{"id":"a95147f9-e6e7-43d6-bcc3-c2e8d8bef47e","arxiv_id":"2501.12961","paper_version":1,"verdict":"UNVERDICTED","confidence":"UNKNOWN","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"Adding up to 25% gold to 2-nm cobalt layers in epitaxial Re/CoAu/Pt stacks tunes perpendicular anisotropy, damping, and interfacial Dzyaloshinskii-Moriya interaction, with a spin-reorientation transition near 13% gold.","lead":"This paper grows atomically flat sandwiches of cobalt-gold alloy between rhenium and platinum, then measures how adding gold changes their magnetism. It reports that gold content tunes perpendicular anisotropy, magnetic damping, and the chiral interaction that stabilizes tiny magnetic textures, which matters for future spintronic memory devices.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The spin-pumping attribution in the abstract is not quantitatively supported: Eq. (3) assumes a fixed bulk Co α0=0.011 for all Au alloys, and the g↑↓ values in Table III are internally inconsistent with Eq. (3) by a factor close to 4π.","rationale":"The reader's weakest_assumption correctly identifies Eq. (3) and the assumption that α0 remains 0.011 for all Au concentrations. My independent check confirms a deeper numerical problem: the g↑↓ values in Table III are a factor of 4π too small relative to the published Eq. (3), as shown by direct substitution of Table II data. This means the abstract's quoted g↑↓ values (2.25–2.91×10^18 m^-2) are not trustworthy as printed. Even after correcting the 4π factor, the physically relevant question remains whether α0 changes with Au content. The paper's own RHEED data indicate increased disorder at high Au concentrations, which would contribute to the intrinsic damping, and the authors explicitly acknowledge that the damping increase for S4–S6 cannot be due to spin pumping only. Thus the headline claim of the abstract—that the damping rise 'can be attributed to the spin pumping effect'—is overstated and unsupported by the evidence presented. The experimental data (SQUID, FMR linewidths, BLS) may well be valid, and the trends in Ms, anisotropy, and DMI are plausible, but the spin-pumping extraction and its interpretation require a major revision. A conditional accept is appropriate: the manuscript should be accepted only after the g↑↓ calculation is corrected and the damping attribution is revised or supported by control experiments (e.g., Cu-spacer or symmetric stacking). No ad hominem is intended; the concern is entirely about the internal consistency and physical assumptions of the analysis.","tokens_in":14199,"tokens_out":9366,"duration_ms":91243,"concrete_test":"First, recompute g↑↓ from Eq. (3) using the Table II values (α, M_s, g, t=20 Å) for all samples; if the results exceed Table III by a factor of 4π, the numerical evaluation of Eq. (3) is wrong. Then, to settle the attribution of the damping rise to spin pumping, fabricate a companion series Re/Co₁₋ₓAuₓ(20 Å)/Cu(t_Cu)/Pt(30 Å) with t_Cu = 0, 1, 2, 3, 4 nm for x = 0, 0.10, 0.20, and measure α by VNA-FMR. If the residual α after subtracting the Cu-saturated (spin-pumping-free) value still increases with x, the damping rise is intrinsic to the alloy, invalidating Eq. (3)'s attribution.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central quantitative claim—that the measured increase in effective damping with Au concentration is attributable to spin pumping—rests on Eq. (3), α−α0 = (g μB / 4π M_s t_FM) g↑↓, with α0 fixed at the bulk Co value 0.011 (ref. 45). Two problems undermine this. First, using the tabulated α, M_s, g, and t=20 Å, Eq. (3) yields g↑↓ ≈ 28×10^18 m^-2 for S1, not the 2.25×10^18 m^-2 in Table III; the tabulated values are exactly a factor of 4π too small, indicating a unit/conversion error in the numerical evaluation. Even after correcting this factor, the values become ≈3×10^19 m^-2, which is high and requires justification. Second, the more fundamental issue is that α0 is assumed unchanged by Au alloying. The paper's own RHEED data show degraded crystallinity for x≥15%, which would raise the intrinsic Gilbert damping independently of spin pumping. The authors themselves concede (in the discussion after Fig. 5c) that for S4–S6, g↑↓ decreases while α continues to rise, so 'an enhancement of the effective damping cannot be due to spin pumping only.' Therefore the abstract's unqualified 'which can be attributed to the spin pumping effect' is not supported by the presented analysis.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports an MBE-grown epitaxial series Al2O3(0001)/Pt(400 Å)/Re(10 Å)/Co1-xAux(20 Å)/Pt(30 Å) with x = 0–25% and characterizes the static and dynamic magnetic properties with SQUID, VNA-FMR, BLS, RHEED, TEM, and MOKE. The central claims are that increasing Au concentration reduces Ms from 1690 to 982 kA/m, drives a spin-reorientation transition at about 13% Au, raises Keff up to 0.33 MJ/m3, increases the effective Gilbert damping from 0.0242 to 0.0386, and decreases the surface iDMI from 2.62 to 1.14 pJ/m. The damping increase is attributed in the abstract to spin pumping, and effective spin mixing conductances are extracted using Eq. (3).","tokens_in":14472,"tokens_out":7052,"duration_ms":71280,"significance":"If the reported trends hold, the paper provides a useful experimental demonstration that Au composition is a practical tuning parameter for PMA, damping, and iDMI in epitaxial Re/Co/Pt-type heterostructures. The multi-technique dataset is substantial and the static trends (Ms, Keff, SRT, iDMI) are internally consistent with the stated analysis. The comparison of the measured iDMI with literature values for Pt/Co/Re, W/Co/Pt, and Pt/Co/Ir is informative. However, the quantitative spin-pumping analysis contains an internal numerical inconsistency, and the abstract's attribution of the damping rise to spin pumping is contradicted by the paper's own discussion. These issues must be resolved before the quantitative claims can be accepted.","major_comments":[{"comment":"The values of g↑↓ in Table III do not follow from Eq. (3) as written. For S1, Eq. (3) with α = 0.0242, α0 = 0.011, g = 2.13, Ms = 1690 kA/m, and tFM = 20 Å gives Δα = 0.0132 and g↑↓ = Δα / [(g μB)/(4π Ms tFM)] ≈ 2.8×10^19 m^-2, not the tabulated 2.25×10^18 m^-2. The tabulated values correspond to omitting the factor 4π from the denominator. This is a factor-4π unit/conversion error in the numerical evaluation. After correction the values are about 2.8–3.7×10^19 m^-2, which is unusually high and requires justification; the manuscript should discuss whether such large effective conductances are physically plausible for these interfaces and should clarify whether both the Re/Co and Co/Pt interfaces are included.","section":"Eq. (3), Tables II and III"},{"comment":"The abstract states that the rise of effective damping with Au concentration 'can be attributed to the spin pumping effect,' and the Summary says 'We attribute this mainly to the spin-pumping phenomena.' This overstates the paper's own analysis. The discussion after Fig. 5(c) states that for S4–S6 the extracted g↑↓ decreases while the damping continues to rise, and concludes that 'an enhancement of the effective damping cannot be due to spin pumping only.' These statements are mutually inconsistent. The abstract and Summary should be revised to reflect the more nuanced conclusion that spin pumping contributes at low Au concentrations but that other channels (magnetic proximity effect, radiative damping, disorder, or the SRT transition) also play a role at higher x.","section":"Abstract and Summary vs. discussion after Fig. 5(c)"},{"comment":"The extraction of g↑↓ assumes α0 = 0.011, the bulk Co value, for every alloy composition. This assumption is load-bearing because the entire excess damping is assigned to spin pumping. The paper provides no justification that the intrinsic Gilbert damping of Co1-xAux is identical to bulk Co, and the RHEED data in Fig. 3 show that the streaks become 'blurry and less distinct' for x = 15–25%, indicating degraded crystalline quality that could raise the intrinsic damping independently of spin pumping. The TEM image for S4 shows a sharp interface, but it does not rule out intralayer alloy disorder. The authors should either model α0(x), provide a control experiment, or at a minimum state explicitly that the reported g↑↓ values are upper bounds under this assumption.","section":"Eq. (3) and Fig. 3"},{"comment":"No error bars or uncertainties are reported for Ms, Keff, α, g↑↓, or Ds. This is particularly important for the claim that g↑↓ has a maximum at S3 (10% Au): the differences between S2 (2.64), S3 (2.91), and S4 (2.71) are of order 10%, and without uncertainties it is impossible to assess whether the non-monotonic behavior is significant. The claimed SRT at approximately 13% Au also lacks an uncertainty. The manuscript should provide error estimates for all extracted parameters, including fit uncertainties from the FMR linewidth and BLS frequency-asymmetry analyses.","section":"Tables II and III, Figs. 2 and 5"}],"minor_comments":[{"comment":"The sentence 'the sign of effective magnetization changes from positive to negative at around 13% Au' should read 'from negative to positive,' since the tabulated μ0Meff values go from -0.44 T for Co to +0.11 T for Co80Au20.","section":"Text near Fig. 2(a)"},{"comment":"Equation (4) is garbled ('D_eff = Δf ... 2πk') and cannot be evaluated as printed. Please provide the standard BLS DMI expression with all symbols and units defined.","section":"Eq. (4)"},{"comment":"The Summary refers to 'Re(10 Å)/Co(20 Å)/Pt(20 Å)' and 'Re(10 Å)/Co75Au25(20 Å)/Pt(20 Å),' but the samples studied have a Pt(30 Å) cap. Please correct the stack notation.","section":"Summary"},{"comment":"The reference list contains malformed entries, e.g., ref. 21 is a patent citation with incomplete authors and title, and ref. 50 is a thesis citation with garbled title. Please format all references consistently.","section":"References"},{"comment":"Several thickness labels in Fig. 6 are corrupted (e.g., 'tS= 19.90Å4.5Å'), which makes the SRT values difficult to read. Please provide a clean version with legible labels.","section":"Fig. 6"},{"comment":"The manuscript contains numerous OCR-type errors in equations and symbols (e.g., 'Keff' typeset inconsistently, 'μBΔH' formatting). A careful proofreading pass is needed.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The experimental dataset is valuable and the static magnetic trends appear sound, but the current quantitative spin-pumping analysis is not correct as presented. The factor-4π discrepancy in Table III is easily fixed by recalculating with Eq. (3) as written, but the corrected values are much larger and will require discussion; the abstract and Summary also need to be aligned with the authors' own caveat that spin pumping alone cannot explain the damping rise for S4–S6. I recommend major revision and re-review after the recalculation and revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a legitimate experimental paper with a useful new data set, but the headline spin-pumping result is not reliable as printed. The static properties—SRT, PMA, Ms, DMI—are likely to be of real use if the numbers hold up. The damping analysis needs a rewrite.\n\nWhat's new: first systematic study of epitaxial Re/Co1-xAux/Pt across x=0-25% for iDMI, damping, and anisotropy. The RHEED/TEM work shows epitaxial growth, and the trends (Ms drops, easy axis flips near 13% Au, DMI decreases with Au) are internally consistent. The SRT thickness vs Re-buffer data for the alloys is also new. The DMI values for pure Co, Ds=2.62 pJ/m, are consistent with additive Re/Co and Co/Pt interfaces.\n\nSoft spots, in order of severity. First, Table III does not follow from Eq. (3). Using their S1 values (α=0.0242, Ms=1690 kA/m, t=20 Å, g=2.13) and α0=0.011, Eq. (3) gives g↑↓ = 2.8×10^19 m^-2, not 2.25×10^18 m^-2. The tabulated values are exactly ~4π too small, which smells like a missing 4π in the numerical code. Second, the abstract says the damping rise 'can be attributed to the spin pumping effect,' but the paper's own discussion (after Fig. 5c) says for S4-S6 'an enhancement of the effective damping cannot be due to spin pumping only.' That is a direct contradiction. Third, fixing α0 at the bulk Co value 0.011 for all alloys ignores that Au doping degrades crystallinity (their own RHEED shows this), which would raise intrinsic damping independently. So the g↑↓ values and the spin-pumping attribution are not quantitatively supported. Fourth, no error bars anywhere on Ms, α, Ds, or g↑↓, which makes it hard to tell whether the 4-9% damping steps are real.\n\nThe positive side is that the static and DMI trends do not depend on the disputed spin-pumping analysis. The SRT transition near 13% Au and the DMI decrease with Au are the real results here.\n\nWho it's for: experimentalists working on composition-tuned PMA/DMI heterostructures for skyrmionics or MRAM, and anyone benchmarking Co alloy damping.\n\nRecommendation: send it to a serious referee, but expect major revision. The authors must fix the 4π factor, correct the abstract, and either justify α0 or measure it as a function of x. If the damping analysis is fixed, the paper has a useful niche.","headline":"Solid new epitaxial CoAu data set undermined by a concrete factor-4π error in the spin-pumping analysis and an abstract that overstates it.","tokens_in":15127,"tokens_out":4025,"would_cite":false,"duration_ms":39687,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["75.70.-i","75.30.Gw","76.50.+g"],"model":"deepseek-v4-flash","headline":"The paper shows that raising the gold fraction in epitaxial Re/CoAu/Pt trilayers from 0 to 25% systematically reorients the easy axis, raises damping, and lowers chiral DMI, making Au content a single tuning knob for spintronic stacks.","keywords":["perpendicular magnetic anisotropy","spin reorientation transition","magnetic damping","spin pumping","Dzyaloshinskii-Moriya interaction","Co-Au alloy","ferromagnetic resonance","Brillouin light scattering"],"falsifier":"Measure the FMR damping of Re/Co$_{1-x}$Au$_x$ stacks with a thin Cu or MgO spacer inserted between the CoAu layer and the Pt cap to block spin pumping: if the damping still rises with Au content, the central attribution to spin pumping would be disproved and the $g^{\\uparrow\\downarrow}$ values extracted from Eq. (3) would be artifacts.","tokens_in":14004,"feed_emoji":"🧲","tokens_out":10806,"duration_ms":91524,"temperature":0.7,"pith_summary":"This paper investigates epitaxial Re/Co$_{1-x}$Au$_x$/Pt heterostructures with $x$ from 0 to 25% and argues that the gold concentration of the 20-Å cobalt layer is a practical tuning parameter for its magnetism. Increasing $x$ lowers the saturation magnetization from 1690 to 982 kA/m, moves the easy axis from in-plane to out-of-plane through a spin-reorientation transition near 13% Au, raises the effective Gilbert damping from 0.0242 to 0.0386, and reduces the interfacial Dzyaloshinskii–Moriya interaction from 2.62 to 1.14 pJ/m. The authors attribute the damping rise largely to spin pumping, with the largest effective spin-mixing conductance $g^{\\uparrow\\downarrow} \\approx 2.91\\times10^{18}$ m$^{-2}$ at 10% Au. If correct, a single composition knob can engineer perpendicular anisotropy, fast magnetization relaxation, and chiral spin textures in this stack.","feed_headline":"Gold doping flips easy axis and tunes damping in Re/Co/Pt stacks","feed_subtitle":"Raising Au from 0 to 25% rotates the magnetization direction while changing anisotropy, damping, and chiral spin interactions.","key_machinery":"The carrying object is the epitaxial trilayer Re(10 Å)/Co$_{1-x}$Au$_x$(20 Å)/Pt(30 Å), grown by molecular-beam epitaxy on a Pt(400 Å) buffer, with asymmetric heavy-metal interfaces that supply both the perpendicular-anisotropy field and the chiral Dzyaloshinskii–Moriya interaction. The quantitative machinery combines SQUID magnetometry for $M_\\mathrm{s}$ and $K_{\\mathrm{eff}}$, vector-network-analyzer ferromagnetic resonance for the resonance condition and linewidth, and Brillouin light scattering for the Stokes–anti-Stokes frequency asymmetry that gives the DMI constant. The spin-pumping interpretation is carried by Eq. (3), which converts the excess damping $\\alpha - \\alpha_0$ into an effective spin-mixing conductance $g^{\\uparrow\\downarrow}$; this is the step that makes the damping data speak about interfacial spin transport.","core_discovery":"The central claim is that diluting the cobalt layer with gold in an epitaxial Re(10 Å)/Co$_{1-x}$Au$_x$(20 Å)/Pt(30 Å) stack gives systematic control over the magnetism: saturation magnetization falls from 1690 to 982 kA/m, the easy axis switches from in-plane to out-of-plane near $x \\approx 0.13$, and the effective anisotropy constant $K_{\\mathrm{eff}}$ reaches 0.33 MJ/m$^3$. The effective damping rises with Au concentration, which the paper attributes mainly to spin pumping, supported by an extracted effective spin-mixing conductance that peaks at $2.91\\times10^{18}$ m$^{-2}$ for Co$_{90}$Au$_{10}$; the paper also notes that damping keeps rising while $g^{\\uparrow\\downarrow}$ falls for higher Au content, so spin pumping alone cannot explain the full trend. The interfacial Dzyaloshinskii–Moriya interaction decreases with Au content, from 2.62 pJ/m in pure Co to 1.14 pJ/m in Co$_{75}$Au$_{25}$, roughly tracking the saturation magnetization. The paper further shows that both Au content and Re buffer thickness raise the thickness at which the easy axis reorients, which is what a device engineer would need for thicker perpendicularly magnetized layers.","pith_inferences":["The non-monotonic $g^{\\uparrow\\downarrow}$ trend, which rises to 10% Au and then falls while damping keeps rising, implies that at high Au content the extra damping comes from channels other than spin pumping; inserting a spacer layer between CoAu and Pt would separate those channels.","The same composition-tuning approach should generalize to other immiscible ferromagnet/noble-metal pairs, giving a generic alloy-composition knob for skyrmion-host materials.","The claimed link between DMI and $M_\\mathrm{s}$ could be tested by varying temperature instead of composition; the correlation should collapse if chemical disorder, rather than magnetization, drives the DMI reduction.","The epitaxial stabilization of the normally immiscible Co–Au alloy suggests MBE-grown alloy films can serve as clean model systems for composition-dependent spin-orbit phenomena."],"forward_implications":["Au concentration provides a one-parameter route to perpendicular magnetic anisotropy in thicker Co layers, because the spin-reorientation threshold moves to larger CoAu thickness as $x$ rises.","Tuning Au from 0 to 25% raises the effective damping from 0.0242 to 0.0386, which is useful for applications that want faster magnetization switching.","The effective spin-mixing conductance peaks at 10% Au, indicating an optimal composition for spin-current transfer before other damping channels grow stronger.","Because interfacial DMI falls from 2.62 to 1.14 pJ/m as Au increases, a device designer can choose a composition at which perpendicular anisotropy, damping, and chiral interaction are mutually compatible.","The observed near-linear tracking of DMI with saturation magnetization implies that the same interfacial spin-orbit mechanism controls both, so one measurement can be used to estimate the other."],"supporting_citations":[{"why":"Supplies the bulk Co damping $\\alpha_0 = 0.011$ used in Eq. (3) to isolate the spin-pumping contribution.","marker":"[45]"},{"why":"Provides the formula connecting excess damping to the effective spin-mixing conductance $g^{\\uparrow\\downarrow}$ in Eq. (3).","marker":"[44]"},{"why":"Gives Kittel's resonance condition (Eq. 1) used to extract effective magnetization and g-factor from FMR field sweeps.","marker":"[30]"},{"why":"Gives the linewidth-versus-frequency relation (Eq. 2) used to extract the effective Gilbert damping.","marker":"[31]"},{"why":"Establishes the additive DMI sign convention for Re/Co and Co/Pt interfaces that the paper assumes also holds for CoAu alloys.","marker":"[26]"},{"why":"Provides the BLS method for quantifying DMI from the Stokes–anti-Stokes frequency asymmetry used here.","marker":"[25]"},{"why":"Demonstrates that a Cu spacer suppresses spin pumping and magnetic proximity effects, the baseline against which the damping interpretation is framed.","marker":"[39]"},{"why":"Reports Au magnetic moments in Au–Co nanoalloys, supporting the proximity-magnetization mechanism invoked for the enhanced damping.","marker":"[13]"}],"fun_headline_variants":["Au doping flips easy axis and boosts damping in Re/Co/Pt","Cobalt-gold alloy drives spin reorientation in Re/Co/Pt stacks","Gold in Co layer tunes anisotropy, damping, and DMI","A little Au in Co switches easy axis and raises loss"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the intrinsic damping of the cobalt layer stays fixed at the bulk value $\\alpha_0 = 0.011$ as gold is added, so the entire measured increase in damping can be assigned to interfacial spin pumping; the paper itself concedes that the spin-mixing conductance falls at higher gold concentrations while damping continues to rise.","fun_headline_variants_meta":{"raw":{"variants":["Au doping flips easy axis and boosts damping in Re/Co/Pt","Cobalt-gold alloy drives spin reorientation in Re/Co/Pt stacks","Gold in Co layer tunes anisotropy, damping, and DMI","A little Au in Co switches easy axis and raises loss"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00074,"raw_usage":{"total_tokens":3460,"prompt_tokens":1256,"completion_tokens":2204,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":872,"completion_tokens_details":{"reasoning_tokens":2127}},"tokens_in":872,"tokens_out":2204,"duration_ms":17504,"temperature":1.0,"reasoning_tokens":2127,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T16:36:00.440901+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the FMR damping of Re/Co$_{1-x}$Au$_x$ stacks with a thin Cu or MgO spacer inserted between the CoAu layer and the Pt cap to block spin pumping: if the damping still rises with Au content, the central attribution to spin pumping would be disproved and the $g^{\\uparrow\\downarrow}$ values extracted from Eq. (3) would be artifacts.","supporting_citations":[{"cited_title":"M., Sinha, J., Hindmarch, A","cited_arxiv_id":null,"evidence_quote":"Supplies the bulk Co damping $\\alpha_0 = 0.011$ used in Eq. (3) to isolate the spin-pumping contribution."},{"cited_title":"A., Kakazei, G","cited_arxiv_id":null,"evidence_quote":"Provides the formula connecting excess damping to the effective spin-mixing conductance $g^{\\uparrow\\downarrow}$ in Eq. (3)."},{"cited_title":"On the theory of ferromagnetic resonance absorption","cited_arxiv_id":null,"evidence_quote":"Gives Kittel's resonance condition (Eq. 1) used to extract effective magnetization and g-factor from FMR field sweeps."},{"cited_title":"& Albrecht, M","cited_arxiv_id":null,"evidence_quote":"Gives the linewidth-versus-frequency relation (Eq. 2) used to extract the effective Gilbert damping."},{"cited_title":"& Ando, K","cited_arxiv_id":null,"evidence_quote":"Establishes the additive DMI sign convention for Re/Co and Co/Pt interfaces that the paper assumes also holds for CoAu alloys."},{"cited_title":"Y., Kim, N","cited_arxiv_id":null,"evidence_quote":"Provides the BLS method for quantifying DMI from the Stokes–anti-Stokes frequency asymmetry used here."},{"cited_title":"C., Reyren, N., Laczkowski, P., Savero, W., Attané, J","cited_arxiv_id":null,"evidence_quote":"Demonstrates that a Cu spacer suppresses spin pumping and magnetic proximity effects, the baseline against which the damping interpretation is framed."},{"cited_title":"& Mattei, G","cited_arxiv_id":null,"evidence_quote":"Reports Au magnetic moments in Au–Co nanoalloys, supporting the proximity-magnetization mechanism invoked for the enhanced damping."}],"review_version":1}