{"id":"2101abf5-ccbd-460a-a01f-8bc64b61ebdc","arxiv_id":"2608.08500","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Fully epitaxial Ga:Ge/Si/Ga:Ge Josephson junctions show strong coupling and a magnetic-field-enhanced switching current, attributed to quasiparticle-assisted thermalization from the Al leads.","lead":"This paper demonstrates Josephson junctions made from a superconductor based on gallium-doped germanium, with an ultra-thin silicon barrier, grown entirely in one epitaxial process. The devices show strong superconducting coupling and a magnetic-field-induced boost in switching current that the authors trace to quasiparticle cooling from aluminum contacts.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central attribution of the switching-current enhancement to quasiparticle-assisted thermalization is underdetermined: no quantitative model links the field-suppressed Al gap to a cooling channel, and the premise that the top Ga:Ge layer relies on Al leads for heat removal is asserted rather…","rationale":"The device demonstration itself is plausible: the epitaxial Ga:Ge/Si/Ga:Ge stack, the area scaling of ISW and RN, and the existence of a supercurrent branch are internally coherent. The reader's conditional verdict is appropriate because the novel mechanistic claim—quasiparticle-assisted thermalization—is supported by qualitative reasoning and a single-field SNS fit, not by a quantitative prediction of the field dependence. My stress-test identifies the same load-bearing assumption: the cooling channel through Al leads must dominate over other thermal paths and over the field-induced suppression of the critical current, yet no independent thermometry or control experiment is provided. The proposed test—measuring the electron temperature independently at zero and finite in-plane field—would settle whether the enhancement is genuinely thermal in origin. Since this does not invalidate the underlying device result but does weaken the central interpretation, the verdict should remain CONDITIONAL.","tokens_in":7706,"tokens_out":10444,"duration_ms":126969,"concrete_test":"Decisive check: on Device A (or a nominally identical device), measure the full switching-current distribution P(ISW) and the dissipative-branch quasiparticle slope at base temperature for B∥=0 and 80 mT, and independently extract the electron temperature from the high-bias conductance (or Johnson noise thermometry). If the extracted electron temperature at 80 mT is not lower than at 0 mT, the QP-thermalization mechanism is falsified, because the ISW enhancement cannot be attributed to cooling. This single measurement tests the causal link directly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing concern is with the central interpretation in the section 'Magnetic field enhancement of the switching current and thermalization process.' The paper states: 'We therefore attribute the ISW-enhancement to enhanced thermalization of the junction by QPs' and specifically assumes that the top Ga:Ge layer 'relies on the Al leads as thermalization path.' For this to be true, the top layer must be thermally isolated from the substrate, and the field-induced quasiparticles in Al must cool the junction more than the field suppresses its critical current. No quantitative model of I_SW(B) is provided; the only quantitative support is a fit of I_SW(T) at a single field (B∥=160 mT) to a disordered short ballistic SNS model with transparency τ=0.8 (Fig. 4a). That fit uses the switching current as a proxy for the critical current and effectively adjusts the electron temperature; it cannot distinguish cooling from other field-dependent changes in junction dynamics, such as increased quasiparticle conductance or damping that moves I_SW closer to I_C. The exclusions of magnetic impurities, π-junction, and vortex mechanisms are negative arguments and do not positively establish the QP-cooling channel. Because the paper's headline phenomenon and its proposed cause rest on this underdetermined link, the central claim is not yet supported at the level claimed in the abstract.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports fully epitaxial Josephson junctions grown by molecular beam epitaxy, consisting of a 0.5 nm Si barrier sandwiched between two Ga-doped Ge superconducting layers, with Al contacts. Six devices with different junction areas are characterized by four-terminal DC transport. The junctions show a clear supercurrent, and the product eR_N I_SW/Delta lies between the Ambegaokar-Baratoff and short-ballistic limits. The central new observation is an enhancement of the switching current at in-plane magnetic fields around 80 mT, together with the emergence of hysteresis at low temperature. The authors attribute this enhancement to quasiparticle-assisted thermalization from the Al contacts, supported by temperature-dependent measurements at zero and 160 mT and a fit to a disordered short ballistic SNS model. The paper concludes with a discussion of the platform's potential for merged-element transmon qubits.","tokens_in":7938,"tokens_out":3488,"duration_ms":40865,"significance":"If the interpretation is correct, the work demonstrates a structurally clean, fully epitaxial, CMOS-compatible Josephson junction platform with strong coupling, which is a meaningful advance for low-disorder superconducting quantum devices. The strengths are the in-situ MBE growth, atomic-scale interface characterization by STEM, the systematic scaling of I_SW and R_N with junction area across six devices, and the clear observation of a reproducible magnetic-field-induced enhancement. The main significance is currently limited by the fact that the headline physical mechanism is supported only qualitatively; the platform demonstration itself is solid and would remain valuable even if the proposed thermalization mechanism were softened.","major_comments":[{"comment":"The central attribution of the switching-current enhancement to quasiparticle-assisted thermalization is underdetermined. The paper states 'We therefore attribute the I_SW-enhancement to enhanced thermalization of the junction by QPs' and assumes that the top Ga:Ge layer 'relies on the Al leads as thermalization path,' but no quantitative model of I_SW(B) is provided, and the thermal isolation of the top layer from the substrate is asserted rather than demonstrated. A fit of I_SW(T) at a single field value (B_parallel = 160 mT) cannot distinguish a genuine cooling channel from other field-dependent effects, such as an increase in quasiparticle conductance or a change in the junction's dynamical damping that moves the switching current closer to the true critical current. A quantitative estimate of the QP cooling power relative to the electron-phonon coupling, and a predicted field dependence that can be compared with Fig. 4b, would be needed to support the central claim.","section":"Magnetic field enhancement of the switching current and thermalization process (Fig. 3, Fig. 4a)"},{"comment":"The quantitative support for the QP-cooling interpretation is a fit of I_SW(T) at B_parallel = 160 mT to a disordered short ballistic SNS model with transparency tau = 0.8, and the superconducting gap is estimated from T_C = 365 mK 'as extracted from the fit in FIG 4a.' Because the switching current is used as a proxy for the critical current, and because tau and T_C are extracted from the same trace that is then used to infer a reduced effective electron temperature, the fit is a circular consistency check rather than an independent test of the cooling scenario. I recommend either obtaining tau and T_C from independent measurements (e.g., normal-state characterization or tunnel-junction spectroscopy) or explicitly constructing the model so that the switching-current dynamics and the field-dependent quasiparticle distribution are treated separately.","section":"Fig. 4a and Table 1"},{"comment":"The arguments against magnetic impurities, vortices, and pi-junction behavior are negative: no magnetic impurities are known, no Fraunhofer pattern is observed, and no field hysteresis is seen. The absence of these signatures does not positively establish that field-generated quasiparticles in the Al leads cool the junction. The paper should provide a falsifiable prediction of the QP-assisted thermalization model—for example, the expected field scale and magnitude of the enhancement, its scaling with contact area and lead volume, and the temperature at which the enhancement should vanish—and compare that prediction with the data in Fig. 4b. Without such a quantitative link, the abstract's claim that the enhancement is 'attributed to quasiparticle-assisted thermalization processes' overstates the evidence.","section":"Magnetic field enhancement: exclusion of alternative mechanisms"}],"minor_comments":[{"comment":"The caption defines the shaded regions as non-hysteretic, transition, and hysteretic, but the exact field values delimiting these regions are not stated; please provide the numerical boundaries.","section":"Fig. 3c"},{"comment":"The legend text appears truncated ('ISW IR'); the curves should be labeled explicitly as I_SW and I_R, and the symbols for the data points should be defined.","section":"Fig. 4a"},{"comment":"Several claims refer to supplementary sections (SI IV, SI V, SI VI, SI IX) that are not included in the manuscript; please ensure these are available and that each referenced section is clearly linked.","section":"Supplementary references"},{"comment":"The phrase 'Cl-terminate the Ga:Ge films in HCl' is used without further explanation; a brief description or a reference for the surface passivation procedure would improve reproducibility.","section":"Introduction"}],"recommendation":"major_revision","confidential_remarks":"The core device demonstration appears sound and the field-enhancement phenomenon is interesting, but the proposed thermalization mechanism is not yet established at the level claimed. I am not asking for new fabrication; rather, the authors should either provide a quantitative thermal model with testable predictions or substantially soften the interpretation in the abstract and conclusion. The work fits the journal's scope, and the platform itself may be publishable even if the enhancement mechanism remains open."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear Colleague,\n\nThe thing to know: this is a genuine new device result—the first vertical Ga:Ge/Si/Ga:Ge Josephson junctions, fully epitaxial, with STEM-verified interfaces and transport that indicates strong Josephson coupling. The field-enhanced switching current is a real observation across all six devices. But the mechanism the paper proposes for that enhancement, quasiparticle-assisted thermalization from the Al contacts, is plausible yet underdetermined, and the abstract and conclusion present it as more established than the evidence warrants.\n\nWhat the paper does well: the device engineering is careful. Area scaling of ISW and RN is clean; eRN ISW/Delta sits between the Ambegaokar-Baratoff and short-ballistic limits; the exclusions of magnetic impurities, pi-junction, and vortex mechanisms are reasonable negative arguments. The STEM/EDS evidence that the structure survives processing is solid. This is a credible new platform with CMOS compatibility, building properly on refs 14 and 15.\n\nWhere it's soft: the central interpretation rests on a qualitative picture plus a single SNS fit at B||=160 mT. The fit takes the switching current as a proxy for the critical current and effectively adjusts an electron temperature; it cannot distinguish cooling from other field-dependent dynamics, such as increased damping, reduced noise, or a shift of ISW toward the true IC. The premise that the top Ga:Ge layer 'relies on the Al leads as thermalization path' is asserted, not verified by any thermal measurement or simulation. There is no quantitative model of ISW(B) linking the Al gap suppression to a cooling power. So the mechanism is a hypothesis with supporting arguments, not a demonstrated conclusion. The conclusion also overreaches on qubit-readiness: no coherence or TLS data exist yet, and 'low disorder' is an inference, not a measurement.\n\nThe circularity worry is minor: using the fit to extract Delta and tau and then using those to normalize the eRN ISW/Delta values is a bit self-referential, but standard practice in transport papers and not a reason to doubt the device result.\n\nBottom line: the platform deserves attention and the observation is real; the interpretation needs either a quantitative thermal model or a softer framing. I would send this to a serious referee, with the expectation that the abstract must be rebalanced to separate observation from interpretation.","headline":"New, credible epitaxial group-IV Josephson junction platform, but the quasiparticle-cooling explanation for the field-enhanced switching current is underdetermined and oversold in the abstract.","tokens_in":8543,"tokens_out":4578,"would_cite":true,"duration_ms":48263,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["74.50.+r"],"model":"deepseek-v4-flash","headline":"This paper demonstrates a fully epitaxial gallium-doped germanium / silicon / gallium-doped germanium Josephson junction platform, and attributes an unconventional magnetic-field rise in switching current to quasiparticle-assisted cooling…","keywords":["Josephson junctions","superconductivity","gallium-doped germanium","crystalline semiconductor","molecular beam epitaxy","switching current","quasiparticle thermalization","quantum computing"],"falsifier":"A direct test would be to add a tunnel-junction thermometer to the top Ga:Ge layer and measure its electronic temperature versus $B_\\parallel$: the quasiparticle-assisted-thermalization explanation predicts a correlated drop in electron temperature as $I_{SW}$ rises, plus the disappearance of the enhancement when the aluminum leads are exchanged for a superconductor with a much higher critical field.","tokens_in":7494,"feed_emoji":"⚛️","tokens_out":12599,"duration_ms":117542,"temperature":0.7,"pith_summary":"Ordinary Josephson junctions rely on an amorphous aluminum-oxide barrier whose atomic disorder is thought to host two-level-system defects that shorten qubit coherence. This paper claims that a fully crystalline alternative can be grown in one molecular-beam-epitaxy run: a superconducting gallium-doped germanium layer, a half-nanometer silicon weak link, and another gallium-doped germanium layer, all with atomically sharp interfaces. In six junctions with different areas the authors find strong Josephson coupling, with $eR_N I_{SW}/\\Delta$ near the ideal tunnel value and below the ballistic short-junction limit, and $I_{SW}$ scaling with area while $R_N$ scales inversely. The paper's central experimental finding is a rise in switching current under an in-plane magnetic field, opposite to the usual suppression, which the authors attribute to quasiparticles generated in the aluminum contacts that open a cooling channel for the top Ga:Ge layer. The stack thereby offers a low-disorder, CMOS-compatible route to merged-element transmon qubits.","feed_headline":"Gallium-doped germanium makes atomically sharp Josephson junctions","feed_subtitle":"Fully epitaxial junctions keep the barrier crystalline and point toward low-defect, CMOS-compatible qubits.","key_machinery":"The load-bearing object is the vertical crystalline stack Ga:Ge/Si/Ga:Ge: two 10 nm superconducting Ga:Ge films separated by a coherently strained 0.5 nm Si barrier, grown by molecular-beam epitaxy with no post-growth junction definition. The measurable that carries the coupling argument is the dimensionless product $eR_N I_{SW}/\\Delta$, which separates tunnel junctions (with value $\\pi/2$) from fully transparent ballistic superconductor–normal–superconductor (SNS) junctions (with value $\\pi$); the six devices scatter between the two, placing them in the short, moderately transparent SNS regime. The magnetic-field effect is carried by a two-channel thermalization picture: at zero field the top Ga:Ge layer cools mainly through electron-phonon coupling, while an in-plane field creates in-gap quasiparticles in the aluminum contacts and opens a second cooling path, so that $I_{SW}$ can rise with field until the aluminum critical field cuts the channel off.","core_discovery":"On its own terms, the central claim is that heavily gallium-doped germanium, made superconducting by extreme substitutional doping, can serve as both the electrodes and the crystalline environment of a Josephson junction, with a 0.5 nm silicon barrier acting as the weak link. The paper reports that the Ga:Ge/Si/Ga:Ge tri-layer, grown entirely in situ, survives device fabrication with coherent lattice contrast across the junction, and that all six measured devices show a supercurrent branch with $eR_N I_{SW}/\\Delta$ clustering around the ideal tunnel-junction value $\\pi/2$ and below the ballistic short-junction value $\\pi$. It then reports that an in-plane magnetic field around 80 mT enhances the switching current, accompanied by growing hysteresis, while a perpendicular field only suppresses it. After excluding magnetic impurities, vortices, and $\\pi$-junction behavior, the authors conclude that field-generated in-gap quasiparticles in the aluminum leads enhance thermalization of the top Ga:Ge layer, an interpretation supported by the temperature dependence: at $B_\\parallel = 160$ mT the switching current continues to rise down to base temperature, and a disordered short ballistic superconductor–normal–superconductor model with transparency $\\tau=0.8$ reproduces the data.","pith_inferences":["If the quasiparticle-cooling interpretation is right, the switching current itself becomes a built-in probe of non-equilibrium quasiparticle density in the aluminum leads, since any change in cooling rate would appear as a shift in $I_{SW}$ at fixed field.","The field-enhanced regime could be deliberately used in qubit operation: parking a transmon near the in-plane field that maximizes cooling would lower the effective electronic temperature at the cost of flux sensitivity, a trade-off the paper does not quantify.","A direct extension would be to replace the aluminum contacts with a higher-critical-field superconductor: the mechanism implies the enhancement peak should move to that material's critical field, offering a clean test of the thermalization channel."],"forward_implications":["Amorphous AlO$_x$ barriers can be replaced by a crystalline group-IV weak link grown in situ, eliminating the multi-step fabrication and interface contamination that usually compromise epitaxial junctions.","The vertical geometry and monolayer control of the silicon barrier thickness give direct tunability of the junction's electrical parameters, including the $E_J/E_C$ ratio needed for charge-noise-insensitive operation.","The reproducible rise of switching current with in-plane field, seen in all six devices and vanishing above about 250 mK, is a concrete signature of quasiparticle-assisted thermalization in this geometry.","Because the stack uses only CMOS-compatible germanium processing, the same growth scheme can be scaled to wafer-scale, densely integrated superconducting qubits with a small footprint and expected low two-level-system density."],"supporting_citations":[{"why":"Establishes the MBE-grown superconducting Ga:Ge platform and its structural characterization, the foundation of the junction stack.","marker":"[14]"},{"why":"Earlier report of superconductivity in heavily gallium-doped germanium grown by MBE, underpinning the material system.","marker":"[15]"},{"why":"Provides the ideal tunnel-junction value $eR_N I_{SW}/\\Delta = \\pi/2$ against which the measured coupling is benchmarked.","marker":"[25]"},{"why":"Gives the short ballistic SNS limit $eR_N I_{SW}/\\Delta = \\pi$, the upper comparison for the six devices.","marker":"[27]"},{"why":"Describes a vortex-based switching-current enhancement that the paper excludes by the absence of field hysteresis.","marker":"[30]"},{"why":"Reports quasiparticle-related switching-current enhancement in nanowire Josephson junctions, the class of mechanism invoked here.","marker":"[31]"},{"why":"Documents magnetic-field-induced quasiparticle thermalization effects used to support the Al-lead cooling channel.","marker":"[32]"},{"why":"Supplies the disordered short ballistic SNS model with transparency $\\tau=0.8$ used to fit the temperature dependence.","marker":"[33]"}],"fun_headline_variants":["Crystalline Ga:Ge junction shows field-enhanced supercurrent","Atomically sharp Josephson junction from doped germanium","Fully epitaxial crystalline junction for low-defect qubits","Crystalline group-IV junction shows unconventional field response","Ga-doped Ge yields crystalline Josephson junction with sharp interfaces"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central explanation assumes that the top Ga:Ge layer loses heat mainly through the aluminum contacts and that in-gap quasiparticles created by the magnetic field open a cooling channel strong enough to overcome the field's direct suppression of the supercurrent, an assumption not tested by a quantitative model.","fun_headline_variants_meta":{"raw":{"variants":["Crystalline Ga:Ge junction shows field-enhanced supercurrent","Atomically sharp Josephson junction from doped germanium","Fully epitaxial crystalline junction for low-defect qubits","Crystalline group-IV junction shows unconventional field response","Ga-doped Ge yields crystalline Josephson junction with sharp interfaces"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000844,"raw_usage":{"total_tokens":3688,"prompt_tokens":970,"completion_tokens":2718,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":586,"completion_tokens_details":{"reasoning_tokens":2639}},"tokens_in":586,"tokens_out":2718,"duration_ms":20487,"temperature":1.0,"reasoning_tokens":2639,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T04:33:22.729987+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct test would be to add a tunnel-junction thermometer to the top Ga:Ge layer and measure its electronic temperature versus $B_\\parallel$: the quasiparticle-assisted-thermalization explanation predicts a correlated drop in electron temperature as $I_{SW}$ rises, plus the disappearance of the enhancement when the aluminum leads are exchanged for a superconductor with a much higher critical field.","supporting_citations":[{"cited_title":"A.; Strohbeen, P","cited_arxiv_id":null,"evidence_quote":"Establishes the MBE-grown superconducting Ga:Ge platform and its structural characterization, the foundation of the junction stack."},{"cited_title":"J.; Brook, A","cited_arxiv_id":null,"evidence_quote":"Earlier report of superconductivity in heavily gallium-doped germanium grown by MBE, underpinning the material system."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the ideal tunnel-junction value $eR_N I_{SW}/\\Delta = \\pi/2$ against which the measured coupling is benchmarked."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the short ballistic SNS limit $eR_N I_{SW}/\\Delta = \\pi$, the upper comparison for the six devices."},{"cited_title":"Physical Review Letters2022,128, 207001","cited_arxiv_id":null,"evidence_quote":"Describes a vortex-based switching-current enhancement that the paper excludes by the absence of field hysteresis."},{"cited_title":"Physical Review B2020,102","cited_arxiv_id":null,"evidence_quote":"Reports quasiparticle-related switching-current enhancement in nanowire Josephson junctions, the class of mechanism invoked here."},{"cited_title":"A.; Zwanenburg, F","cited_arxiv_id":null,"evidence_quote":"Documents magnetic-field-induced quasiparticle thermalization effects used to support the Al-lead cooling channel."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the disordered short ballistic SNS model with transparency $\\tau=0.8$ used to fit the temperature dependence."}],"review_version":1}