provides direct, element-resolved evidence of incongruent dissolution in a single-phase multi-principal-element alloy, with supporting surface chemistry from XPS. The qualitative picture—preferential Al dissolution, suppressed Ti dissolution, and a film enriched in Al oxide with metallic species—is consistent with prior indirect work and is likely to be of interest to the corrosion and high-entropy alloy communities. However, the quantitative partial currents and the associated claims of precise dissolution rates depend on assumed oxidation states for dissolved ions, which the manuscript itself flags as uncertain; this limits the strength of the quantitative conclusions as currently presented.","major_comments":[{"comment":"The conversion of ICP-MS elemental mass fluxes to partial current densities (iAl3+, iTi4+, iV3+, iCr3+) via Faraday's law assumes fixed oxidation states (Al3+, Ti4+, V3+, Cr3+). The manuscript's own XPS data (General discussion; Fig. 3d) show that the fraction of V4+/V5+ in the surface film increases from ~30% to ~60% and that Cr5+ also increases after anodic polarization. If dissolved V and Cr carry higher charges (e.g., V5+ or Cr6+), then iV3+ and iCr3+ are systematically underestimated—by up to ~67% for V and ~100% for Cr. This directly affects the quantitative claims in the Conclusions that Al3+ dissolution is 1–2 orders of magnitude greater than V, Cr, and Ti, and the 'ground truth dissolution rates' wording in the Results. Please provide a sensitivity analysis for the assumed valences or clearly frame the partial currents as lower-bound estimates, and temper the 'ground truth' language.","section":"Methods (AESEC); Results; General discussion"},{"comment":"The inference that itot < ipstat at all times indicates the development of a surface film rests on the sum of partial currents computed under the same fixed-valence assumption. If V and Cr dissolve with higher valences, itot would increase and the margin below ipstat would shrink, weakening the quantitative support for the film-formation inference. In addition, ipstat includes non-faradaic contributions (capacitive charging, film growth), so the comparison should be discussed with this caveat. The qualitative conclusion that a film is present may still hold, but the quantitative statement as written is not robust without addressing these effects.","section":"Results (Fig. 2b/c)"},{"comment":"The claim that Ti essentially does not dissolve until breakdown near +0.7 VSCE is based on Ti concentrations that may be close to the ICP-MS detection limit (0.5 µg/L for Ti, as stated in Methods). Without presenting the raw concentration profiles or signal-to-noise information, it is not clear whether the reported Ti dissolution is truly negligible or merely below the instrument's detection capability. Please include detection-limit-relative data or error estimates to support the 'essentially no dissolution' statement.","section":"Methods (AESEC); Results (Fig. 2)"}],"minor_comments":[{"comment":"The method is called 'atomic emission spectroelectrochemistry (AESEC),' but the experimental setup uses ICP-MS, which measures mass spectra rather than atomic emission. This terminology is historically associated with ICP-OES; please clarify the relationship or use a more accurate descriptor such as 'inline ICP-MS spectroelectrochemistry.'","section":"Title/Abstract/Methods"},{"comment":"The 'repeat' test after 60 min of etching compares a native surface briefly exposed to air with an anodically polarized surface kept under vacuum for three days; these two conditions differ in both prior electrochemical history and storage environment, so the comparison of re-oxidation behavior should be qualified accordingly.","section":"Fig. 3 caption and text"},{"comment":"The sentence 'It is noteworthy that Ti4+ is the fastest metal cation that reaches to a lower and stable ion dissolution current' contains a grammatical error ('reaches to') and is ambiguous in meaning; please rephrase for clarity.","section":"Results, paragraph 3"},{"comment":"The term 'in-congruent' is used with a hyphen; the standard spelling is 'incongruent' (also in the Conclusions).","section":"Throughout"},{"comment":"The phrase 'the precise quantification of ion dissolution rates indicates the corrosion rate' is imprecise because only anodic dissolution is measured; suggest using 'anodic dissolution rate' to avoid implying that cathodic or other contributions are included.","section":"General discussion"},{"comment":"The introduction cites the 2004 Yeh paper for the HEA concept; for readers, a more recent review of HEAs and CCAs (beyond the authors' own references) would improve context.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript's central qualitative claim—incongruent dissolution in a CCA—is well supported, but the quantitative partial currents are built on a fixed-valence assumption that the paper's own XPS data call into question. The authors acknowledge this in the General discussion but do not reflect it in the Results or Conclusions, where the numbers are presented as definitive. The requested sensitivity analysis is straightforward and should be feasible within the manuscript's scope. The paper also relies heavily on the authors' prior work on the same alloy (refs 4, 11, 21), which is appropriate method continuity rather than a novelty concern; however, the fit with the journal's applied-physics scope is somewhat indirect, as the work is primarily corrosion science."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a look. This is the first time AESEC has been applied to a compositionally complex alloy, and it delivers what prior ex situ work could only infer: real-time, element-resolved dissolution rates. The qualitative picture is well supported. During anodic polarisation in 0.1 M NaCl, Al dissolves far more than V, Cr, or Ti, and Ti basically sits still until breakdown near +0.7 V_SCE. The XPS correlation, showing a surface film dominated by Al oxide with unoxidised metal from all four elements, is a nice complement and consistent with the dissolution ranking. That is a solid new application, and the paper is honest about being a first demonstration.\n\nThe soft spot is the one the stress-test note flags, and the authors actually flag it too. The partial current densities i_Al3+, i_Ti4+, i_V3+, i_Cr3+ are computed from ICP-MS mass fluxes using fixed charges. But the paper's own XPS data show the V4+/V5+ fraction rising from about 30% to 60% and an increase in Cr5+ after polarisation. If dissolved V and Cr carry higher charges, the computed partial currents are systematically low by up to 67% and 100%. The direction of the error is known and bounded, so the qualitative claim of incongruent dissolution, with Al dominant and Ti inert, survives. But the precise numbers, and the use of i_tot < i_pstat to infer film formation, are less secure than the 'ground truth dissolution rates' wording in the Results implies. There are also no replicate error bars shown, which matters when comparing partial currents across elements.\n\nNone of this is fatal. The paper deserves peer review, and a good referee should push for a sensitivity analysis of the valence assumption, or at least a softening of the quantitative claims, plus replicate statistics. The self-citations are method continuity, not padding. If I worked on CCA corrosion, I would cite this. I'd bring it to a reading group only if the group cares about operando corrosion methods; otherwise it is a worthwhile but narrow contribution.","headline":"First real-time element-resolved dissolution data for a CCA; the qualitative incongruent-dissolution story holds, but the quantitative partial currents rest on a valence assumption the paper's own XPS undercuts.","tokens_in":10170,"tokens_out":1413,"would_cite":true,"duration_ms":15552,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A single-phase Al1.5TiVCr alloy corrodes incongruently: aluminum dissolves one to two orders of magnitude faster than vanadium, chromium, and especially titanium, which stays put until breakdown.","keywords":["high entropy alloy","compositionally complex alloy","corrosion","atomic emission spectroelectrochemistry","inductively coupled plasma mass spectrometry","X-ray photoelectron spectroscopy","incongruent dissolution","oxide film"],"falsifier":"Run the same AESEC polarization while measuring dissolved vanadium and chromium speciation (e.g., by ion chromatography coupled to ICP-MS, or by UV-vis on collected fractions). If dissolved V is predominantly V5+ and Cr is Cr6+, then the reported iV3+ and iCr3+ values would need to be multiplied by roughly 5/3 and 6/3, respectively, which could shift the relative dissolution rates and the inference of which element dominates.","tokens_in":9108,"feed_emoji":"🧪","tokens_out":6432,"duration_ms":55111,"temperature":0.7,"pith_summary":"The paper reports the first element-resolved, real-time measurement of dissolution for a compositionally complex alloy. Using atomic emission spectroelectrochemistry—an electrochemical flow cell feeding an ICP-MS—the authors show that Al1.5TiVCr in 0.1 M NaCl dissolves incongruently: aluminum leaves far faster than vanadium or chromium, and titanium essentially does not dissolve until the alloy breaks down near +0.7 V vs SCE. Throughout polarization the sum of element dissolution currents stays below the applied potentiostat current, evidence that a surface film is forming. XPS of the film shows it is dominated by aluminum oxide and also contains unoxidized metal of all four elements, supporting the picture that corrosion resistance comes from rapid oxide formation rather than inertness. If correct, the work makes a previously assumed behavior—selective oxidation in CCAs—directly observable and quantifiable.","feed_headline":"In a complex alloy, aluminum dissolves first; titanium barely moves","feed_subtitle":"Live element-by-element ICP tracking reveals incongruent corrosion and a surface film that explains the alloy's resistance.","key_machinery":"The central object is atomic emission spectroelectrochemistry (AESEC), a scanning electrochemical flow cell whose outlet is coupled to an ICP-MS so that dissolved Al, Ti, V, and Cr are quantified continuously during open-circuit exposure and potentiodynamic polarization. Measured ion concentrations are converted to instantaneous mass dissolution rates via v_M = C_M f / A, and then to partial elemental current densities i_Mx+ via Faraday's law using fixed oxidation states (Al3+, Ti4+, V3+, Cr3+). The sum itot is compared with the potentiostat current ipstat to infer surface film formation. X-ray photoelectron spectroscopy depth profiling, using argon cluster sputtering, supplies the companion surface film composition and metal/oxide balance. The machinery's work is to make element-specific dissolution rates visible in real time, which no conventional electrochemical test can do.","core_discovery":"The central claim is that dissolution of the single-phase CCA Al1.5TiVCr is incongruent in real time and in situ. In quiescent 0.1 M NaCl, the partial ion dissolution currents follow iAl3+ >> iV3+ >> iCr3+ > iTi4+ at open circuit; during potentiodynamic polarization, Al3+ remains the dominant dissolved species, Cr and V show distinct increases at around +0.3 and +0.45 V, and Ti4+ only rises sharply at ~+0.7 V SCE, the alloy breakdown potential. Since itot < ipstat at all times, not all applied charge goes into metal dissolution; a surface film forms. XPS reveals that film is principally Al oxide, with the less reactive elements partly trapped as unoxidized metal, and that the outer film after polarization is more fully oxidized. The paper presents this as the first AESEC study of a CCA and as direct validation that CCA corrosion involves incongruent dissolution.","pith_inferences":["If oxidation states shift as XPS suggests (V to V4+/V5+, Cr to higher valence) during polarization, the true partial currents for V and Cr would be larger than reported, potentially closing some of the gap with aluminum; the element ranking might change if this correction were applied.","The same AESEC+XPS protocol could be applied to other single-phase CCAs to test whether the 'aluminum-first, titanium-last' pattern is general, or whether it is controlled by the relative Gibbs energies of oxide formation of the constituents.","The observation that the outer surface film after polarization is almost fully oxidized, while the native film contains substantial metal, suggests a testable prediction: pre-polarized surfaces should show different early-stage dissolution kinetics than as-polished surfaces, because the oxide/metal balance differs."],"forward_implications":["The element-resolved dissolution rates provide a quantitative basis for predicting selective leaching from CCAs: aluminum is the preferential anodic species, so Al-rich CCAs may lose load-bearing Al while retaining Ti, V, and Cr in a surface film.","Because itot < ipstat throughout, the passivity of Al1.5TiVCr is consistent with film formation consuming most of the anodic charge; equivalent corrosion models that assume all charge goes to dissolution would overestimate the corrosion rate for this alloy.","The near-zero Ti dissolution until ~+0.7 V SCE identifies titanium as a key stabilizer of the film: if Ti's repassivation contributes to suppressing V, Cr, and Al release, alloy designs that raise Ti content should improve resistance.","The XPS finding of unoxidized metal inside the surface film means that the film is not a simple passive oxide; any model of CCA passivity must account for a mixed oxide-metal layer whose protective character is not yet mechanistically explained."],"supporting_citations":[{"why":"Introduces atomic emission spectroelectrochemistry, the inline ICP method that provides element-resolved dissolution rates.","marker":"[14]"},{"why":"Demonstrates operando multi-element passivation analysis by ICP-MS, the direct precedent for applying this approach to a multi-principal element alloy.","marker":"[20]"},{"why":"Prior XPS surface analysis of the native oxide on the same alloy family; supplies the spectral analysis method and the prior observation of unoxidized metal in the film.","marker":"[21]"},{"why":"Earlier corrosion study of AlTiVCr that posited incongruent dissolution; the present paper's quantitative result directly tests that proposal.","marker":"[11]"},{"why":"Reports the lightweight single-phase AlTiVCr compositionally complex alloy and its B2 structure, the material studied here.","marker":"[4]"},{"why":"Review of high entropy alloy corrosion that frames the expectation of high corrosion resistance and the need for mechanistic dissolution data.","marker":"[3]"},{"why":"Source of the flow-cell equation used to convert measured ion concentration into instantaneous mass dissolution rate.","marker":"[31]"}],"fun_headline_variants":["In complex alloy, aluminum dissolves first, titanium holds on","Dissolving a complex alloy: Al leaves, Ti stays until breakdown","Real-time view: incongruent dissolution in a 5-metal alloy","Aluminum exits first; titanium clings to complex alloy surface","Complex alloy corrosion: Al preferential, Ti resistant until the end"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The reported dissolution currents are computed from ICP-MS mass fluxes using fixed oxidation states (Al3+, Ti4+, V3+, Cr3+), yet the paper's own XPS data show vanadium and chromium reaching higher valence states after polarization, so if those higher charges are the ones dissolving, the calculated currents and the element ranking would change.","fun_headline_variants_meta":{"raw":{"variants":["In complex alloy, aluminum dissolves first, titanium holds on","Dissolving a complex alloy: Al leaves, Ti stays until breakdown","Real-time view: incongruent dissolution in a 5-metal alloy","Aluminum exits first; titanium clings to complex alloy surface","Complex alloy corrosion: Al preferential, Ti resistant until the end"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000662,"raw_usage":{"total_tokens":3011,"prompt_tokens":914,"completion_tokens":2097,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":530,"completion_tokens_details":{"reasoning_tokens":2010}},"tokens_in":530,"tokens_out":2097,"duration_ms":13266,"temperature":1.0,"reasoning_tokens":2010,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:40:47.336366+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same AESEC polarization while measuring dissolved vanadium and chromium speciation (e.g., by ion chromatography coupled to ICP-MS, or by UV-vis on collected fractions). If dissolved V is predominantly V5+ and Cr is Cr6+, then the reported iV3+ and iCr3+ values would need to be multiplied by roughly 5/3 and 6/3, respectively, which could shift the relative dissolution rates and the inference of which element dominates.","supporting_citations":[],"review_version":1}