{"id":"b085a994-251c-4348-afcd-1e2a481b32dc","arxiv_id":"2504.18043","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"ESR-STM magnetic resonance imaging of organic radical anions produces resonant-slice rings that map the delocalized spin density at sub-molecular resolution and can distinguish similar species by their 3D geometry.","lead":"A magnetic scanning tunneling microscope detected electron spin resonance in single organic molecules and imaged the location of their unpaired electrons with sub-molecular resolution. The method distinguishes nearly identical molecules by the three-dimensional shape of their spin density.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The paper's own site-D phase-cancellation explanation shows the MRI rings encode phase-dependent exchange integrals rather than spin density; without a quantitative inversion, the spin-density tomography claim is unsupported.","rationale":"The empirical observations are credible: the ESR spectra, the shift of resonance with tip height, the appearance of resonant-slice rings that move systematically with RF frequency and tip height, and the DFT adsorption geometries all support the reality of the measured effect. The concern is not about data quality but about the interpretation. The paper's central claim that the rings visualize spin density and enable spin-density tomography relies on the exchange coupling being a local, phase-insensitive function of spin-density magnitude. The paper's own explanation of the missing site-D signal contradicts that assumption, as the reader's weakest_assumption also noted. The concern is load-bearing because the abstract and conclusions make a stronger claim than the presented analysis can support: no algorithm or quantitative model is given to convert the observed rings into a three-dimensional spin-density map, and the site-D cancellation shows that phase information matters. I therefore agree with the reader's conditional verdict. The concern could be settled by a first-principles calculation of the position-dependent exchange field and a simulated MRI comparison, as described in concrete_test. The missing Supplementary Materials are an additional reason not to upgrade the verdict, but they are secondary to the phase-dependence issue.","tokens_in":15184,"tokens_out":6650,"duration_ms":79078,"concrete_test":"Recompute the MRI rings from first principles for the same DAF and fluorenone adsorption geometries: take the relaxed DFT structures, place a model Fe-terminated tip at the experimental heights, and evaluate the tip-position-dependent exchange field J(r_tip) two ways: (a) J proportional to the spatial overlap of the tip spin density with the molecular spin-density magnitude, and (b) J from full phase-dependent exchange integrals evaluated with the SOMO wavefunctions including minority-spin density. Generate simulated MRI images for the frequencies and heights of Figs. 3I-K and 3O-Q and compare against the measured ring positions, radii, and the missing site-D ring. If model (b) is required to reproduce the data, the claimed spin-density tomography is not established; if model (a) also reproduces all rings, including suppression at site D, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim converts observed resonant-slice rings into a map of molecular spin density. That conversion requires the tip-molecule exchange field to be a scalar function of the local spin-density magnitude at the tip position, independent of orbital phase and tip-apex structure. The paper itself supplies two counterexamples in the paragraph after Fig. 4: (i) negative spin density at the N atoms of DAF and the corresponding carbons of fluorenone is 'expected to reduce the total exchange integral' near site D, and (ii) 'the exchange integral depends on the wavefunction phase of the delocalized molecular spin,' so SOMO lobes of opposite phase around site D cancel. Both statements imply the exchange coupling is not proportional to the spin-density magnitude. Because the exchange matrix element between a tip state and the molecular SOMO is phase-sensitive and nonlocal, the rings may reflect SOMO phase and tip electronic structure rather than spin density. No quantitative relation between ring radii or positions and the spin-density field is given, and no tomographic inversion is demonstrated. This is an internal consistency problem, not merely a disagreement with the community consensus. The explicit statement that Supplementary Materials are not available further prevents cross-checking of the image filtering and transport-model details.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports electron spin resonance (ESR) and scanning magnetic resonance imaging (MRI) measurements on individual all-organic radical anions (DAF, fluorenone, DBF) adsorbed on a 2-ML MgO film on Ag(001). The authors observe ESR peaks with g-factors near 2, and MRI images that show ring-shaped resonant slices whose size and position change with RF frequency and tip height. They compare these rings with DFT-computed spin densities of the molecular anions, and use DFT-optimized adsorption geometries to explain differences between DAF and fluorenone. A non-equilibrium transport model with a single Anderson impurity is used to simulate bias-dependent ESR spectra. The paper claims that the MRI rings map the delocalized spin density with sub-molecular resolution, enabling spin-density tomography that distinguishes similar molecular species.","tokens_in":15456,"tokens_out":3855,"duration_ms":40639,"significance":"If the central interpretation is correct, this work extends ESR-STM to all-organic radicals and demonstrates a real-space probe of delocalized molecular spins, which is of considerable interest for molecular magnetism, nanographene spins, and spin-based electronics. The experiments appear carefully conducted, and the observation of frequency- and height-dependent resonant slices on molecular radical anions is a substantial technical advance. However, the step from ring positions to a quantitative spin-density map is not established. The paper's own discussion of site-D introduces phase-dependent and nonlocal exchange contributions that conflict with the simple 'exchange proportional to spin-density overlap' assumption. The text also explicitly states that Supplementary Materials are unavailable, which limits reproducibility. The qualitative visual comparison to DFT is compelling but not a quantitative tomography. The central claim is therefore defensible only in a weakened form, and the manuscript requires substantial revision to justify the tomography language.","major_comments":[{"comment":"The assumption that 'the exchange interaction between the tip and molecule is nearly isotropic and with a magnitude proportional to the overlap between the tip spin and each spin lobe' is contradicted by the two explanations given in the same paragraph for the missing site-D signal. First, negative spin density at the N atoms (DAF) or corresponding carbons (fluorenone) is said to 'reduce the total exchange integral'; second, the exchange integral is said to depend on the wavefunction phase, with opposite-phase SOMO lobes around site D canceling. Both statements imply that the observed rings encode a phase-sensitive, nonlocal exchange matrix element, not the local magnitude of the spin density. Since the tomographic claim relies on the proportionality assumption, the paper needs either a quantitative relation between ring radii/positions and spin density or an explicit model that accounts for phase and nonlocality. As written, the abstract's 'spin-density tomography' claim is not supported.","section":"Results, paragraph after Fig. 4"},{"comment":"The MRI sequences are not tomographic in the standard sense: each image is a 2D slice of a constant tip-molecule interaction isosurface, not a direct map of spin density at that height. The paper acknowledges this in Fig. 3A,B but then refers to 'spin-density tomography' without performing an inversion. The comparison to DFT spin density (Fig. 3G,M and Fig. 4D,H) is purely visual; no quantitative metric (e.g., correlation, RMS deviation, or a fitted interaction model) is provided. A load-bearing component of the central claim is therefore missing.","section":"Results, Fig. 3 and Fig. 4"},{"comment":"The manuscript states that 'Supplementary Materials (SM) are not available for this version of this manuscript.' However, the text refers to the SM for essential technical details: high-pass filtering of MRI images (Fig. 3 caption), detailed transport-model discussions (Results), the spin-torque mechanism for negative-bias contrast reversal (Results), and the site-dependent ESR signal reversal (Results). Without these materials, readers cannot verify the image processing, the model assumptions, or the proposed mechanisms. This is a reproducibility concern that must be resolved before publication.","section":"Supplementary Materials statement"},{"comment":"The transport model is used to support the bias-dependent ESR interpretation, but the simulation 'reproduces most of the main features' and explicitly fails to capture the contrast inversion at site B. The fit parameters include the ionization energy ε, Hubbard U, tip coupling Γ_tip, tip polarization P, RF driving amplitude, and temperature; the paper states that ε, U, and Γ_tip are fitted from the ESR resonance position using exchange-field theory. Given the number of adjusted parameters and the model's failure at site B, the transport-model section should be framed as illustrative rather than as a quantitative validation of the experimental interpretation.","section":"Results, transport model paragraph"}],"minor_comments":[{"comment":"The text states that CO-tip itProbe images were acquired of 'fluorenyl and DAF,' but the Fig. 1 caption describes panels (E,F) as 'fluorenone (E) and DAF (F).' This inconsistency should be corrected.","section":"Figure 1 caption vs. text"},{"comment":"The quoted g-factors, 1.97 ± 0.04 and 1.98 ± 0.03, have large uncertainties. It would be helpful to state how many measurements and tip apexes contribute to these values, and to report the confidence intervals more explicitly.","section":"Fig. 2 and Results"},{"comment":"The phrase 'For most tips tested' appears three times without a number. Please give the total number of distinct tip apexes and how many exhibited the stated behavior.","section":"Results, tip dependence"},{"comment":"The MRI images are described as 'high-pass filtered for clarity (see SM).' Since the SM is unavailable, the filter type, cutoff, and any artifacts should be described in the main text or a footnote.","section":"Fig. 3 and Fig. 4 captions"},{"comment":"Reference 19 lacks volume and page numbers (it is given as 'acsnano.4c14327'). Reference 52 is an unpublished manuscript; if it is not yet accepted, please provide a preprint identifier or remove it from the reference list.","section":"References"},{"comment":"The paragraph on contrast reversal proposes three distinct mechanisms (SOMO/SUMO channel switching, tip gating, and spin-torque initialization). To avoid the appearance of ad hoc explanation, please clearly label which parts are directly supported by data and which are hypotheses, and consider consolidating the discussion.","section":"Results, contrast reversal paragraph"}],"recommendation":"major_revision","confidential_remarks":"The experimental data appear novel and valuable, but the 'spin-density tomography' claim is not quantitatively supported and the paper's own site-D discussion undermines the proportionality assumption. The missing Supplementary Materials is an additional concern for reproducibility. I would recommend major revision with the expectation that the authors either provide a quantitative inversion or substantially soften the claims to 'imaging of exchange-interaction isosurfaces correlated with spin density.'"},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: the experimental work is genuine and the data are the first ESR-MRI of all-organic radical anions, but the spin-density tomography claim is a step ahead of what the paper actually demonstrates.\n\nThe ring patterns that appear as you sweep RF frequency or tip height behave exactly as resonant slices through a constant-interaction isosurface. The g-factors are measured cleanly, and the comparison with DFT spin density is visually strong. The observation that DAF and fluorenone show rings in different order as the tip is lowered is a clever result, and the adsorption-geometry explanation is plausible. That is real progress.\n\nThe soft spot is the central mapping claim. The authors say the exchange interaction is 'nearly isotropic and proportional to the overlap between the tip spin and each spin lobe of the delocalized radical.' Then they explain the missing site-D signal by pointing out that the exchange integral depends on the wavefunction phase and that negative spin density reduces it. Those two statements contradict the simple spin-density picture. If phase and sign matter, the rings are not a direct image of spin-density magnitude. The paper never gives a quantitative relation between ring positions and a spin-density field, and no tomographic inversion is shown. So 'spin-density tomography' is an overstatement.\n\nThe missing supplementary materials make this worse—you can't check the filtering or the transport model details. The transport model also has a lot of fitted parameters, so its agreement with a few spectra doesn't carry much weight. The contrast-reversal mechanism is speculative, though the authors say so themselves.\n\nThese are not fatal problems. The raw observations stand, and the paper is honest about the site-D puzzle. But the abstract and the word 'tomography' need to be toned down, and a serious revision should include either a quantitative model or a clear statement of what is qualitative.\n\nI'd send this to peer review. It deserves a real referee. The experimental core will survive the scrutiny; the claims will need to be trimmed.","headline":"Genuinely new experimental data, but the 'spin-density tomography' claim is ahead of the evidence.","tokens_in":15986,"tokens_out":2645,"would_cite":true,"duration_ms":27524,"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":"ESR-STM now images the unpaired-electron spin density of single organic radicals with sub-molecular resolution.","keywords":["ESR-STM","magnetic resonance imaging","spin density","organic radicals","fluorenone","exchange interaction","sub-molecular resolution","SOMO"],"falsifier":"Perform the same MRI sequence on a radical whose high-spin-density lobe has zero SOMO wavefunction amplitude at the tip position or is surrounded by opposite-phase lobes: a purely spin-density map would still ring there, while phase-dependent exchange would not. Alternatively, rotate a single adsorbed molecule by 90 degrees on the surface and check whether the resonant-slice positions track the fixed spin-density lobes or shift with the molecular orientation relative to the tip's crystal axes; a shift would indicate anisotropic exchange rather than isotropic spin-density mapping.","tokens_in":14985,"feed_emoji":"🧲","tokens_out":5119,"duration_ms":52167,"temperature":0.7,"pith_summary":"The paper sets out to show that electron spin resonance scanning tunneling microscopy (ESR-STM) can image the spatial distribution of the unpaired electron in single organic radical anions with sub-molecular resolution. On fluorene-derived anions adsorbed on a two-monolayer MgO film, the magnetic tip's highly localized exchange field creates narrow resonant-slice rings that trace constant tip-molecule interaction around lobes of high spin density. Because the rings depend on tip height and frequency, a sequence of images amounts to three-dimensional spin-density tomography that can tell structurally similar molecules apart. The authors argue this opens a route to atomic-scale visualization of spin density and magnetic interactions in organic systems such as nanographenes.","feed_headline":"MRI of single molecules maps unpaired-electron spin density","feed_subtitle":"Resonant rings trace the spin lobes of fluorene radicals, telling nearly identical molecules apart.","key_machinery":"The central object is the exchange interaction between the magnetic tip apex and the molecular spin, approximated as nearly isotropic and proportional to the overlap between the tip's spin and each spin lobe. Scanning at fixed RF frequency, the ESR signal appears only where the lateral tip position makes the tip field shift the spin resonance into the drive frequency; the observed rings are the intersection of the scan plane with a constant-exchange-energy isosurface. The SOMO/SUMO cotunneling dichotomy is the second mechanism: which orbital dominates tunneling at a given lateral position sets the sign of the magnetoresistive ESR signal, and spin-torque initialization accounts for the bias-dependent reversal.","core_discovery":"The paper's central claim is that a magnetic STM tip's exchange field can serve as a sub-molecular probe of a delocalized radical's spin density. ESR spectra of three fluorene-derived radical anions on MgO/Ag(001) show spin $S=1/2$ resonances with $g\\approx 1.97$ to $1.98$, and scanning the tip at fixed RF frequency produces narrow resonant-slice rings that enclose the density-functional-theory (DFT) predicted high-spin-density lobes of the anion's singly occupied molecular orbital (SOMO). Sequences taken at different frequencies or tip heights map the same interaction isosurface in three dimensions, so the molecules' adsorption tilt and distortion become visible: DAF shows the first ring at the ketone-carbon lobe, fluorenone at the opposite lobe. The paper also accounts for missing contrast at one high-density lobe by wavefunction-phase cancellation of the exchange integral and for light/dark contrast reversal by switching between SOMO and SUMO cotunneling channels, with spin-torque initialization flipping the overall contrast at negative bias.","pith_inferences":["If the isotropic-overlap picture holds, the same resonant-slice technique could image spin-density sign, not just magnitude, because negative-spin-density regions suppress the exchange integral; the paper's site-D result hints at this but does not claim it.","The SOMO/SUMO contrast reversal suggests MRI could be used as an orbital-selective probe: by tuning bias and tip gating, the same physical spin density might be imaged through either orbital, mapping SOMO and SUMO wavefunctions separately.","A quantitative test that follows from the paper's model is that ring radius versus tip-height curves at fixed RF detuning should collapse onto one exchange-field isosurface; if they do not, the assumed isotropic-overlap model needs revision."],"forward_implications":["ESR-STM MRI can distinguish closely related organic radicals that look identical in STM topography, using only their spin-density distribution and three-dimensional adsorption geometry.","The method gives a direct three-dimensional map of where a delocalized spin resides, so it can guide tip placement for coherent spin driving and sensing on individual molecules.","The combination of neV energy resolution and atomic spatial resolution lets MRI detect differences in molecular tilt and distortion at the picometer scale.","Applied to nanographenes and graphene nanoribbons, the same exchange-field imaging should expose the spatial structure of edge-state spins and intramolecular spin coupling."],"supporting_citations":[{"why":"Supplies the foundational ESR-STM demonstration of a single spin on a surface that this paper extends to molecular radicals.","marker":"[12]"},{"why":"Establishes the resonant-slice MRI method for single atoms that the paper adapts to organic molecules.","marker":"[33]"},{"why":"Attributes frequency shifts above about 1 GHz to the tip's exchange field, the basis for interpreting the resonant rings.","marker":"[43]"},{"why":"Describes the combined exchange and dipole tip field that sets the resonance condition used in the MRI images.","marker":"[42]"},{"why":"Demonstrates spin-torque-driven ESR in a pentacene molecule, providing the mechanism for the bias-dependent contrast reversal.","marker":"[23]"},{"why":"Provides the many-body non-equilibrium transport model used to simulate the measured ESR spectra.","marker":"[50]"},{"why":"Demonstrates atomic-scale spin sensing by exchange interaction with a magnetic molecule at the tip, the predecessor of this imaging.","marker":"[45]"},{"why":"Shows probing and imaging of spin interactions with a magnetic single-molecule sensor at sub-nanometer resolution.","marker":"[46]"}],"fun_headline_variants":["Sub-molecular MRI of single organic radicals maps spin density","Spin-density MRI at sub-molecular scale for single molecules","Single-molecule MRI reveals unpaired-electron spin lobes","Magnetic tip images spin density of individual radicals","Sub-molecular spin maps from STM electron spin resonance"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the tip-molecule exchange interaction is nearly isotropic and grows with the overlap between the tip spin and each spin-density lobe, so that the position of a resonant slice is a direct readout of spin-density magnitude.","fun_headline_variants_meta":{"raw":{"variants":["Sub-molecular MRI of single organic radicals maps spin density","Spin-density MRI at sub-molecular scale for single molecules","Single-molecule MRI reveals unpaired-electron spin lobes","Magnetic tip images spin density of individual radicals","Sub-molecular spin maps from STM electron spin resonance"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000293,"raw_usage":{"total_tokens":1678,"prompt_tokens":885,"completion_tokens":793,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":501,"completion_tokens_details":{"reasoning_tokens":715}},"tokens_in":501,"tokens_out":793,"duration_ms":7151,"temperature":1.0,"reasoning_tokens":715,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T10:25:37.568344+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Perform the same MRI sequence on a radical whose high-spin-density lobe has zero SOMO wavefunction amplitude at the tip position or is surrounded by opposite-phase lobes: a purely spin-density map would still ring there, while phase-dependent exchange would not. Alternatively, rotate a single adsorbed molecule by 90 degrees on the surface and check whether the resonant-slice positions track the fixed spin-density lobes or shift with the molecular orientation relative to the tip's crystal axes; a shift would indicate anisotropic exchange rather than isotropic spin-density mapping.","supporting_citations":[{"cited_title":"Baumann, W","cited_arxiv_id":null,"evidence_quote":"Supplies the foundational ESR-STM demonstration of a single spin on a surface that this paper extends to molecular radicals."},{"cited_title":"Willke, K","cited_arxiv_id":null,"evidence_quote":"Establishes the resonant-slice MRI method for single atoms that the paper adapts to organic molecules."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Attributes frequency shifts above about 1 GHz to the tip's exchange field, the basis for interpreting the resonant rings."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Describes the combined exchange and dipole tip field that sets the resonance condition used in the MRI images."},{"cited_title":"Kovarik, R","cited_arxiv_id":null,"evidence_quote":"Demonstrates spin-torque-driven ESR in a pentacene molecule, providing the mechanism for the bias-dependent contrast reversal."},{"cited_title":"Reina-Gálvez, C","cited_arxiv_id":null,"evidence_quote":"Provides the many-body non-equilibrium transport model used to simulate the measured ESR spectra."},{"cited_title":"Verlhac, N","cited_arxiv_id":null,"evidence_quote":"Demonstrates atomic-scale spin sensing by exchange interaction with a magnetic molecule at the tip, the predecessor of this imaging."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows probing and imaging of spin interactions with a magnetic single-molecule sensor at sub-nanometer resolution."}],"review_version":1}