{"id":"60705107-26ba-4a64-b4d3-61bc66b0bba2","arxiv_id":"2608.09611","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A new superconducting alloy PtIrSiGe with 1.9 T out-of-plane critical field and 1.85 K critical temperature is integrated with a high-mobility buried germanium quantum well, showing gate-tunable Josephson supercurrents.","lead":"Researchers made a superconducting thin film alloy, platinum iridium germanosilicide, that stays superconducting in out-of-plane magnetic fields up to 1.9 tesla and at temperatures near 1.85 kelvin. They built it into gate-controlled Josephson junctions on a germanium quantum well without degrading the high hole mobility, a step toward germanium-based hybrid quantum devices.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the gate-tunable supercurrent and SQUID interference data substantiate the proximitisation claim, and the remaining limitations are explicitly acknowledged without undermining the central result.","rationale":"The paper's central claim has two components: (i) PtIrSiGe is a superconductor with high out-of-plane critical field and accessible T_c; (ii) a buried Ge quantum well is proximitized in a gate-defined Josephson junction/SQUID. The first component is supported by four-terminal thin-film transport measurements up to B_perp = 1.9 T at 84 mK and T_c = 1.85 K, with a second sample on the same chip as the SQUID showing 1.7 T. The second component is supported by gate-tunable switching currents, Fraunhofer-like oscillations, and SQUID interference with the expected loop area. The reader's weakest assumption—the absence of a parasitic superconducting path—is substantially mitigated by the gate-tunability: a parasitic shunt would not vanish at V_g = 0 in an accumulation-mode device. The acknowledged absence of an induced-gap measurement and the non-uniform PtIrSiGe interface are limitations for future qubit use, but they do not falsify the demonstrated proximity-induced supercurrent. I considered whether the lack of phase/composition identification of 'PtIrSiGe' could be a load-bearing issue, but the central device claim is compositional rather than phase-specific, and the transport properties are attributed to the reacted layer as a whole. Overall, the argument is internally consistent, the evidence is appropriate for a proof-of-principle claim, and the paper's own limitations are stated candidly. I therefore maintain the ACCEPT verdict without change, while noting that a direct induced-gap measurement would considerably strengthen the proximity claim.","tokens_in":10248,"tokens_out":10716,"duration_ms":110094,"concrete_test":"Perform tunnelling spectroscopy on a quantum dot or gate-defined tunnel barrier in the same 25-nm-deep Ge quantum well proximitized by PtIrSiGe, measuring dI/dV versus bias at T ≈ 20 mK; observation of a hard induced gap with magnitude consistent with the parent superconducting gap would directly confirm proximitisation and rule out residual metallic shunting. If no gap is resolvable, the central 'proximitisation' claim would need to be downgraded to 'Josephson supercurrent demonstrated', though the transport evidence already reported would still support the latter.","verdict_should_be":"UNCHANGED","load_bearing_attack":"After reading the full manuscript, I do not find a load-bearing concern that invalidates the central claim. The reader's weakest assumption—that the supercurrent flows through the buried Ge quantum well rather than a parasitic PtIrSiGe shunt—is largely addressed by the data in Fig. 3(a) and (c): in this accumulation-mode heterostructure, no transport is observed at V_g = 0 for junction 2, and the switching current turns on only when the gate accumulates holes in the quantum well. A metallic or superconducting shunt would not be switched off by the top gate. The SQUID oscillations in Fig. 3(e), fit with an effective loop area of 10.47–10.48 µm² against a geometric area of 10 µm², provide independent evidence of coherent two-junction interference through the gated semiconductor. The remaining weaknesses are real but non-fatal: no induced superconducting gap is reported, and the STEM cross-section shows non-uniform PtIrSiGe diffusion. Both are acknowledged in the Discussion ('further process development is required to establish a uniform and reproducible interface' and 'further work is needed to establish the size and quality of the superconducting gap'). These limit the strength of the proximity claim for qubit applications but do not contradict the demonstrated Josephson supercurrent. The claimed material is a platinum iridium germanosilicide, and while the paper does not establish single-phase composition, the device-level claims do not require phase purity. I therefore raise no significant objection.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a lithographically defined superconducting contact material, PtIrSiGe, formed by annealing Pt/Ir on a Ge/SiGe heterostructure, and its use in a gate-defined Josephson junction and SQUID on a buried Ge quantum well. Thin-film four-terminal measurements on two samples give T_c about 1.85 K and out-of-plane critical fields of 1.7 T and 1.9 T at base temperature. A Hall bar with PtIrSiGe contacts shows a hole mobility of 1.3e6 cm2/Vs at 4.3e11 cm-2, indicating that the 500 C anneal does not destroy the quantum well. HAADF-STEM and EDX show a PtIrSiGe alloy that diffuses unevenly into the heterostructure and contacts the well in places. In the SQUID, the switching currents of two independently gated junctions are tunable, and SQUID oscillations fit Eq. (1) with a loop area of 10.47-10.48 um2 against a lithographic area of 10 um2. The authors explicitly acknowledge that the interface is not uniform and that the induced superconducting gap has not yet been characterized.","tokens_in":10524,"tokens_out":8930,"duration_ms":83784,"significance":"The result is significant for Ge-based hybrid quantum devices because it combines a superconductor with out-of-plane critical field approaching 2 T and T_c near 1.85 K with a fabrication process that is compatible with standard ohmic-contact lithography and preserves a high-mobility buried Ge well. The main claims rest on direct measurements: thin-film superconductivity is reproduced in two independent samples, mobility retention is quantified on a Hall bar, and gate-tunable supercurrent plus SQUID interference provide functional evidence of proximitisation. The fit to Eq. (1) is a consistency check rather than a fitted conclusion, since the extracted loop area agrees with the lithographic value. The acknowledged limitations--non-uniform PtIrSiGe diffusion and the absence of an induced-gap measurement--moderate the strength of the proximity claim for qubit applications but do not contradict the demonstrated Josephson supercurrent.","major_comments":[],"minor_comments":[{"comment":"The gate voltage for junction 2 is given as -1.9 mV in the text and caption, but the corresponding measurement and the stated asymmetry I_c2 about 2 I_c1 imply -1.9 V; please correct the unit.","section":"Fig. 3(d) and Section IV"},{"comment":"The notation 'I_c1,2(B A_1,2)' is ambiguous because the equation as printed treats I_c1 and I_c2 as constants; please clarify how the individual Fraunhofer envelopes are incorporated into the fit.","section":"Eq. (1)"},{"comment":"The defining criteria for T_c = 1.85 K and for B_perp,c are not stated; please specify, for example, whether these are zero-resistance or 50% normal-resistance thresholds.","section":"Section II"},{"comment":"The Hall-bar mobility comparison would be easier to evaluate if the measurement temperature and the Hall-bar geometry were given in the main text.","section":"Section III"},{"comment":"The statement that 'at zero gate voltage... there is no transport across junction 2' is confusing in the context of Fig. 3(a), where V_g,2 = 0 while V_g,1 is swept; please clarify that both junctions are off at V_g = 0.","section":"Section IV"},{"comment":"The claim that annealing at 550 C leads to a breakdown of superconductivity is only in the Supplementary Material; a brief explanation in the main text would make the thermal budget clear.","section":"Discussion"},{"comment":"Adding explicit labels for the Al2O3, SiGe barriers, and Ge quantum well directly on the STEM/EDX images would help the reader evaluate the 'buried quantum well' and the uneven diffusion.","section":"Fig. 2(d)-(e)"},{"comment":"The material name 'platinum iridium germanosilicide' could be read as implying a stoichiometric compound, while the EDX data show an intermixed alloy; wording such as 'alloy formed by the process' would be safer.","section":"Section III"}],"recommendation":"minor_revision","confidential_remarks":"The manuscript is a well-executed empirical device demonstration and is within the scope of cond-mat.mes-hall. I found no load-bearing technical error; the minor revisions are largely clarifications. The explicit data availability statement and the transparency about the missing induced-gap measurement are strengths."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a serious referee. The real news is PtIrSiGe: a lithography-compatible superconductor on Ge with out-of-plane critical field up to 1.9 T and Tc ~1.85 K, demonstrated in two thin-film samples. That is roughly an order of magnitude better in B_perp than the earlier PtSiGe contacts, and it survives a ~500 C anneal without wrecking a 25 nm deep Ge quantum well (mobility 1.3e6 cm2/Vs). The device-level data back the proximitisation claim: the gate-tunable switching current in the accumulation-mode SQUID turns on only when the gate accumulates holes, and the Fraunhofer-like and SQUID oscillations fit with a loop area 10.47-10.48 um2 versus 10 um2 geometric. That is coherent two-junction interference through the gated Ge well, not a metallic shunt.\n\nThe soft spots are genuine but explicitly acknowledged: there is no induced-gap measurement, the PtIrSiGe/Ge interface is visibly uneven in STEM, and the paper does not establish a single-phase composition. The uneven interface is actually used to explain the non-ideal Fraunhofer pattern, which is consistent. I agree with the stress-test that the parasitic-shunt worry is not supported: if a continuous metallic path were carrying the supercurrent, the top gate would not turn it off.\n\nOne thing I would like in revision is more detail on how the switching current was extracted and some statistics or error bars on the extracted loop areas. But that is a referee-level request, not a blocker. The citation and data-sharing practices are fine: the data are on Zenodo and the key comparisons (PtSiGe, IrSiGe films, Ta germanide, granular Al) are cited.\n\nThis paper is for the Ge-hybrid and superconductor-semiconductor community. It deserves a serious referee. I would send it out.","headline":"A solid experimental demonstration of a new high-field superconductor integrated with a buried Ge quantum well; the proximity claim is well supported, with the missing gap measurement openly flagged.","tokens_in":11163,"tokens_out":3306,"would_cite":true,"duration_ms":27355,"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":"Annealing a Pt/Ir stack into platinum iridium germanosilicide yields a superconductor that tolerates 1.9 T out-of-plane and induces superconductivity in a buried germanium quantum well without degrading its mobility.","keywords":["platinum iridium germanosilicide","germanium quantum well","Josephson junction","SQUID","proximity effect","hole gas mobility","critical magnetic field","superconducting hybrid device"],"falsifier":"Measure the local density of states in the quantum well under a Josephson junction by tunneling spectroscopy: if no induced superconducting gap appears while the switching current persists, or if a fully depleted well still conducts supercurrent, the proximitisation claim is refuted.","tokens_in":10048,"feed_emoji":"🧲","tokens_out":16138,"duration_ms":126007,"temperature":0.7,"pith_summary":"This paper claims a new lithography-friendly superconductor for germanium quantum devices: platinum iridium germanosilicide (PtIrSiGe), produced by annealing a 2 nm platinum / 8 nm iridium stack on a Ge/SiGe heterostructure at about $500\\,^\\circ\\mathrm{C}$. The thin film superconducts up to an out-of-plane magnetic field of $B_{\\perp,\\mathrm{c}} = 1.9\\,\\mathrm{T}$ at 84 mK, with a critical temperature near $T_\\mathrm{c}\\sim1.85\\,\\mathrm{K}$, and remains superconducting at 400 mK in an in-plane field of 1 T. A Hall bar made with PtIrSiGe contacts keeps a hole mobility of $\\mu=1.3\\times10^{6}\\,\\mathrm{cm^2\\,V^{-1}\\,s^{-1}}$, showing the anneal preserves the buried quantum well. In a gate-defined SQUID, the authors observe gate-tunable switching currents and Fraunhofer-like oscillations, which they interpret as proximity-induced supercurrent through the quantum well. If correct, the result would make high-critical-field hybrid devices—spin qubits at fields above 1 T, gatemons, Andreev spin qubits, and Kitaev chains—accessible with standard top-down fabrication.","feed_headline":"A 1.9-tesla superconductor lands on a germanium quantum well","feed_subtitle":"A lithography-made PtIrSiGe film turns a buried Ge well into a tunable SQUID without losing its high hole mobility.","key_machinery":"The central object is the PtIrSiGe alloy formed by solid-phase reaction of a thin Pt/Ir bilayer with the SiGe barrier during rapid thermal annealing at $500\\,^\\circ\\mathrm{C}$. It does two jobs at once: it is the superconducting material, and it is the contact that reaches the quantum well without an etch step. The load-bearing detail is the platinum interlayer: iridium alone reacts by outward silicon and germanium migration, creating voids and damaging the heterostructure, while the 2 nm Pt layer mediates inward diffusion so that the alloy contacts the quantum well in some regions, with the contact depth varying laterally. That variation is what the authors use to explain the non-ideal Fraunhofer patterns, and the alloy's high critical field is what allows the SQUID to operate in out-of-plane fields up to the tesla range.","core_discovery":"The paper demonstrates that PtIrSiGe is a superconductor with critical temperature $T_\\mathrm{c}\\sim1.85\\,\\mathrm{K}$ and out-of-plane critical field $B_{\\perp,\\mathrm{c}}$ up to $1.9\\,\\mathrm{T}$ (with a second film reaching $1.7\\,\\mathrm{T}$), and that the same alloy can be used as the superconducting contact to a $25\\,\\mathrm{nm}$-deep germanium quantum well. The platinum underlayer is decisive: without it, iridium pulls germanium and silicon out of the heterostructure and leaves voids, whereas with it the alloy diffuses downward in places, contacting the well while the quantum well remains structurally pristine after the $500\\,^\\circ\\mathrm{C}$ anneal. Transport on a Hall bar yields a peak hole mobility of $\\mu=1.3\\times10^{6}\\,\\mathrm{cm^2\\,V^{-1}\\,s^{-1}}$ at density $p=4.3\\times10^{11}\\,\\mathrm{cm^{-2}}$, comparable to values reported for quantum wells contacted by normal metals. In a SQUID with two gate-tunable Josephson junctions, the critical current responds to top-gate voltage, shows Fraunhofer-like oscillations in out-of-plane field, and the SQUID interference pattern fits an asymmetric-junction model with a loop area near the designed $10\\,\\mu\\mathrm{m}^2$. The authors therefore claim a proof-of-principle proximitisation of a buried, high-mobility germanium quantum well by a high-critical-field, top-down-lithography-compatible superconductor.","pith_inferences":["The two-metal recipe may be a general materials design rule: for any metal that reacts with a semiconductor by outward diffusion and tends to form voids, a thin inner layer that diffuses inward could suppress voiding and enable contacts to deep quantum wells.","If the high out-of-plane critical field is partly a thin-film or interface effect, narrowing the PtIrSiGe leads—a strategy already noted for PtSiGe—could push the critical field beyond the 1.9 T reported here.","A natural next experiment is to embed the same junction in a gatemon or transmon circuit, testing whether the proximitised well supports coherent superconducting qubits rather than only d.c. supercurrent.","Systematically varying the Pt:Ir ratio would reveal whether the trade-off between field resilience and interface uniformity is tunable, potentially giving a reproducible contact while retaining most of the $1.9\\,\\mathrm{T}$ ceiling."],"forward_implications":["Germanium spin-qubit devices could operate at out-of-plane fields above 1 T while the contacts remain superconducting, easing spin-blockade readout and exploiting the larger out-of-plane $g$-factor.","Hybrid germanium architectures such as gatemons, Andreev spin qubits, and Kitaev chains could be fabricated with the same top-down lithography flow used for ohmic contacts, without etching or ultra-shallow wells.","Because the $500\\,^\\circ\\mathrm{C}$ anneal is comparable to the strained germanium epitaxy temperature, the contact step does not by itself impose a mobility ceiling on deep, low-disorder quantum wells.","The laterally varying diffusion depth must be included in quantitative device models, since it directly produces the irregular Fraunhofer pattern and an effectively position-dependent junction length."],"supporting_citations":[{"why":"Supplies the earlier PtSiGe-based proximitisation of germanium and the hard-gap measurement that this work extends.","marker":"[4]"},{"why":"Provides the low-disorder Ge/SiGe heterostructure and the mobility benchmark used to judge whether the 500 °C anneal degrades the well.","marker":"[22]"},{"why":"Reports epitaxial tantalum germanide with an out-of-plane critical field of 1.88 T and critical temperature near 2 K, the comparator for a high-field germanium-compatible superconductor.","marker":"[26]"},{"why":"Systematically characterises superconducting platinum-group germanide and germanosilicide films, including IrSiGe with critical fields up to 2.8 T, the property target motivating PtIrSiGe.","marker":"[27]"},{"why":"Explains that platinum silicide forms by inward platinum diffusion while iridium silicide forms by outward silicon migration, the mechanistic basis for the platinum underlayer.","marker":"[31]"},{"why":"Shows that platinum incorporation suppresses voiding and oxidation in germanosilicide formation, supporting the alloy design used here.","marker":"[32]"},{"why":"Gives the asymmetric-SQUID interference formula used to fit the measured SQUID oscillations and extract the effective loop area.","marker":"[33]"}],"fun_headline_variants":["PtIrSiGe superconductor hits 1.9 T on a buried Ge well","Superconducting PtIrSiGe preserves Ge mobility in SQUID","1.9-T superconductor integrates with high-mobility Ge well","Ge quantum well proximitized by field-resilient PtIrSiGe","Tunable SQUID from buried Ge well and 1.9-T superconductor"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the supercurrent observed in the SQUID flows through the buried germanium quantum well, not through an unintended superconducting or metallic path created by the uneven PtIrSiGe diffusion, since the STEM images show contact to the well only in some regions and no induced gap has yet been measured.","fun_headline_variants_meta":{"raw":{"variants":["PtIrSiGe superconductor hits 1.9 T on a buried Ge well","Superconducting PtIrSiGe preserves Ge mobility in SQUID","1.9-T superconductor integrates with high-mobility Ge well","Ge quantum well proximitized by field-resilient PtIrSiGe","Tunable SQUID from buried Ge well and 1.9-T superconductor"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000254,"raw_usage":{"total_tokens":1638,"prompt_tokens":1087,"completion_tokens":551,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":703,"completion_tokens_details":{"reasoning_tokens":447}},"tokens_in":703,"tokens_out":551,"duration_ms":5506,"temperature":1.0,"reasoning_tokens":447,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T14:05:19.462192+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the local density of states in the quantum well under a Josephson junction by tunneling spectroscopy: if no induced superconducting gap appears while the switching current persists, or if a fully depleted well still conducts supercurrent, the proximitisation claim is refuted.","supporting_citations":[{"cited_title":"Tosato, V","cited_arxiv_id":null,"evidence_quote":"Supplies the earlier PtSiGe-based proximitisation of germanium and the hard-gap measurement that this work extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports epitaxial tantalum germanide with an out-of-plane critical field of 1.88 T and critical temperature near 2 K, the comparator for a high-field germanium-compatible superconductor."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Systematically characterises superconducting platinum-group germanide and germanosilicide films, including IrSiGe with critical fields up to 2.8 T, the property target motivating PtIrSiGe."},{"cited_title":"Morgan, R","cited_arxiv_id":null,"evidence_quote":"Explains that platinum silicide forms by inward platinum diffusion while iridium silicide forms by outward silicon migration, the mechanistic basis for the platinum underlayer."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows that platinum incorporation suppresses voiding and oxidation in germanosilicide formation, supporting the alloy design used here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the asymmetric-SQUID interference formula used to fit the measured SQUID oscillations and extract the effective loop area."}],"review_version":1}