{"id":"d89ddb9b-ed44-4db2-8968-240bb273f29a","arxiv_id":"2412.02822","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"The superconducting dome in electron-doped MoS2 is recreated from first principles and traced to the 1x1 H to 2x2 charge-density-wave transition and later structural phases.","lead":"Computer simulations trace the rise and fall of superconductivity in a single layer of molybdenum disulfide to changes in the crystal structure itself. The result offers a structural explanation for a well-known experimental dome that earlier theories blamed on disorder or electron interactions.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 800 K DFT smearing chosen for the phase diagram leaves the claimed match with the experimental dome peak unverified, since nc shifts from 1.5 to 2.4e14 cm^-2 between 100 K and 800 K.","rationale":"After reading the paper in good faith, I find the mechanism credible and the calculations extensive. The identified soft A1 phonon at M, its backfolding into a 2x2 CDW, the 1T' phase, and the nonadiabatic reduction of lambda are internally consistent and mutually supporting. The TBA model is clearly labeled as qualitative, and no circular fitting to the experimental dome is apparent. However, the central claim is phrased as a reconciliation with experiment, and the quantitative anchor of that reconciliation is the doping at which the H phase softens and the CDW grows. That anchor is the most sensitive input: the paper itself reports a 60% swing in nc when the DFT Fermi-Dirac temperature is reduced from 800 K to 100 K, and the 800 K value is chosen for computational convenience rather than because it is physically representative. The experimental peak sits at the low-T end of this range, so the agreement displayed in Fig. 5(b) depends on the high-T choice. This is exactly the kind of self-identified limitation that should be resolved before accepting the quantitative claim. It does not overturn the qualitative structural mechanism, so the verdict should remain conditional. The reader identified the same weakest assumption, and I agree. The most effective check is a direct low-T recomputation of the phase boundaries and dome position, which is expensive but feasible with current methods.","tokens_in":20063,"tokens_out":7200,"duration_ms":79741,"concrete_test":"Re-run the 1x1 H M-point A1 phonon dispersion and the 2x2 CDW relative stability as a function of doping at T_scf = 100 K (and, if feasible, extrapolated 0 K) with k-grids converged so that nc is stable to <0.05e14 cm^-2, e.g., 48x48x1 for the 1x1 cell and equivalent 2x2 sampling. Regenerate Fig. 5(b). If the H instability and the H/2x2 CDW coexistence shift to about 1.5e14 cm^-2 while preserving the phase sequence, the central claim is confirmed; if the boundaries remain near 2.4e14 cm^-2 or the phase sequence changes, the dome match is an artifact of 800 K smearing.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the superconducting dome is caused by the 1x1 H to 2x2 CDW structural transition, with Tc rising in the H phase and peaking at H/CDW coexistence. For this claim to be quantitatively anchored, the doping nc at which the H phase goes soft and the 2x2 CDW becomes stable must match the experimental dome maximum. The paper computes this boundary with T_scf = 800 K and reports nc = 2.1 to 2.38e14 cm^-2, while the experimental peak sits near 1.5e14 cm^-2. The authors themselves state that lowering T_scf from 800 K to 100 K moves nc from 1.5 to 2.4e14 cm^-2 (Results and Supplementary Figure 1). Thus the plotted dome in Fig. 5(b) is anchored at the high-temperature end of a factor-of-1.6 uncertainty; the claimed reconciliation would shift by roughly 0.6 to 0.9e14 cm^-2 if the low-temperature boundary is physical. This is a verification gap rather than a contradiction, because the low-T shift is toward the experimental value, but the central quantitative comparison in the abstract is not established until the low-T (or 0 K) phase boundary is computed with converged k-grids.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript presents first-principles DFT/DFPT/EPW calculations combined with a large-supercell tight-binding model to explain the superconducting dome in electron-doped monolayer MoS2. The authors compute the doping-dependent stability of the 1x1 H, 2x2 CDW, and 1T' phases and find that Tc rises with doping in the H phase, peaks near the H/2x2-CDW coexistence region, and decreases once the 1T' phase and high-doping CDW phases stabilize, producing a dome. A tight-binding model on an 18√3 x 18√3 supercell predicts polaronic distortions and partial CDWs in the intermediate doping regime. The calculated dome is compared with experimental Tc data from Refs. [3,36] and with Raman A1g phonon frequencies from Li-doped MoS2.","tokens_in":20324,"tokens_out":9146,"duration_ms":94183,"significance":"If the proposed mechanism holds, it would resolve a long-standing puzzle by attributing the superconducting dome in electron-doped MoS2 to a phonon-softening-induced structural transition, rather than to extrinsic disorder or to a doping-dependent Coulomb pseudopotential. The paper is significant because it connects CDW, polaronic, and superconducting instabilities in a single ab initio framework, and it explicitly treats nonadiabatic electron-phonon coupling and spin-orbit coupling, both of which are known to be important in TMDs. The manuscript is technically strong: it uses state-of-the-art methods, provides a comprehensive set of calculations across several phases, and is candid about its remaining approximations (anharmonicity, idealized gating geometry, and the electronic-temperature smearing). There is no circularity: the experimental dome is compared only after the calculations are completed, and the fixed parameters (mu* = 0.13, TB couplings) come from prior literature or previous DFPT fits.","major_comments":[{"comment":"The central quantitative comparison with the experimental dome is not established because the critical doping nc at which the H phase softens, and at which the calculated Tc peaks, depends strongly on the DFT electronic temperature T_scf. The paper reports nc = 2.1–2.38 × 10^14 cm^-2 for T_scf = 800 K but acknowledges (and Supplementary Figure 1 shows) that nc = 1.5 × 10^14 cm^-2 for T_scf = 100 K; the experimental dome maximum is near 1.5 × 10^14 cm^-2. Since the abstract and Discussion claim that the dome is 'successfully create[d]' and 'reconcile[s]' experimental observations, the use of the high-temperature boundary leaves the peak position shifted by roughly 0.6–0.9 × 10^14 cm^-2 (about 40%). The low-temperature (or converged zero-temperature) phase boundary should be computed and used for the comparison before the quantitative claim can be assessed.","section":"The low-doping 1x1 H phase and Discussion"},{"comment":"The predicted peak Tc is a factor of 2–3 larger than the measured values: the authors obtain 29.9 K (21.6 K with SOC) at n = 2.04 × 10^14 cm^-2, whereas the experimental maxima are 10.9 K and 11.6 K. The authors attribute the discrepancy to anharmonicity and to the difference between uniform doping and a gating geometry, but these corrections are not computed in this work. Because the paper's stated goal is to 'recreate' the experimental dome, the large absolute overshoot means that the dome is reproduced only in shape, not in magnitude. This should be stated explicitly in the abstract if no further calculations are added.","section":"Discussion"}],"minor_comments":[{"comment":"The abstract says the work 'successfully recreat[es] the so far unresolved superconducting dome'; in view of the quantitative discrepancies discussed in the main text, a softer formulation such as 'reproducing the qualitative dome shape' would be more accurate.","section":"Abstract"},{"comment":"The text states that the H-to-1T' transition has a 'large energy barrier of around 10 eV'. This value is not justified; typical barriers in TMDs are on the order of 1 eV per formula unit. Please specify the supercell size and whether the barrier is per formula unit, or correct the value if it is a typographical error.","section":"Results, The 1T' phase"},{"comment":"Figure 5(a) compares theoretical phonon frequencies with data from Li-doped MoS2 using an upper axis in atomic ratio x, but the conversion from the theoretical electron density n to x is not stated. Please provide the conversion or explicitly mark the upper axis as schematic.","section":"Discussion, Fig. 5(a)"},{"comment":"The paper uses the term 'coexistence' for the 1x1 H and 2x2 CDW phases at the same doping, but does not define the thermodynamic condition (e.g., equal chemical potentials). Clarify whether 'coexistence' means that both structures are dynamically stable at the same doping or that they are true coexisting equilibrium phases.","section":"Results, Coexistence of 2x2 CDW and 1x1 H phases"},{"comment":"In the nonadiabatic phonon self-energy expression, Eq. (12), the static term (second term) is subtracted at the unrenormalized frequencies; the notation for the Fermi factors f_nk and f_mk+q could be clarified by explicitly defining the band indices m,n in the text.","section":"Methods, Eq. (12)"}],"recommendation":"major_revision","confidential_remarks":"This is a well-executed computational study with a plausible and novel mechanism for the superconducting dome in MoS2. The main obstacle is the sensitivity of the predicted dome peak to the DFT electronic temperature: the 800 K used in the calculations shifts the peak by ~40% relative to the experimental position. This is a verification gap rather than a fundamental error, and it can be addressed by additional calculations at lower T_scf with converged k-grids. The quantitative Tc overestimation is acknowledged by the authors and is likely correctable by including anharmonicity and realistic gating geometry, but those corrections are not yet in the manuscript. I recommend major revision with the expectation that the low-temperature phase boundary and possibly anharmonic/gating corrections will be added."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is the first attempt I know of to build the whole doping phase diagram of monolayer MoS2 from DFT and use it to explain the SC dome, and it is largely credible. But the abstract's claim that it \"recreates\" and \"reconciles\" the experimental dome goes beyond what the calculations actually pin down, because the critical doping depends strongly on a computational choice.\n\nWhat is genuinely good: they compute dynamically stable phases—1x1 H, 2x2 CDW, 1T'—across doping; they show the 2x2 CDW comes from three M-point instabilities, matching STM; they compute Tc with nonadiabatic Migdal-Eliashberg and show the H-phase Tc rises monotonically, peaks near H/CDW coexistence, then falls as CDW and 1T' stabilize. The TB model for polaronic distortions and partial CDWs is a reasonable extension, and they are upfront that it is qualitative. No circularity: mu*=0.13 and the TB parameters come from earlier work, not from fitting the measured dome.\n\nThe main soft spot is exactly what the stress-test note says. They choose T_scf = 800 K for the phase diagram, which moves nc from 1.5 (at 100 K) to 2.4 x 10^14 cm^-2 (at 800 K). The experimental peak is near 1.5. So the plotted dome is anchored at the high-temperature end of a factor-of-1.6 range. The low-T end goes in the right direction, which suggests the mechanism is not wrong, but the claimed quantitative match is not established until they either converge the low-T boundary or explain why 800 K is physically appropriate. They also overestimate Tc by 2-3x (34.6 K vs 10-11 K), and while they mention this, it weakens \"recreates.\" The intermediate-doping downslope relies partly on the TB model with fitted nearest-neighbor EPC; that part is illustrative, not first-principles.\n\nMinor: the \"10 eV\" barrier for H to 1T' is almost certainly a typo (should be ~1 eV or meV); easy fix.\n\nBottom line: the qualitative mechanism—soft phonon drives Tc up, structural transitions and mode hardening drive it down—is well supported and connects to a broader class of materials (SrTiO3, TMDs). The paper deserves a serious referee and would be a solid contribution after the T_scf question is addressed and the claims are softened. I'd send it to review, with a request to recompute the phase boundary at lower T_scf or at least present the 100 K result as the primary one, and to temper the abstract.\n\nWho it's for: anyone working on 2D superconductivity or soft-phonon mechanisms; a useful reference even in current form.","headline":"A credible first-principles mechanism for the MoS2 superconducting dome via structural phase transitions, but the quantitative match to experiment is not yet pinned down because the critical doping shifts with the DFT electronic temperature.","tokens_in":20914,"tokens_out":3578,"would_cite":true,"duration_ms":32986,"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":"The superconducting dome in electron-doped monolayer MoS2 is caused by a doping-driven lattice transition, not by pairing physics in a fixed structure.","keywords":["MoS2 monolayer","superconducting dome","charge density wave","electron-phonon coupling","soft phonon mode","nonadiabatic effects","phase transition","first-principles calculation"],"falsifier":"Measure the phonon dispersion of a gated monolayer MoS2 around $2\\times 10^{14}\\,\\mathrm{cm}^{-2}$ with inelastic scattering: the M-point acoustic phonon should soften and then stiffen just as the $2\\times2$ CDW superlattice appears, and the $T_c$ peak should coincide with that stiffening. If the soft mode appears at a clearly different doping, or if the lattice stays $1\\times1$ H while $T_c$ falls, the structural explanation fails.","tokens_in":19805,"feed_emoji":"❄️","tokens_out":8324,"duration_ms":81470,"temperature":0.7,"pith_summary":"This paper sets out to explain the superconducting dome in electron-doped monolayer MoS2, the rise and fall of the critical temperature $T_c$ with added electrons, which experiments have seen but theory has not reproduced. The authors argue that the dome is a structural effect: within the undistorted $1\\times1$ H phase $T_c$ rises monotonically, peaks near the doping at which a $2\\times2$ charge-density-wave (CDW) reconstruction becomes stable, and then falls as the CDW, polaronic distortions, and the $1T'$ phase take over and harden the phonon modes that mediate pairing. Their first-principles calculations trace the rise to a strongly coupled acoustic phonon that softens with doping, and the fall to the same mode stiffening in the new phases. If correct, the result turns a puzzling dome shape into a phase-transition signature and should apply to other doped transition-metal dichalcogenides with competing CDW order.","feed_headline":"Superconducting dome in MoS2 traced to a lattice switch","feed_subtitle":"Tc rises until a 2x2 charge-density-wave transition hardens the pairing phonon and pushes it back down.","key_machinery":"The load-bearing object is the $A_1$ acoustic phonon at the M point of the $1\\times1$ Brillouin zone, a motion of molybdenum atoms toward sulfur atoms that couples strongly to the conduction electrons. Its frequency softens with doping and it dominates the Eliashberg spectral function; nonadiabatic phonon self-energy corrections renormalize and broaden it, roughly halving the electron-phonon coupling constant. The same mode, backfolded to the $\\Gamma$ point of the $2\\times2$ cell (the $M^*$ point), hardens when the CDW or $1T'$ structure stabilizes, providing the single mechanism that raises and then lowers $T_c$. A tight-binding model on an $18\\sqrt{3}\\times 18\\sqrt{3}$ supercell extends the argument to intermediate dopings, where localized polaronic distortions and partial CDWs appear before the full CDW.","core_discovery":"The central discovery is that the experimentally observed superconducting dome in electron-doped MoS2 monolayer can be recreated from first principles only when the doping-induced structural phase transitions are included. The paper finds a sequence of dynamically stable phases: the $1\\times1$ H phase up to roughly $2.1\\text{--}2.4\\times 10^{14}\\,\\mathrm{cm}^{-2}$, a coexisting $2\\times2$ CDW phase with triangular Mo displacements in a narrow window, a metastable $1T'$ phase separated by a large barrier, and a high-doping $2\\times2$ CDW with a $2\\sqrt{3}\\times 2\\sqrt{3}$ modulation. $T_c$ rises in the H phase, reaches a maximum of about 34.6 K near the H/$2\\times2$ CDW coexistence (a representative value at $2.04\\times 10^{14}\\,\\mathrm{cm}^{-2}$ is 29.9 K, reduced to 21.6 K with spin-orbit coupling), and decreases in the CDW and $1T'$ phases because the relevant phonon mode hardens and the Fermi-level density of states drops. The computed $T_c$ values overestimate the measured maxima of about 10.9–11.6 K, which the paper attributes to anharmonicity and the difference between a jellium background and a realistic gate. The paper concludes that the $1\\times1$ H to $2\\times2$ CDW transition is the leading origin of the dome.","pith_inferences":["The paper leaves implicit that the dome peak in other TMDs such as WS2, TiSe2, and WTe2 should sit at the doping where their own soft phonon goes unstable, making a dome shape a fingerprint of a nearby CDW or ferroelectric instability.","The reported sensitivity to the 800 K electronic smearing suggests that using lower smearing would shift the theoretical phase boundaries toward the experimental maximum near $1.5\\times 10^{14}\\,\\mathrm{cm}^{-2}$ while keeping the structural mechanism intact.","A testable extension is strain or substrate engineering: biaxial strain changes the CDW instability doping and should shift the $T_c$ dome peak in the direction predicted by the softening mode.","The intermediate polaronic and partial-CDW regime implies that transport and Raman anomalies seen at moderate doping in gated MoS2 may be intrinsic precursors of the CDW rather than extrinsic disorder."],"forward_implications":["In the $1\\times1$ H phase, $T_c$ rises monotonically with doping, so the falling side of the dome is necessarily a structural effect rather than an electronic pairing effect.","The $2\\times2$ CDW with triangular Mo displacements and the high-doping $2\\sqrt{3}\\times 2\\sqrt{3}$ CDW should appear at the doping windows predicted here, in line with STM observations.","The metastable $1T'$ phase, with its large roughly 10 eV barrier from H, explains why gated samples do not show the $1T'$ transition while intercalated samples can.","Nonadiabatic corrections halve the electron-phonon coupling constant, and spin-orbit coupling reduces $T_c$ by about 30% at the highest dopings, so quantitative agreement with experiments requires both ingredients.","Between the H and full-CDW regions, polaronic distortions and partial CDWs suppress the Fermi-level density of states and contribute to the reduction of $T_c$."],"supporting_citations":[{"why":"Provides the experimental superconducting dome in gate-tuned MoS2 that the paper aims to reproduce.","marker":"[3]"},{"why":"Provides tunnelling-spectroscopy measurements of the dome used as the primary experimental reference for $T_c$.","marker":"[36]"},{"why":"Reports the $2\\times2$ and $2\\sqrt{3}\\times 2\\sqrt{3}$ CDW phases with triangular Mo distortions identified by STM, the structural feature the paper explains.","marker":"[14]"},{"why":"Earlier phase-diagram study that predicted CDW instabilities in doped dichalcogenides and whose treatment the paper extends beyond the $1\\times1$ H phase.","marker":"[26]"},{"why":"Field-effect-doped calculation that reproduces $T_c$ values but finds no dome; the baseline the paper's structural mechanism must beat.","marker":"[62]"},{"why":"Supplies the nonadiabatic electron-phonon coupling formalism used to renormalize $T_c$ downward.","marker":"[63]"},{"why":"Establishes nonadiabatic phonon renormalization in MoS2 and WS2 monolayers, central to the computed phonon frequencies.","marker":"[65]"},{"why":"Provides the tight-binding electron-lattice downfolding model used for large-supercell polaronic and partial-CDW calculations.","marker":"[70]"}],"fun_headline_variants":["Doping's double act: MoS2 superconductivity dome explained","Lattice switch behind MoS2 superconducting dome","How a 2x2 CDW caps MoS2's Tc dome","From H to CDW: the origin of MoS2's superconducting dome","Superconductivity dome in MoS2: a structural phase transition story"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation uses a fictitious electronic temperature of 800 K to smear the Fermi surface, and the doping at which the H lattice softens moves from about $1.5\\times 10^{14}\\,\\mathrm{cm}^{-2}$ to $2.4\\times 10^{14}\\,\\mathrm{cm}^{-2}$ when that temperature is lowered to 100 K, so the position of the theoretical dome peak depends on this numerical setting.","fun_headline_variants_meta":{"raw":{"variants":["Doping's double act: MoS2 superconductivity dome explained","Lattice switch behind MoS2 superconducting dome","How a 2x2 CDW caps MoS2's Tc dome","From H to CDW: the origin of MoS2's superconducting dome","Superconductivity dome in MoS2: a structural phase transition story"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000359,"raw_usage":{"total_tokens":1964,"prompt_tokens":990,"completion_tokens":974,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":606,"completion_tokens_details":{"reasoning_tokens":882}},"tokens_in":606,"tokens_out":974,"duration_ms":7510,"temperature":1.0,"reasoning_tokens":882,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T23:05:00.634735+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the phonon dispersion of a gated monolayer MoS2 around $2\\times 10^{14}\\,\\mathrm{cm}^{-2}$ with inelastic scattering: the M-point acoustic phonon should soften and then stiffen just as the $2\\times2$ CDW superlattice appears, and the $T_c$ peak should coincide with that stiffening. If the soft mode appears at a clearly different doping, or if the lattice stays $1\\times1$ H while $T_c$ falls, the structural explanation fails.","supporting_citations":[{"cited_title":"Charge density waves in electron-doped molybdenum disulfide","cited_arxiv_id":"2108.07015","evidence_quote":"Reports the $2\\times2$ and $2\\sqrt{3}\\times 2\\sqrt{3}$ CDW phases with triangular Mo distortions identified by STM, the structural feature the paper explains."},{"cited_title":"Phase Diagram of Electron Doped Dichalcogenides","cited_arxiv_id":"1404.4295","evidence_quote":"Earlier phase-diagram study that predicted CDW instabilities in doped dichalcogenides and whose treatment the paper extends beyond the $1\\times1$ H phase."},{"cited_title":"Anisotropic Migdal-Eliashberg theory using Wannier functions","cited_arxiv_id":"1211.3345","evidence_quote":"Field-effect-doped calculation that reproduces $T_c$ values but finds no dome; the baseline the paper's structural mechanism must beat."},{"cited_title":"Competition between phonon mediated superconductivity and carriers localization in field-effect doped molybdenum dichalcogenides","cited_arxiv_id":"2209.01973","evidence_quote":"Supplies the nonadiabatic electron-phonon coupling formalism used to renormalize $T_c$ downward."},{"cited_title":"Phonon Self-Energy Corrections: To Screen, or Not to Screen","cited_arxiv_id":"2212.11806","evidence_quote":"Establishes nonadiabatic phonon renormalization in MoS2 and WS2 monolayers, central to the computed phonon frequencies."},{"cited_title":"Zheliuk, J","cited_arxiv_id":null,"evidence_quote":"Provides the tight-binding electron-lattice downfolding model used for large-supercell polaronic and partial-CDW calculations."}],"review_version":1}