{"id":"675217db-718e-4741-b10b-62e22fec22c0","arxiv_id":"2509.16361","paper_version":5,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Spin-resolved ARPES of 2.7 nm epitaxial RuO2 reveals coexisting mirror-even and mirror-odd momentum-dependent spin polarization, consistent with an emergent in-plane magnetic order (m'm2') stabilized by epitaxial strain.","lead":"Ultra-thin, fully strained RuO2 films show a mix of mirror-odd and mirror-even spin polarization in angle-resolved photoemission, unlike bulk RuO2. The patterns point to a strain-stabilized, nonrelativistic spin structure, possibly altermagnetic, with potential uses in oxide spintronics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The magnetic conclusion rests on the mirror-even [001] photoelectron polarization being an initial-state property; final-state multiple scattering is an untested nonmagnetic alternative that the paper itself flags as a condition.","rationale":"The paper has real strengths: the mirror-even component is reproduced at two photon energies, the structural characterization is careful, the SHG confirms mm2 symmetry, and the DFT surface-state assignment is plausible. The reader's CONDITIONAL verdict is appropriate. The load-bearing concern is not that the authors are wrong, but that the central claim's conclusion depends on an untested condition that the authors themselves flag. The one-step photoemission test would settle whether the mirror-even [001] polarization can be generated without intrinsic magnetism. A secondary fragility is the constant-k110 assumption used to reinterpret k001 sigma110 as k110 k001 sigma110, but that would only affect the specific B1- assignment, not the more fundamental question of whether the observed polarization is intrinsic at all. I would keep the reader's CONDITIONAL verdict unchanged.","tokens_in":18538,"tokens_out":4458,"duration_ms":43519,"concrete_test":"Run a one-step spin-resolved photoemission calculation for a nonmagnetic, spin-unpolarized but spin-orbit-coupled strained RuO2(110) slab using the experimental geometry (55 eV and 62 eV p-polarized light, VLEED detector axis along [001]) at the same k-points used in Figs. 4 and 5. If the computed P[001] reproduces the observed mirror-even sign pattern with magnitude comparable to the data, the multiple-scattering explanation holds and the m'm2' conclusion is not justified; if the computed P[001] is negligible, the final-state objection is retired and the initial-state assignment is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The core inference is that the (1-10)- and (001)-mirror-even photoelectron spin polarization along [001] (Figs. 4K-L, 5I-J, fig. S8I-J) reflects the intrinsic initial-state spin texture of the 2.7 nm RuO2 film. The symmetry analysis in the Discussion and Table 1 only classifies allowed spin-splitting terms of the initial-state Hamiltonian under mm2.1', so it rules out nonmagnetic spin-splitting terms in the initial state; it does not constrain spin-dependent final-state multiple scattering, which is known to produce momentum-dependent photoelectron spin polarization even for nonmagnetic surfaces (refs. 66-72). The manuscript itself states the magnetic conclusion conditionally ('if the observed mirror-even spin polarization arises from intrinsic magnetism regardless of photoemission multiple scattering') and acknowledges 'potential complications on the photoelectron spin polarization from multiple scattering.' No one-step photoemission calculation or nonmagnetic control measurement is provided. Because the same VLEED analyzer and photon energies were used for all spin-resolved data, a multiple-scattering origin could mimic exactly the observed mirror-even pattern, especially since the effect appears near EF where the alpha-NBs dominate. The abstract's claim that a 'comprehensive symmetry analysis rules out nonmagnetic origins' is therefore stronger than the evidence supports unless the final-state channel is independently excluded.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports spin-resolved ARPES measurements on 2.7 nm fully strained epitaxial RuO2 films grown on TiO2/Nb:TiO2(110) by hybrid molecular beam epitaxy. The authors observe narrow bands near the Fermi level and a photoelectron spin polarization that combines a mirror-odd component (consistent with inversion-symmetry breaking) and a mirror-even in-plane [001] component that is not expected for the paramagnetic state. Using a group-theory classification of spin-splitting terms for the paramagnetic point group mm2.1', they show that the observed mirror-even terms can be accommodated by a magnetic order parameter transforming as B1-, which would correspond to the magnetic point group m'm2' with in-plane moments, compatible with either ferromagnetism or d-wave altermagnetism. The magnetic conclusion is explicitly stated to hold only if the mirror-even photoelectron spin polarization is an intrinsic initial-state property and not a final-state multiple-scattering artifact. The paper is an experimental first look at the spin structure in the ultra-thin, fully strained regime of RuO2.","tokens_in":18931,"tokens_out":3077,"duration_ms":29671,"significance":"If the central claim holds, the paper provides direct spectroscopic evidence for a nonrelativistic spin texture stabilized by epitaxial strain in ultra-thin RuO2, a material whose bulk and thick-film forms have increasingly been argued to be nonmagnetic. This would be an important advance for the altermagnetism debate and for oxide heterostructure spintronics. The authors should be credited for the careful sample characterization (XRD, XRR, RHEED, AFM, SHG), the two-photon-energy internal consistency checks of the spin polarization, the propagated Poisson error bars, and the transparent group-theory classification in Table 1, which does not fit any parameter and is a genuine symmetry analysis. The main weakness is that the load-bearing premise—that the mirror-even polarization reflects the intrinsic initial-state spin texture rather than spin-dependent final-state multiple scattering—is explicitly conditional and is not independently tested. This limits the strength of the conclusions as currently worded.","major_comments":[{"comment":"The abstract claims that 'a comprehensive symmetry analysis rules out nonmagnetic origins of this spin texture,' but the symmetry analysis in the Discussion and Table 1 only classifies spin-splitting terms of the initial-state Hamiltonian. It does not constrain spin-dependent final-state multiple scattering, which is known to produce momentum-dependent photoelectron spin polarization even for nonmagnetic surfaces (refs. 66-72). The manuscript itself acknowledges this condition, stating that the magnetic conclusion holds 'if the observed mirror-even spin polarization arises from intrinsic magnetism regardless of photoemission multiple scattering,' and notes 'potential complications on the photoelectron spin polarization from multiple scattering.' No one-step photoemission calculation or nonmagnetic control measurement is provided. The abstract's unconditional wording therefore overstates the evidence. This is a load-bearing issue because the mirror-even [001] polarization is the sole basis for the time-reversal-symmetry-broken conclusion.","section":"Abstract and Discussion"},{"comment":"The assignment of the observed k001*sigma110 term to the B1- irrep rests on the assumption that the out-of-plane momentum k110 can be treated as a constant, allowing the term to be reinterpreted as the quadratic k110*k001*sigma110 term. The manuscript states this assumption but does not provide quantitative justification, such as an estimate of the kz broadening from the 2.7 nm film thickness or a photon-energy dependence test. If kz conservation is not fully suppressed, the term would transform as B1+ under mm2.1', which the authors themselves note would require a different magnetic point group (m.1') incompatible with the structural mm2 symmetry. Because this assumption directly determines the proposed m'm2' magnetic point group, it needs either additional experimental support or an explicit sensitivity analysis showing that the qualitative conclusion is robust to partial kz coherence.","section":"Discussion, paragraph 4"},{"comment":"The text states 'Judging from the calculated spin polarization presented in Fig. 4 (K and L)' in reference to data-derived spin polarization curves. The curves in Fig. 4(K,L) are converted from measured spin-resolved EDCs, not from a calculation. This appears to be a typo (likely 'measured' instead of 'calculated'), but because the figure is central to the mirror-even claim, the wording should be corrected to avoid ambiguity about whether any theoretical spin-resolved simulation is being shown.","section":"Fig. 4, panels K-L and Results section"}],"minor_comments":[{"comment":"The abstract says '2-nanometer-thick' while the main text and Methods consistently state 2.7 nm; this should be harmonized.","section":"Abstract"},{"comment":"The Sherman function is quoted as S = 0.2 without an uncertainty. Since the absolute polarization values are not central to the symmetry classification, this is acceptable, but adding a nominal systematic uncertainty would improve the error discussion.","section":"Methods, Eq. (1)"},{"comment":"The spin-resolved EDCs are normalized using counts in kinetic-energy windows that are assumed to be background-dominated. It would be helpful to state explicitly whether those windows were checked to have negligible spin polarization, since a spin-polarized background would bias the normalized asymmetry.","section":"Methods, normalization paragraph"},{"comment":"The arrows marking the alpha, gamma, and delta bands are described in the text and figure caption, but the figure panel itself is busy; adding the labels directly on the panel would improve readability.","section":"Fig. 4D"},{"comment":"The phrase 'could be associated with time-reversal-symmetry breaking' is appropriately cautious, but the earlier sentence in the same paragraph ('is beyond intrinsically nonmagnetic origins') is stronger; aligning these two statements would help the reader track the level of certainty.","section":"Discussion, last paragraph"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a highly contested topic (magnetism in RuO2), where several recent high-profile experiments have concluded that bulk and thick-film RuO2 are nonmagnetic. The authors are appropriately careful in the Discussion by making the magnetic conclusion conditional on excluding final-state multiple scattering, but the abstract does not carry that condition. Given the field's history, I would urge the editor to require either a quantitative final-state check (e.g., a one-step photoemission simulation or a nonmagnetic control) or a substantial rewording of the central claims in the abstract and conclusions. The k110-constant assumption is an additional point that needs explicit defense; as it stands, the m'm2' assignment is one of several possible interpretations. The group-theory table and the spectroscopy itself are solid contributions that warrant publication after these issues are addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear Colleague,\n\nHere's my take. The genuinely new thing in this paper is the spin-resolved ARPES on 2.7 nm fully strained RuO2: nobody has measured spin texture in that regime, and the coexistence of mirror-odd and mirror-even momentum-dependent photoelectron spin polarization is striking. The sample quality looks excellent (hMBE, XRR, XRD, RHEED, AFM, SHG all confirm the intended structure), and the data are presented honestly with raw counts and propagated Poisson errors. The symmetry analysis in Table 1 is a real classification of spin-splitting terms under the paramagnetic group mm2.1', and the deduction of the m'm2' magnetic point group from the observed terms is a legitimate exercise. I also appreciate that the paper is transparent about the controversy around RuO2 and doesn't overclaim individual band assignments beyond what the data show.\n\nThe soft spot is the one the authors themselves flag but do not close. The mirror-even [001] polarization near EF is the load-bearing observation, and the magnetic interpretation depends on it being an initial-state property. The text says 'if the observed mirror-even spin polarization arises from intrinsic magnetism regardless of photoemission multiple scattering,' and refs 66–72 show this is a known artifact for nonmagnetic surfaces. The symmetry analysis rules out nonmagnetic spin-splitting terms in the initial-state Hamiltonian, but it cannot constrain final-state multiple scattering. Without a one-step photoemission calculation or a control measurement on a nonmagnetic reference, the abstract's claim that 'a comprehensive symmetry analysis rules out nonmagnetic origins' is too strong. Also, interpreting the k001σ110 term as a quadratic k110k001σ110 term relies on taking k110 as a constant; that is plausible for a 2.7 nm film, but it is an extra assumption the authors acknowledge. The m'm2' assignment is therefore conditional, not proven.\n\nWho should read it: anyone working on altermagnetism and spin-ARPES, and anyone trying to reconcile the RuO2 debate. The paper deserves a serious referee, but the referee should push for a multiple-scattering-aware calculation or a complementary probe on the same films, and for a revision that makes the abstract match the conditional tone of the discussion.\n\nI'd bring it to reading group. It would generate a useful discussion about evidence standards in altermagnetism.","headline":"Genuinely new spin-ARPES data on ultra-thin strained RuO2 with a solid symmetry analysis, but the abstract overstates the conclusion by glossing over the unaddressed multiple-scattering alternative.","tokens_in":19477,"tokens_out":2886,"would_cite":true,"duration_ms":24723,"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":"Ultra-thin epitaxially strained RuO2 shows a momentum-dependent spin texture whose mirror-even component implies time-reversal symmetry breaking and the m′m2′ magnetic point group.","keywords":["RuO2","altermagnetism","spin-resolved ARPES","spin texture","epitaxial strain","magnetic point group","time-reversal symmetry breaking","ultrathin films"],"falsifier":"A spin-resolved ARPES measurement of a nonmagnetic reference film with the same surface termination and measurement geometry, or a one-step photoemission calculation that includes spin-dependent final-state scattering, could settle it: if the mirror-even [001] polarization survives without magnetism, the magnetic conclusion collapses. A second check would be varying the photon energy to test whether the out-of-plane momentum $k_{110}$ is truly constant; if the $k_{001}\\sigma_{110}$ term is a genuine linear term, the symmetry analysis would instead require a lower-symmetry magnetic group incompatible with the mm2 structural symmetry.","tokens_in":18309,"feed_emoji":"🧲","tokens_out":11550,"duration_ms":97206,"temperature":0.7,"pith_summary":"Using spin-resolved ARPES on 2.7-nm epitaxially strained RuO2 films grown on a conducting TiO2 substrate, the paper reports a momentum-dependent photoelectron spin texture with two coexisting symmetry classes: a mirror-odd component expected from the polar, inversion-broken interface, and a mirror-even in-plane [001] component that the authors argue no nonmagnetic mechanism can produce. They conclude that the mirror-even component signals time-reversal symmetry breaking and that the observed pattern of spin-splitting terms is consistent with the magnetic point group m′m2′, with moments in the film plane. That magnetic point group admits both a ferromagnetic and a d-wave altermagnetic order parameter, so the phase would be a nonrelativistic spin structure stabilized by epitaxial strain in the ultra-thin limit, distinct from the nonmagnetic behavior established for bulk and strain-relaxed RuO2. The same measurements show strain-shifted narrow surface bands near the Fermi level, which accompany the new spin texture.","feed_headline":"Mirror-even spin texture signals magnetism in ultra-thin RuO2","feed_subtitle":"In a 2.7-nm strained film, a mirror-even electron spin polarization appears that symmetry analysis says no nonmagnetic state can produce.","key_machinery":"The load-bearing machinery is the irreducible-representation classification of spin-splitting terms for the paramagnetic point group $mm2.1'$, up to quadratic order in momentum (Table 1). The classification shows which combinations of spin polarization components $\\sigma_i$ and momentum products $k_i k_j$ transform under each irrep. The decisive move is treating the out-of-plane momentum $k_{110}$ as a constant in the ultra-thin film, so that the observed linear-looking $k_{001}\\sigma_{110}$ term is reinterpreted as the quadratic term $k_{110}k_{001}\\sigma_{110}$. Together with the mirror-even uniform ($\\sigma_{001}$) or $k_{110}^{2}\\sigma_{001}$ term and the Rashba term $k_{1\\bar{1}0}\\sigma_{001}$, all observed splittings then belong to the single irrep $B_1^-$, whose condensation gives the magnetic point group $m'm2'$. The spin-resolved ARPES measurements use very-low-energy electron diffraction spin detectors to select the in-plane [001] and out-of-plane [110] polarization directions.","core_discovery":"On its own terms, the paper's central discovery is that the photoelectron spin polarization of fully strained 2.7-nm RuO2 contains, near the Fermi level, an in-plane [001] component that is even under both the (1̄10) and (001) mirrors, in addition to the expected mirror-odd spin texture. Because such a mirror-even component is not allowed by any nonmagnetic spin-orbit mechanism in the polar point group mm2, the paper attributes it to time-reversal symmetry breaking. The symmetry analysis of the observed terms — the uniform $\\sigma_{001}$ or quadratic $k_{110}^{2}\\sigma_{001}$, the linear $k_{001}\\sigma_{110}$, and the Rashba-type $k_{1\\bar{1}0}\\sigma_{001}$ — yields, under the assumption that $k_{110}$ is a constant, a single magnetic order parameter $B_1^-$ and the magnetic point group $m'm2'$ with in-plane moments. The paper states that this magnetic group is consistent with both ferromagnetism and d-wave altermagnetism, and presents the result as direct spectroscopic evidence of a nonrelativistic spin structure in the ultra-thin strained regime of RuO2.","pith_inferences":["A decisive test the paper leaves implicit: a one-step photoemission calculation with spin-dependent final-state scattering, or a spin-resolved ARPES control on a nonmagnetic sample with identical geometry, would determine whether the mirror-even [001] polarization survives without magnetism.","If $m'm2'$ is the true magnetic group, the in-plane [001] moments should show up as a characteristic crystalline-axis dependence in magneto-optical Kerr rotation and planar Hall transport, which could also help distinguish ferromagnetism from d-wave altermagnetism.","A photon-energy-dependent spin-resolved ARPES scan on a series of strained film thicknesses could test the assumption that $k_{110}$ is a good constant; if the $k_{001}\\sigma_{110}$ term is genuinely linear in momentum, the magnetic point group assignment would have to be revised.","The strain-shifted surface-derived $\\alpha$ narrow bands sit at the same energies as the mirror-even polarization; spin-polarized slab calculations that include the surface $d_{z^2}$ states could reveal whether those bands enhance the magnetic instability."],"forward_implications":["The fully strained ultra-thin regime of RuO2 is a distinct electronic state: the nonrelativistic spin texture and broken time reversal appear below the roughly 4-nm fully strained thickness, while bulk and strain-relaxed films remain nonmagnetic.","The proposed magnetic point group $m'm2'$ has in-plane moments, distinguishing the low-temperature phase from the out-of-plane-moment $m'm'2$ phase inferred from room-temperature second-harmonic generation, even though both correspond to d-wave altermagnetism.","Because the mirror-odd Rashba-type component coexists with the mirror-even magnetic component, the system combines spin-orbit and nonrelativistic spin splittings, which could be exploited for spin-charge conversion.","A magnetic order parameter transforming as $B_1^-$ predicts additional spin-splitting terms beyond those measured, for example $k_{1\\bar{1}0}k_{001}\\sigma_{1\\bar{1}0}$ and $(k_{1\\bar{1}0}^{2}-k_{001}^{2})\\sigma_{001}$, providing a checklist for future spin-resolved measurements."],"supporting_citations":[{"why":"Defines altermagnetism as a collinear magnetic phase whose spin sublattices are related by rotations, the framework the paper claims strained ultra-thin RuO2 realizes.","marker":"1"},{"why":"Reports a d-wave spin texture in RuO2 single crystals by spin-resolved ARPES, providing the prior claim the present ultra-thin measurement extends and contrasts with.","marker":"44"},{"why":"Uses muon spin rotation and neutron diffraction to rule out magnetic order in bulk RuO2, establishing the nonmagnetic baseline that the ultra-thin strained result departs from.","marker":"45"},{"why":"Finds no altermagnetic spin splitting in spin-resolved ARPES of bulk and strain-relaxed RuO2 films, the main counter-result the new mirror-even polarization must be distinguished from.","marker":"54"},{"why":"Shows theoretically that magnetism in RuO2 is fragile and sensitive to parameters, supporting the paper's claim that strain can shift the system into a magnetic phase.","marker":"55"},{"why":"Characterizes fully strained ultra-thin RuO2/TiO2 as polar mm2 and reports second-harmonic and magneto-optical signatures of a time-reversal-symmetry-broken phase, the structural and phenomenological basis for the magnetic assignment.","marker":"57"},{"why":"Reports metallicity and anomalous Hall signals in atomically thin strained RuO2, corroborating a strain-stabilized magnetic state with transport probes.","marker":"60"},{"why":"Documents a spin-polarization effect in photoemission from nonmagnetic surfaces, the known multiple-scattering artifact the paper must assume does not generate the mirror-even signal.","marker":"66"}],"fun_headline_variants":["Ultra-thin strained RuO2 reveals mirror-even spin texture","Mirror-even spin texture in 2.7-nm RuO2 hints at magnetism","Strained RuO2 ultrathin film shows forbidden spin texture","Epitaxial strain triggers mirror-even spin texture in RuO2"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion rests on the assumption that the measured photoelectron spin polarization reflects the intrinsic spin of the initial electronic states, and not a spin polarization produced by spin-dependent scattering of the outgoing photoelectrons.","fun_headline_variants_meta":{"raw":{"variants":["Ultra-thin strained RuO2 reveals mirror-even spin texture","Mirror-even spin texture in 2.7-nm RuO2 hints at magnetism","Strained RuO2 ultrathin film shows forbidden spin texture","Epitaxial strain triggers mirror-even spin texture in RuO2"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00069,"raw_usage":{"total_tokens":3163,"prompt_tokens":1023,"completion_tokens":2140,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":639,"completion_tokens_details":{"reasoning_tokens":2060}},"tokens_in":639,"tokens_out":2140,"duration_ms":14281,"temperature":1.0,"reasoning_tokens":2060,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T15:49:23.095156+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A spin-resolved ARPES measurement of a nonmagnetic reference film with the same surface termination and measurement geometry, or a one-step photoemission calculation that includes spin-dependent final-state scattering, could settle it: if the mirror-even [001] polarization survives without magnetism, the magnetic conclusion collapses. A second check would be varying the photon energy to test whether the out-of-plane momentum $k_{110}$ is truly constant; if the $k_{001}\\sigma_{110}$ term is a genuine linear term, the symmetry analysis would instead require a lower-symmetry magnetic group incompatible with the mm2 structural symmetry.","supporting_citations":[{"cited_title":"URL https://www.nature.com/ articles/s44306-024-00055-y","cited_arxiv_id":null,"evidence_quote":"Uses muon spin rotation and neutron diffraction to rule out magnetic order in bulk RuO2, establishing the nonmagnetic baseline that the ultra-thin strained result departs from."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Finds no altermagnetic spin splitting in spin-resolved ARPES of bulk and strain-relaxed RuO2 films, the main counter-result the new mirror-even polarization must be distinguished from."},{"cited_title":"& Feder, R","cited_arxiv_id":null,"evidence_quote":"Documents a spin-polarization effect in photoemission from nonmagnetic surfaces, the known multiple-scattering artifact the paper must assume does not generate the mirror-even signal."}],"review_version":2}