{"id":"dffd40ac-217f-4966-8c0c-661d21c9c1f7","arxiv_id":"2507.01585","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"The paper demonstrates numerically that a tunable 3x3 dot array in graphene can be optimized to steer and split electron beams into target directions.","lead":"A theoretical study shows that a 3x3 array of graphene quantum dots with independently tunable voltages, optimized by an evolutionary algorithm, can redirect or split an incoming electron beam. The platform could serve as a reconfigurable electron-optic component for future graphene electronics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Plane-wave far-field optimization may not control electron flow in realistic focused-beam devices; the paper's own Discussion and SI concede this gap, and beam splitters are never validated in the near field.","rationale":"The reader's weakest assumption correctly identifies the gap between the single-valley plane-wave scattering model and a real device with focused beams, disorder, and intervalley scattering. I narrow this to the most load-bearing part: the optimization objective itself is the far-field scattered current of an infinite plane wave, and the paper concedes that realistic devices use focused beams and that near-field and far-field behavior can differ. This matters because the central claim is about controlling electron flow, not merely about computing an asymptotic cross-section. I did not choose angular-momentum truncation as the primary concern because the low-energy parameters (kR = 0.45, k d = 1) make truncation risks plausible but less central than the acknowledged transfer gap. I also did not make DE robustness the primary concern because the random-vs-DE comparison in Fig. 2 gives some support for the optimizer, whereas no evidence at all is provided for focused-beam performance. The paper is a reasonable computational proposal, so the verdict should remain conditional rather than being rejected: the missing focused-beam and near-field validation is exactly the condition that needs to be met before the platform claim is accepted. My agreement with the reader is partial because the reader framed the issue broadly around model realism, while my attack focuses on the specific mismatch between the optimized far-field observable and the finite, focused-beam setting the paper itself identifies as realistic.","tokens_in":15934,"tokens_out":20188,"duration_ms":228020,"concrete_test":"Take the optimized TDP configurations from Figs. 3(e) and 3(f) and rerun the scattering calculation with a finite-width focused incident beam, for example a Gaussian superposition of plane waves with transverse momentum spread delta_k/k of roughly 0.1 to 0.3, and compute the angular distribution of total current (incident plus scattered) on a detector circle at a finite radius such as r = 100. Measure the fraction of transmitted power entering each target angular range and compare with the far-field plane-wave split ratio. If the split ratio degrades by more than roughly 30% or the main lobes shift outside the target ranges, the plane-wave far-field optimization does not transfer to realistic electron-optic conditions, and the claim of precise current control would need to be restricted to the idealized plane-wave limit.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that a tunable 3x3 dot array can precisely control electron flow by steering or splitting an incident beam. What must be true for this claim is that optimizing the far-field scattered current for an infinite plane wave, via the fitness function in Eq. (5), also produces the intended current flow in a realistic device. That bridge is not established, and the authors themselves flag it. The Discussion states that realistic electron-optic devices will use focused beams rather than the plane wave considered throughout, and the Supporting Information warns that near-field and far-field patterns need not correspond and that far-field-optimized configurations may be unsuitable for probes placed near the scatterer. For the beam-splitter targets in Figs. 3(e) and 3(f), only far-field angular scattering is shown; the near-field maps in Fig. S3 show cases where the visible current flow does not match the far-field peaks. The near-field optimization in Fig. 4 addresses a single target angular range at one fixed radius, not the splitter geometries that are highlighted as a key functionality. Thus the designed configurations are optimal for an idealized asymptotic observable, but the paper's broader claim about controlling electron flow in graphene devices requires performance under finite illumination, finite detection distance, and interference between incident and scattered waves. Without quantitative evidence for a focused or finite-size beam, the central claim remains conditional on a transfer step that the text explicitly leaves open.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a tunable dot platform (TDP), a 3x3 array of electrostatically gated quantum dots in graphene with independently adjustable potentials. Using a single-valley Dirac model and a generalized Mie-scattering formalism, the authors compute angular scattering patterns and use differential evolution to optimize the dot potentials so that the far-field scattered current is concentrated in chosen angular windows. They demonstrate single-angle targeting (Fig. 3(a)-(d)), two-angle beam splitting (Fig. 3(e)-(f)), and one example of near-field current optimization (Fig. 4). The paper claims that the TDP enables precise control of electron flow and is a promising platform for electron-optics devices. The Supporting Information provides the multiple-scattering derivation, the DE implementation details, and additional near-field maps, and it explicitly notes that near-field and far-field behaviors do not always correspond.","tokens_in":16168,"tokens_out":6679,"duration_ms":69472,"significance":"The proposal is timely: it combines an experimentally accessible architecture (independently gated dots in graphene) with a numerical optimization scheme (differential evolution) to design electron-optical functionality, extending earlier work on Dirac Mie scattering and multi-dot arrays. The comparison against random search in Fig. 2 is a good control for the optimizer, and the near-field optimization in Fig. 4 shows that the method can be adapted to a local observable. If the designed configurations are validated under realistic illumination (focused beams, finite detector distance, disorder), the platform could enable reprogrammable beam deflectors and splitters. However, the current manuscript's central claim is stronger than what is demonstrated, because the optimization metric is the far-field angular current for a plane wave, and the paper's own Supporting Information and Discussion concede that this metric does not necessarily transfer to near-field or finite-beam settings. The manuscript also lacks mode-convergence and robustness tests that are essential for numerical credibility.","major_comments":[{"comment":"The central claim 'precise control of electron flow' is supported only for the far-field angular scattering of an infinite plane wave. The beam-splitter targets in Figs. 3(e)-(f) are optimized and evaluated exclusively through the far-field quantity in Eq. (5) / Eq. (S18). The near-field maps in Fig. S3 show cases where the real-space current flow does not match the far-field peaks, and the SI explicitly states that far-field-optimized configurations may not be suitable for probes near the scatterer. To make the stated claim, the authors should demonstrate that optimized configurations also produce the intended current flow for a finite-size or focused illumination and at finite detection distances, or at least qualify the claim accordingly.","section":"Optimization / Fig. S3"},{"comment":"The manuscript does not report the truncation order M of the angular momentum expansion used in the calculations, nor does it provide any mode-convergence test. Since the far-field current in Eq. (S14) and the near-field maps both rely on truncated sums over m, the absence of such a test leaves the numerical results unvalidated. Please add a convergence study (e.g., showing jsc,r(θ) and selected near-field quantities for increasing M) for representative TDP configurations.","section":"Supporting Information, Details of the scattering theory"},{"comment":"The fitness function in Eq. (5) is the same quantity used to report success: the integrated far-field current in the target angular window. The optimized patterns therefore match the target by construction, so the demonstration that peaks appear in the target range is essentially a check that the optimizer works. The comparison with random search is a good control for the optimizer, but it does not test the physical generalization. The paper should evaluate the optimized configurations with an independent metric, such as the current transmitted through a finite angular sector at a finite radius, or the overlap of the real-space current density with a desired beam profile, to justify the claim of controlling electron flow.","section":"Optimization, Eq. (5)"},{"comment":"The DE results appear to come from a single run with a fixed population size (np=30), mutation factor F=0.3, fitness weight W=0.4, and discrete potential step δV=0.1. There is no reproducibility test across independent DE runs, no sensitivity analysis with respect to these hyperparameters, and no robustness test of the optimized configurations against small perturbations of the dot potentials. Given the run-to-run variability typical of evolutionary algorithms, the claim that DE 'reliably' finds suitable configurations needs support from multiple restarts and from stability checks of the reported solutions.","section":"Optimization and SI (DE parameters)"}],"minor_comments":[{"comment":"The stated configuration space of 2^19 for 9 dots with 21 potential levels is incorrect: the space has 21^9 ≈ 7.9×10^11 configurations. Please clarify the counting or correct the number.","section":"Optimization"},{"comment":"The sentence 'This strong correspondence between near- and far-field scattering is not universal, and can be masked by interference with additional notes provided in the Supporting Information' is incomplete and should be rewritten.","section":"Main text after Fig. 2(b)"},{"comment":"Please report the actual value of M used in all calculations, in addition to stating that low-order modes suffice.","section":"Supporting Information"},{"comment":"The reference list is disordered and contains duplicate numbers; for example, refs. (7), (6), (1), and (3) appear more than once. This must be corrected.","section":"References"},{"comment":"Please justify why the first (reward) term in the near-field fitness uses ∆j(r,θ) while the second (penalty) term uses only its positive component ∆j+(r,θ); the asymmetry is not explained.","section":"Supporting Information, Eq. (S19)"},{"comment":"The sentence stating that realistic devices will use focused beams should be reconciled with the central claim; consider adding a brief remark in the conclusions about how the far-field-optimized configurations can be translated to focused-beam settings.","section":"Discussion"}],"recommendation":"major_revision","confidential_remarks":"The manuscript fits the scope of the journal. The main concern is the gap between the demonstrated far-field plane-wave optimization and the claimed device-level control; this is, in my view, fixable within the manuscript's scope by adding near-field/focused-beam validations and convergence tests. The paper would also benefit from a careful correction of the reference list and the configuration-space estimate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper shows a 3x3 array of independently gated dots in graphene can, in a single-valley Dirac model, be optimized by differential evolution to scatter an incident plane wave into specified angular windows, including a two-lobed splitter pattern. That's a genuinely new building block for graphene electron optics, and the authors deserve credit for framing it clearly and for comparing against random search to show the optimization is doing real work.\n\nThe scattering formalism is standard — generalized Mie theory for Dirac fermions, extended to multiple dots via Graf's addition theorem — and the implementation appears correct. The extension to near-field current optimization is a reasonable, thoughtful addition, and the authors are transparent that near-field and far-field behavior need not correspond. They even provide an example in the SI where a far-field peak does not match the near-field map.\n\nWhere it gets soft: the paper makes a broad claim about 'controlling electron flow in graphene,' but every target configuration is optimized for plane-wave far-field scattering. The Discussion concedes realistic devices will use focused beams, and the SI warns that far-field-optimized designs may be unsuitable for probes placed near the scatterer. The beam splitters in Fig. 3 are only validated in the far field; the near-field optimization in Fig. 4 addresses a single angular window at one fixed radius, not the splitter geometry. So the transfer from this idealization to a device is untested. That's not a fatal flaw for a proof-of-concept, but the text should be careful not to overstate it.\n\nThere are also a few more mundane issues. No mode-convergence tests or error estimates are provided; the SI says a few low-order modes suffice, but no table or plot demonstrates that. No code or data is shipped, which is a reproducibility minus. And the reference list is mangled — duplicate numbers, missing numbers, inconsistent formatting. These are fixable, but they should be fixed.\n\nOverall, this is a solid, useful paper for the graphene electron-optics community. The DE-based design of a tunable multi-dot array is new and likely to be picked up by others. The main request in revision should be: either validate the splitter in the near field (even with focused beams) or substantially tone down the device-control claims. I'd send this to a serious referee.","headline":"A new optimization-driven twist on graphene electron optics, soundly executed within its idealized model, but the device-level claims outrun what the simulations support.","tokens_in":16695,"tokens_out":3283,"would_cite":false,"duration_ms":36736,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A 3x3 array of independently gated quantum dots in graphene can be tuned to redirect or split an incident electron beam into chosen angular ranges, with differential evolution finding the required gate settings.","keywords":["tunable dot platform","graphene","quantum dots","electron optics","Mie scattering","multiple scattering","differential evolution","beam splitting"],"falsifier":"A concrete check would be a full tight-binding or experimental measurement of the angular current pattern for one of the paper's optimized 3x3 configurations: if the scattered current does not show a pronounced peak in the target angular range, for example because intervalley scattering, disorder, or the finite width of a focused beam destroys the plane-wave Mie prediction, the central claim fails. A simpler numerical falsifier is to compute the same optimized configuration with a multi-valley or tight-binding solver and compare the resulting angular scattering to the generalized Mie result.","tokens_in":15718,"feed_emoji":"⚡","tokens_out":11709,"duration_ms":118802,"temperature":0.7,"pith_summary":"The paper introduces a 'tunable dot platform' (TDP): a 3x3 array of circular gated quantum dots embedded in graphene, where each dot's potential can be set independently. It argues that, because the dots scatter massless Dirac electrons like tunable Mie scatterers, the array can be configured to steer an incident electron beam into a chosen angular range, or to split it between two ranges, producing asymmetric angle-dependent current patterns that a single dot cannot create. The central technical claim is that a differential evolution search over the nine dot potentials reliably finds configurations that maximize a fitness function measuring how much scattered current lands in the target angles, both in the far field and near a fixed radius. If correct, the TDP would be a reconfigurable, all-electrostatic building block for graphene electron optics: one device that can be reprogrammed in situ to deflect, split, or locally deliver current.","feed_headline":"Gated 3x3 dot array redirects electron beams in graphene","feed_subtitle":"Differential evolution finds gate settings that deflect or split an incoming electron wave into chosen angles.","key_machinery":"The central object is the tunable dot platform itself: a 3x3 array of circular potential steps in graphene, each with an independently adjustable potential. The carrying mechanism is generalized Mie scattering for multiple dots: the wavefunction is expanded in angular momentum modes around every dot, Graf's addition theorem translates scattered waves between dot-centered reference frames, and matching boundary conditions at all nine edges produces a linear system of size $MN \\times MN$ for the scattering coefficients. The far-field scattered current is then built from effective TDP coefficients $S_m$ as $j_{\\mathrm{sc},r}(r\\to\\infty,\\theta)=\\frac{2\\eta_0}{\\pi k r}\\sum_{m,n}S_m S_n^* e^{i(m-n)\\theta}$, and differential evolution maximizes a fitness function that rewards scattered current inside the target angular range while penalizing current outside it.","core_discovery":"The paper's central discovery is that a small, gate-defined array of quantum dots in graphene can act as a programmable scatterer whose angular response is shaped by the pattern of dot potentials. Using a generalized Mie-theory multiple-scattering calculation for single-valley Dirac fermions, the authors show that a uniform 3x3 TDP behaves essentially like one larger effective dot, whereas non-uniform potential patterns break the left-right symmetry and produce scattering lobes in directions determined by the configuration. They then invert the design problem with differential evolution: for targets such as a single angular window or two simultaneous windows, the algorithm converges to dot configurations whose far-field scattered current is concentrated in the desired directions, with comparable success for near-field current at a chosen radius. The paper claims this makes the TDP a reconfigurable electron optic component capable of beam deflection, beam splitting, and local current delivery, with the same optimization framework applicable to other geometries, wave profiles, and physical quantities.","pith_inferences":["If the single-valley plane-wave idealization is relaxed, intervalley scattering and short-range disorder would likely blur the optimized angular peaks; the same fitness-and-evolution scheme could be rerun with a multi-valley or tight-binding forward model to test how much directionality survives.","The authors' observation that near-field and far-field patterns can disagree implies that an 'optimized' configuration is only optimal for the specific detection geometry; a device designer would need to choose the fitness function to match where the current is actually measured.","The exponential growth of the configuration space with array size suggests differential evolution is doing real work here, and larger arrays or continuous potential landscapes could be explored with the same method, although convergence may become harder.","Pairing the TDP with gapped or bilayer graphene regions, as suggested in the discussion, could turn the same optimization machinery into a valley-selective beam splitter; one testable extension is to define the fitness function on valley-resolved scattered current."],"forward_implications":["A single TDP can be reprogrammed in situ by changing gate voltages, so one device can serve as a beam deflector, a beam splitter, or a near-field current source without changing its geometry.","Because all parameters enter through products like $kR$, an optimized configuration can be rescaled to dot sizes, energies, and voltages within current fabrication tolerances, as the authors note.","The optimization is not limited to far-field angles: the same fitness approach can concentrate current along a circle of fixed radius, which matters when a detector or probe sits close to the scattering region.","The method generalizes to other dot geometries, incident wave profiles, and target physical quantities, so the TDP concept is a starting point for a broader class of programmable electron optics."],"supporting_citations":[{"why":"Provides the single-dot Mie-scattering formalism for a circular potential step in graphene, which the TDP generalizes to nine dots.","marker":"1"},{"why":"Extends Dirac-electron scattering to a cluster of electrostatically defined quantum dots, giving the multiple-scattering framework the TDP uses.","marker":"3"},{"why":"Supplies the Graf addition theorem bounds used to re-express scattered waves between dot-centered reference frames.","marker":"4"},{"why":"Is the original Mie theory whose generalization underpins the scattering calculation.","marker":"50"},{"why":"Defines the differential evolution algorithm used to optimize the nine dot potentials.","marker":"10"},{"why":"Demonstrates independently gate-defined quantum dots in graphene, the experimental basis for the TDP's programmability.","marker":"51"},{"why":"Provides an experimental electrical analogue of Mie scattering in graphene, supporting the feasibility of scattering-based electron optic devices.","marker":"73"}],"fun_headline_variants":["Gated dot array steers electron beams in graphene","Dot potentials bend and split electron waves in graphene","Programmable graphene dot array focuses electron flow","Graphene dot array redirects electron beams to chosen angles"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The single-valley Dirac plane-wave scattering model, truncated to a small number of angular momentum modes, correctly predicts the current flow that a real gated device would produce despite finite contacts, focused beams, disorder, and possible intervalley scattering.","fun_headline_variants_meta":{"raw":{"variants":["Gated dot array steers electron beams in graphene","Dot potentials bend and split electron waves in graphene","Programmable graphene dot array focuses electron flow","Graphene dot array redirects electron beams to chosen angles"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00051,"raw_usage":{"total_tokens":2428,"prompt_tokens":839,"completion_tokens":1589,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":455,"completion_tokens_details":{"reasoning_tokens":1527}},"tokens_in":455,"tokens_out":1589,"duration_ms":11867,"temperature":1.0,"reasoning_tokens":1527,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T20:46:59.744308+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete check would be a full tight-binding or experimental measurement of the angular current pattern for one of the paper's optimized 3x3 configurations: if the scattered current does not show a pronounced peak in the target angular range, for example because intervalley scattering, disorder, or the finite width of a focused beam destroys the plane-wave Mie prediction, the central claim fails. A simpler numerical falsifier is to compute the same optimized configuration with a multi-valley or tight-binding solver and compare the resulting angular scattering to the generalized Mie result.","supporting_citations":[{"cited_title":"a ge zur Optik tr \\","cited_arxiv_id":null,"evidence_quote":"Is the original Mie theory whose generalization underpins the scattering calculation."},{"cited_title":"E.; Fuchs, J.-N","cited_arxiv_id":null,"evidence_quote":"Defines the differential evolution algorithm used to optimize the nine dot potentials."},{"cited_title":"Gate-defined electron--hole double dots in bilayer graphene","cited_arxiv_id":null,"evidence_quote":"Demonstrates independently gate-defined quantum dots in graphene, the experimental basis for the TDP's programmability."},{"cited_title":"M.; Connaughton, S.; Ott, C.; Weber, H","cited_arxiv_id":null,"evidence_quote":"Provides an experimental electrical analogue of Mie scattering in graphene, supporting the feasibility of scattering-based electron optic devices."}],"review_version":1}