{"id":"9437ed20-8fe9-47d3-ac5c-9eaee7d9b937","arxiv_id":"1908.09430","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"TDDFT calculations on III-V quantum dots predict a LUMO symmetry crossover in GaP dots near 1.5 nm, a nearly linear size scaling of the InP optical gap, and a Vegard-like behavior in GaInP alloy dots with a small positive bowing parameter.","lead":"This paper uses large-scale density functional theory and time-dependent density functional theory to compute the electronic and excitonic properties of indium phosphide, gallium phosphide, and their alloy quantum dots containing up to about a thousand atoms.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The predicted GaP LUMO crossover near 1.5 nm rests on an uncalibrated single-particle level ordering; the same B3LYP/pseudohydrogen model yields GaP optical gaps significantly above experiment, so the crossover location is not anchored.","rationale":"The reader's weakest assumption is the pseudohydrogen surface model, and I agree that is the critical vulnerability. I want to sharpen it: the crossover is a level-ordering prediction between two close-lying conduction states, and the paper itself contains two explicit admissions that the quantitative energy scale of the model is not calibrated for GaP. First, Sec. III B (Fig. 3(c)) states the calculated GaP optical gaps are significantly larger than the only available experimental data, which the authors attribute to high-temperature measurements rather than validating the model. Second, Sec. III A says the calculated exciton binding energies are significantly smaller than experiment. Neither discrepancy is necessarily fatal, but together they mean the relative Gamma5/Gamma1 conduction-state energy that sets the crossover has no external anchor. Near-degeneracies are exactly the quantities most sensitive to functional, basis, passivation, and surface details, so a crossover-diameter prediction from this model needs an independent check. The unsupported 'sudden drop' near x = 0.8 in Sec. III C is also a concern, but it is not the central claim. The recommended verdict remains CONDITIONAL/UNCHANGED because the paper's claims are plausible and testable, but the quantitative crossover prediction is not yet established.","tokens_in":15299,"tokens_out":11043,"duration_ms":117333,"concrete_test":"Recompute the GaP QD series (D = 1.07, 1.5, 2.29, 3.0 nm) at the same B3LYP/def2-SVP level with (i) the original 1.25/0.75 pseudohydrogens, (ii) an explicitly capped surface (e.g., H/OH or Cl/H pairs), and (iii) a second hybrid functional such as PBE0. In each case, project the lowest conduction states onto bulk Gamma1c and X-derived Bloch states and record the LUMO symmetry as a function of D. If the Gamma5/Gamma1 crossing diameter shifts by more than ~0.3 nm between setups, or if the LUMO at D = 3 nm is not Gamma5 in any setup, the crossover prediction is not robust to the passivation/functional choice and should not be presented as a quantitative spectroscopic signature.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing quantity is not the HOMO-LUMO gap but the relative energy of two conduction-band states in GaP QDs: the zone-center Gamma1-derived state and the X-derived Gamma5 state. Their near-degeneracy is what makes the LUMO switch symmetry near D = 1.5 nm. That relative energy is never calibrated against any GaP-specific reference. The pseudohydrogen passivation (charges 1.25/0.75, Sec. II) is transferred from previous studies and tested only indirectly; the paper's own GaP comparison in Sec. III B (Fig. 3(c)) shows calculated optical gaps 'significantly larger' than the available experimental data, attributed to high-temperature measurements, and Sec. III A reports exciton binding energies 'significantly smaller' than experiment. Even if those discrepancies are due to experimental conditions or to the absence of spin-orbit coupling, they mean the computed conduction-state ladder for GaP has no confirmed absolute or relative accuracy. A shift of a few tenths of an eV in the relative Gamma5/Gamma1 ordering, which is well within the expected error of B3LYP on a 65-atom passivated cluster, would move the crossing diameter by several tenths of a nanometer or suppress the crossover entirely. The prediction is therefore plausible but currently contingent on an unvalidated level-ordering calculation.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports large-scale DFT and TDDFT calculations of the electronic and excitonic optical properties of InP, GaP, and GaInP colloidal quantum dots with up to ~1100 atoms. Using PBE for geometry, B3LYP for single-particle states, and linear-response TDDFT for excitonic effects, the authors compute size-dependent optical gaps, exciton binding energies, radiative lifetimes, singlet-triplet splittings, and absorption spectra. The main findings are: (i) InP QD optical gaps scale nearly linearly with inverse diameter and agree well with experiment; (ii) the radiative lifetime increases linearly with dot size; (iii) for GaP QDs, the LUMO symmetry is predicted to switch from Γ5-like at larger diameters to Γ1-like below about 1.5 nm, accompanied by enhanced band-edge absorption; and (iv) GaInP alloy QDs obey Vegard's law down to ultra-small sizes, with a positive, size-dependent bowing parameter dominated by volume deformation. The paper frames these as first-principles predictions that can guide spectroscopic studies of III-V QDs.","tokens_in":15560,"tokens_out":5269,"duration_ms":53264,"significance":"If the central predictions hold, the paper provides valuable quantitative guidance for III-V QD optoelectronics, particularly the falsifiable GaP LUMO-crossover signature near 1.5 nm and the linear size dependence of the InP radiative lifetime. The use of a hybrid functional with TDDFT on dots of up to 1100 atoms is a strength, as is the systematic comparison among InP, GaP, and their alloy. The bowing decomposition following Bernard and Zunger is also a useful contribution. However, the quantitative claims are weakened by the absence of error bars on all fitted exponents and bowing parameters, by the uncalibrated GaP conduction-band level ordering that underpins the crossover prediction, and by the thoroughness of the alloy configurational averaging. These issues currently limit the strength of the conclusions.","major_comments":[{"comment":"The predicted Γ5→Γ1 LUMO crossover near D = 1.5 nm is a central claim, but the relative energy of the two conduction-band states (Γ5-like versus Γ1-like) is never calibrated against any GaP-specific reference. The paper itself reports in Sec. III B (Fig. 3(c)) that the calculated GaP optical gap is 'significantly larger' than the available experimental data, and in Sec. III A that exciton binding energies are 'significantly smaller' than experiment. These discrepancies imply that the computed conduction-state ladder for GaP has no confirmed absolute or relative accuracy. A shift of a few tenths of an eV in the relative Γ5/Γ1 ordering, well within the expected error of B3LYP on a passivated cluster of 65–100 atoms, would move the crossover diameter by several tenths of a nanometer or even suppress the crossover entirely. Please provide a sensitivity analysis (for example, varying the relative offset, testing a second hybrid functional, or comparing with a bulk GaP band-structure benchmark) and discuss the resulting uncertainty in the crossover diameter.","section":"Sec. III A and III B, Eqs. (1) and (2)"},{"comment":"The fits for the size-scaling exponents and prefactors (Cg = 1.12 and 1.2 for InP; exponents 1.23 and 1.47 for GaP; β = 0.77 for InP exciton binding energy; exponents for the singlet-triplet splittings in Fig. 3(d)) are reported without any statistical measures such as error bars, confidence intervals, number of fit points, or residuals. This is load-bearing because the comparisons to effective-mass theory (Cg = 2) and to empirical pseudopotential results (Cg = 1.36) rest entirely on these fitted values. Please report the fitting details and the associated uncertainties, and discuss how the systematic error of the pseudohydrogen surface model affects the fitted exponents.","section":"Sec. III C, Fig. 5(d) and Eq. (3)"},{"comment":"The alloy gap bowing is computed from an average over only ten random configurations per composition, with no convergence test with respect to the number of configurations and no error bars on the averaged gaps. In addition, the decomposition in Eq. (3) yields b = b_vd + b_ce + b_sr = 0.30 eV, while the text states that the parabolic fit of Fig. 5(d) gives about 0.22 eV; this 0.08 eV discrepancy is not explained and undermines the claim that the sum 'well reproduces' the fitted bowing. Please provide a configurational convergence study, error propagation on the averaged gaps and bowing, and an explicit reconciliation of the two values.","section":"Fig. 1 caption and Sec. III B"},{"comment":"Spin-orbit coupling is neglected throughout, yet the paper reports quantitative values for radiative lifetimes (e.g., τ_GaP = 5.86 μs and τ_InP = 7.66 ns at D = 1.5 nm), singlet-triplet splittings, and oscillator strengths. For InP and GaP the valence-band spin-orbit splitting is of order 0.1 eV, which is comparable to or larger than several of the reported fine-structure splittings. While the LUMO crossover itself is a conduction-band phenomenon, the exciton manifold, lifetimes, and absorption intensities will be affected. Please estimate the impact of spin-orbit coupling on the reported quantitative predictions, or state explicitly that the fine-structure energies and lifetimes are expected to change once it is included.","section":"Sec. II"}],"minor_comments":[{"comment":"The text writes 'Turbolmole' but the program is TURBOMOLE, as in Ref. 32.","section":"Sec. II"},{"comment":"The axis label 'Exction lifetime (ns)' in the figure panel contains a typo; it should read 'Exciton lifetime (ns)'.","section":"Fig. 2(c)"},{"comment":"The phrase 'Vergard's law' appears in the first paragraph of Sec. III C; it should be 'Vegard's law'.","section":"Sec. III C"},{"comment":"The radiative lifetime formula uses a refractive index n, but the numerical values used for InP, GaP, and CdSe are never specified. Please state the n values used in the calculations.","section":"Sec. II"},{"comment":"The comparison with GaP experiments in Fig. 3(c) is brief; please provide a few sentences describing the sample quality, size distribution, and any thermal corrections used to interpret the high-temperature (≈400°C) data.","section":"Sec. III B"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and offers a useful computational survey, but the GaP crossover prediction is currently insufficiently anchored. I would like the editor to ensure that the requested sensitivity analysis and uncertainty quantification are treated as substantive, not cosmetic, revisions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper delivers three concrete predictions: a GaP LUMO symmetry crossover near 1.5 nm, a positive size-dependent bowing in GaInP QDs, and a linear size scaling for InP exciton lifetimes. The computations are large-scale hybrid B3LYP plus TDDFT on dots up to ~1100 atoms, and the InP optical gaps match experiments well. The alloy bowing decomposition follows the standard Bernard-Zunger scheme and the Vegard check on bond lengths is a nice touch.\n\nThe main soft spot is the GaP crossover. The load-bearing quantity is the relative energy of the Γ5 and Γ1 conduction states, and that relative energy is never calibrated against a GaP-specific reference. The paper's own GaP optical gaps sit well above the experimental data, and the exciton binding energies sit well below, so the conduction-state ladder has no confirmed accuracy. A few tenths of an eV of error in the level ordering could move the crossing diameter by several tenths of a nanometer, or suppress the crossover entirely. This does not kill the prediction, but it should be labeled as a prediction contingent on level-ordering accuracy.\n\nOther soft spots: the scaling and bowing fits have no error bars; the alloy averages use only ten random configurations with no convergence test; spin-orbit coupling is neglected, which is relevant for near-degenerate GaP conduction states; and the x≈0.8 absorption drop is asserted without being shown. The pseudohydrogen surface model is carried over from prior work and not tested against ligand-covered dots, which is a caveat for the predicted absolute gaps. The citation pattern looks normal; the relevant atomistic and effective-mass literature is cited.\n\nNone of these are fatal. The paper is technically competent, honest about most of its limitations, and the InP results give confidence in the method. The GaP part needs either a calibration check or a softer claim. The GaInP bowing analysis is solid and well decomposed.\n\nRecommendation: send to peer review. Referees should ask for error bars on fits, a convergence check on the alloy averaging, an estimate of spin-orbit corrections to the level ordering, and for the x≈0.8 claim to be shown or removed. After revision, this will be a useful reference for the III-V QD community. If your reading group works on QDs, the crossover is a good discussion item; otherwise it is not essential reading.","headline":"Solid TDDFT roadmap with a testable GaP LUMO crossover prediction, but the crossover's level ordering is uncalibrated and the paper needs fit error bars, alloy-convergence checks, and a bit more restraint in the claims.","tokens_in":16135,"tokens_out":4720,"would_cite":true,"duration_ms":46882,"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":"This paper predicts that gallium phosphide quantum dots undergo a LUMO symmetry crossover at about 1.5 nm diameter, sharply enhancing their band-edge absorption.","keywords":["quantum dots","excitons","time-dependent density functional theory","indium phosphide","gallium phosphide","GaInP alloy","band gap bowing","LUMO symmetry crossover"],"falsifier":"Measure the absorption onset and oscillator strength of monodisperse, size-selected GaP quantum dots with diameters from about 1.2 to 2.0 nm. The crossover prediction requires a sharp jump in band-edge absorption intensity and singlet-triplet splitting as the diameter falls below roughly 1.5 nm; a smooth, featureless size dependence across that range would rule it out.","tokens_in":15108,"feed_emoji":"⚛️","tokens_out":9018,"duration_ms":80008,"temperature":0.7,"pith_summary":"This paper builds atomistic models of indium phosphide, gallium phosphide, and gallium-indium phosphide quantum dots with up to about a thousand atoms and computes their optical properties with density functional theory plus time-dependent density functional theory. The authors claim two quantitative results that experiments can check. For InP dots, the optical gap matches measured values and scales nearly linearly with inverse diameter, and the radiative exciton lifetime grows linearly with dot size. For GaP dots, they predict that the lowest unoccupied molecular orbital changes symmetry from $\\Gamma_5$-like to $\\Gamma_1$-like as the diameter falls below about 1.5 nm, and that this crossover sharply increases the absorption intensity of the band-edge exciton. They also report that the lattice constant of alloyed GaInP dots follows the linear alloy rule down to very small sizes, with a positive, size-dependent band-gap bowing dominated by volume deformation.","feed_headline":"GaP quantum dots flip symmetry near 1.5 nm, brightening absorption","feed_subtitle":"A symmetry switch in the lowest electron state would make tiny GaP dots absorb light much more strongly.","key_machinery":"The carrying mechanism is the size-dependent ordering of the conduction-band states under quantum confinement in an atomistic model with tetrahedral symmetry. The paper computes single-particle states with a hybrid exchange-correlation functional and then adds excitonic effects through linear-response time-dependent density functional theory, using pseudohydrogen atoms with modified nuclear charges to passivate the surface. The decisive object is the symmetry of the LUMO state: as the dot shrinks, the $\\Gamma_1$-like state is pulled below the $\\Gamma_5$-like state, and this level crossing at about 1.5 nm diameter for GaP reorganizes the exciton manifold, enlarges the singlet-triplet splitting, and increases the oscillator strength of the lowest allowed transition.","core_discovery":"The central discovery is a predicted electronic-state crossover in GaP quantum dots. In bulk GaP the conduction-band minimum sits at the X point, but in a confined dot the lowest unoccupied state is not simply the folded bulk minimum. The calculations show that for dots larger than about 1.5 nm this LUMO has $\\Gamma_5$ symmetry, while for smaller dots it becomes $\\Gamma_1$-like, the same symmetry as the InP conduction minimum at the zone center. Because the HOMO is always $\\Gamma_5$-like, the crossover changes the band-edge transition from a $\\Gamma_5\\to\\Gamma_5$ transition with weak oscillator strength to a $\\Gamma_5\\to\\Gamma_1$ transition with much stronger absorption. The paper further claims that the optical gap of InP dots scales as an inverse power of diameter with exponent near 1.2, that exciton binding energies scale as $D^{-0.77}$, and that radiative lifetimes increase linearly with size, and that in GaInP alloy dots the linear composition rule for lattice constants holds with a small, positive, size-dependent bowing parameter.","pith_inferences":["If the GaP crossover is real, the same physics may appear in other indirect-gap III-V nanocrystals; the crossover diameter should depend on the position of the conduction-band valleys and could be shifted by strain or by shell material, a testable prediction beyond the paper.","The crossover prediction could be checked immediately by measuring excitation or absorption spectra of size-selected GaP dots across the 1.2-2.0 nm range; the paper's surface model predicts a threshold, so a null result would indicate that real ligands alter the level ordering.","The linear lifetime scaling found here may be a general feature of strongly confined direct-gap dots, not just InP; comparing with CdSe dots of identical shape would separate material-specific from universal confinement effects.","Because the paper averages over ten random alloy configurations, the bowing parameters carry configurational variance; a follow-up could report the spread to show when the parabolic fit is meaningful."],"forward_implications":["For InP quantum dots, the nearly linear inverse-diameter scaling of the optical gap gives a simple design rule for tuning emission color by size.","GaP dots below about 1.5 nm should show a sharp increase in band-edge absorption intensity and a larger singlet-triplet splitting, providing a spectroscopy-visible signature of the predicted crossover.","The linear growth of radiative exciton lifetime with dot size implies that larger InP dots emit more slowly, which matters for LED efficiency and lifetime engineering.","GaInP alloy dots obey the linear lattice-constant rule and have a small, positive bowing, so composition can tune the gap predictably even in ultra-small dots.","Excitonic corrections leave the scaling exponents nearly unchanged, so single-particle calculations already capture the main size trends for these III-V dots."],"supporting_citations":[{"why":"Experimental InP quantum-dot optical gaps used as the benchmark for the computed gap scaling.","marker":"29,31"},{"why":"Experimental GaP quantum-dot optical gaps that the calculated gaps are compared against.","marker":"30"},{"why":"Supply the expression connecting exciton energy and transition dipole moment to the radiative decay lifetime.","marker":"37,38"},{"why":"Provides the decomposition of alloy band-gap bowing into volume deformation, charge exchange, and structure relaxation.","marker":"57"},{"why":"Gives the bulk GaInP direct-gap bowing parameter to which the quantum-dot bowing is compared.","marker":"56"},{"why":"Provides the empirical pseudopotential scaling exponent for InP dots that the present exponent is contrasted with.","marker":"42"},{"why":"Experimental CdSe exciton binding energy scaling used to validate the computational approach.","marker":"45"}],"fun_headline_variants":["GaP dot symmetry flip brightens tiny quantum dots","Symmetry crossover brightens GaP dots below 1.5 nm","GaP quantum dot LUMO symmetry switch boosts absorption","Size-driven symmetry switch enhances GaP dot absorption","Tiny GaP dots get brighter via LUMO symmetry change"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole prediction rests on the assumption that replacing the real surface ligands with pseudohydrogen atoms of modified nuclear charge reproduces the near-band-edge electronic structure of actual colloidal dots closely enough that the computed gaps, lifetimes, and especially the GaP crossover diameter near 1.5 nm survive in synthesized dots.","fun_headline_variants_meta":{"raw":{"variants":["GaP dot symmetry flip brightens tiny quantum dots","Symmetry crossover brightens GaP dots below 1.5 nm","GaP quantum dot LUMO symmetry switch boosts absorption","Size-driven symmetry switch enhances GaP dot absorption","Tiny GaP dots get brighter via LUMO symmetry change"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000227,"raw_usage":{"total_tokens":1554,"prompt_tokens":1111,"completion_tokens":443,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":727,"completion_tokens_details":{"reasoning_tokens":360}},"tokens_in":727,"tokens_out":443,"duration_ms":3985,"temperature":1.0,"reasoning_tokens":360,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:11:33.594477+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the absorption onset and oscillator strength of monodisperse, size-selected GaP quantum dots with diameters from about 1.2 to 2.0 nm. The crossover prediction requires a sharp jump in band-edge absorption intensity and singlet-triplet splitting as the diameter falls below roughly 1.5 nm; a smooth, featureless size dependence across that range would rule it out.","supporting_citations":[{"cited_title":"Micic \\ and\\ author A","cited_arxiv_id":null,"evidence_quote":"Experimental GaP quantum-dot optical gaps that the calculated gaps are compared against."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the decomposition of alloy band-gap bowing into volume deformation, charge exchange, and structure relaxation."},{"cited_title":"Merle , author D","cited_arxiv_id":null,"evidence_quote":"Gives the bulk GaInP direct-gap bowing parameter to which the quantum-dot bowing is compared."},{"cited_title":"Fu \\ and\\ author A","cited_arxiv_id":null,"evidence_quote":"Provides the empirical pseudopotential scaling exponent for InP dots that the present exponent is contrasted with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Experimental CdSe exciton binding energy scaling used to validate the computational approach."}],"review_version":1}