{"id":"db75badb-089f-431c-ae6d-87a3f405e026","arxiv_id":"2412.11884","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Direct majority-minority sublattice coupling of dibenzotriangulene units on Au(111) produces ferromagnetic dimers (quintet) and trimers (septet) with an inter-unit exchange of 7 meV, bypassing the spin-quenching pentagonal rings that plagued earlier triangulene couplings.","lead":"The paper builds tiny chains of a carbon-based molecule called dibenzotriangulene on a gold surface and shows that the chains can be ferromagnetically coupled, with an exchange energy of 7 meV in the dimer. It matters because it offers a route to purely organic magnetic materials without metal atoms, a step toward molecular spintronics and quantum simulation of spin chains.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"All key high-spin claims hinge on the AFM-based structural assignment of intact majority-minority DBT coupling; a misassigned or cyclodehydrogenated isomer would erase the sublattice imbalance and the ferromagnetism.","rationale":"The reader's weakest_assumption identifies the same load-bearing concern that I find: the entire ferromagnetic assignment is downstream of the structural assignment of 3k and 5FM as intact, bipartite, majority-minority coupled DBT units. The paper's own narrative makes this explicit—previous direct majority-minority triangulene couplings failed precisely because cyclodehydrogenation formed pentagonal rings and quenched spin. The DBT gulf-region design is the proposed fix, and the only direct evidence that the fix worked is the AFM-based structural determination. The spectroscopic data are consistent with the assigned high-spin states but are not independently discriminating: the 14 meV step in 3k could, in principle, be a spin excitation of a different, non-bipartite isomer, and the zero-bias Kondo feature only establishes open-shell character, not S=2. For the trimer, the quantitative discrepancy between the observed 23 meV excitation and the calculated 17 meV state further weakens the already model-dependent energetic assignment. I therefore agree with the reader that this structural premise is the crux and that the verdict should remain CONDITIONAL. No additional concern appears severe enough to move the verdict to REJECT; the evidence presented is substantial, and the proposed computational test of the alternative isomer would settle the ambiguity.","tokens_in":16534,"tokens_out":9954,"duration_ms":98318,"concrete_test":"Compute CASSCF-NEVPT2 (using the same active-space protocol as Fig. 4) for the pentagonal-ring cyclodehydrogenated isomer of the 3k dimer and for the mis-coupled majority-majority and minority-minority isomers, and compare their ground-state spins and lowest spin-excitation energies with the measured 14 meV step and Kondo feature. If a non-bipartite isomer also yields an open-shell ground state with a low-energy excitation near 14 meV, the AFM-based structural assignment becomes the sole discriminator, and the ferromagnetism claim should remain conditional; if its spectrum is clearly incompatible, the structural concern is mitigated. The same check should be repeated for the 5FM trimer against its 6 meV and 23 meV excitations.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—quintet (septet) ground states and 7 meV ferromagnetic exchange in 3k (5FM)—depends on the structural assignment of the coupled dimer/trimer as intact DBT units joined through majority-minority sublattice sites. The AFM images (Fig. 2f–h, S6) are the only direct evidence for this connectivity, and specifically for the absence of the pentagonal-ring cyclodehydrogenation that previously quenched spin in direct triangulene couplings (Fig. 1c, refs 21,22). The measured 14 meV inelastic step in 3k is assigned to a quintet-to-triplet transition, but a hypothetical pentagonal-ring isomer would be non-bipartite and could exhibit a low-energy spin excitation at a similar energy; the zero-bias Kondo feature indicates an open-shell ground state but does not by itself fix S=2. For 5FM, the same structural premise is load-bearing, and the 23 meV versus 17 meV mismatch for the second quintet already shows that the CASSCF-NEVPT2 assignment is not quantitatively secure, so the structural assignment cannot be relaxed. The paper's gulf-region argument is plausible but is validated only by qualitative AFM inspection, not by quantitative simulation of the AFM contrast or by computing the spectrum of the alternative cyclodehydrogenated structure. If a different coupling motif or a pentagonal ring had formed, the global sublattice imbalance would vanish and the ferromagnetism claim would collapse.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports the on-surface synthesis of dibenzotriangulene (DBT) dimers and trimers on Au(111) and their characterization by STM/AFM and inelastic electron tunneling spectroscopy. Depending on the regioselective coupling of the constituent DBT units, the dimers exhibit antiferromagnetic exchange (majority-majority and minority-minority couplings) or ferromagnetic exchange (majority-minority coupling), with the ferromagnetic dimer 3k showing a quintet ground state and an extracted intermolecular exchange of 7 meV. A majority-minority coupled trimer 5FM is shown to have a septet ground state. The authors argue that the benzo extension creates gulf regions that prevent the pentagonal-ring cyclodehydrogenation that previously quenched spin in direct triangulene couplings, providing a route to organic ferromagnetic quantum spin chains.","tokens_in":16811,"tokens_out":11176,"duration_ms":94120,"significance":"If the structural assignments hold, the paper provides a clear design principle for direct, spacer-free ferromagnetic coupling between organic radicals, with an exchange energy (7 meV) an order of magnitude larger than previous spacer-based ferromagnetic dimers. The combination of STM/AFM structural data, spin-excitation spectroscopy, and CASSCF-NEVPT2 benchmarks is a strength, as is the use of the Heisenberg dimer model to extract J_eff from the measured quintet-triplet gap. The demonstration of a septet trimer, even with the quantitative discrepancy noted below, is a step toward longer ferromagnetic spin chains. The principal risk is that the central ferromagnetism claim depends on the AFM-based exclusion of cyclodehydrogenated isomers, which is not yet quantitatively supported.","major_comments":[{"comment":"The quintet and septet ground states claimed for 3k and 5FM depend entirely on the assignment of these structures as intact DBT units coupled through majority-minority sublattice sites with intact gulf regions. The AFM images are the only direct evidence against the alternative cyclodehydrogenated isomer containing a pentagonal ring, which is the exact failure mode previously identified for direct triangulene couplings (refs. 21,22). The gulf-region argument is plausible but is validated only by qualitative inspection. I request that the authors either (i) simulate the AFM constant-height images for both the intact gulf structure and the pentagonal-ring isomer and compare them with experiment, or (ii) compute the electronic spectrum of the pentagonal-ring isomer and show explicitly that it cannot account for the observed 14 meV excitation and the S=2 Kondo signature. Without this, the central ferromagnetism claim rests on an unverified structural assumption.","section":"Fig. 2f-h and section \"On-surface synthesis of dibenzotriangulene dimers\""},{"comment":"The 23 meV spin excitation observed on the central DBT unit of 5FM is assigned to the second quintet state calculated at 17 meV, a discrepancy of 6 meV that is not discussed. This assignment is used to confirm the septet ground state and the spectral-weight pattern (Fig. 5e). The authors should either provide a quantitative estimate of the surface-induced renormalization for this trimer (e.g., following ref. 26) or perform a state-averaged CASSCF calculation with a larger active space to verify the second-quintet energy. As written, the mismatch leaves the trimer confirmation quantitatively insecure.","section":"Fig. 5c,d and section \"Magnetic excitations in ferromagnetic dibenzotriangulene trimers\""},{"comment":"The spectral weights shown in Fig. 5e are obtained from an exact diagonalization of a CAS-Hubbard model (CAS(6,6), t = -2.7 eV, U = |t|), whereas the energy levels and the 6 meV and 17 meV assignments come from CASSCF-NEVPT2. The relationship between these two models is not established; in particular, it is not shown that the CAS-Hubbard states at the energies of the first and second quintets have the same spatial character as the CASSCF-NEVPT2 states. The authors should clarify how the spectral-weight calculation relates to the CASSCF-NEVPT2 wavefunctions, or compute the spectral weights directly from the CASSCF states, to justify the spatial assignment of the 23 meV excitation.","section":"Fig. 5e and Methods (CASSCF calculations)"}],"minor_comments":[{"comment":"In the mean-field Hubbard Hamiltonian, the second sum contains operators a_i↑^† a_j↑ with an undefined index j; it should be a_i↑^† a_i↑ (i.e., the number operator), or the sum should be written accordingly. Please correct this typo.","section":"Methods, Eq. (1)"},{"comment":"The introduction states that the spacer-based ferromagnetic exchange is \"about 1 meV\", while the conclusion states \"< 1 meV\". Please make these two characterizations consistent.","section":"Abstract and Conclusion"},{"comment":"In the bullet-list comparison of experimental and calculated excitation energies, the calculated quintet-triplet gap for 3k is not given, while the values for 2k (35 meV) and 4k (2.8 meV) are. Please provide the calculated value for 3k to facilitate comparison with the measured 14 meV.","section":"Conclusions, item (2)"},{"comment":"The statistical distribution of the three dimer classes (6 majority-majority, 9 majority-minority, 11 minority-minority out of 26 dimers) is only reported in the Supporting Information. Mentioning this in the main text would strengthen the regioselectivity claim.","section":"Fig. S6 caption"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a strong candidate for publication in a high-impact venue, but the structural assignment of the ferromagnetic dimers/trimer is the key vulnerability. I recommend asking the authors to add explicit computational evidence against the pentagonal-ring isomer and to address the trimer excitation energy mismatch. These are within the scope of a major revision and would substantially strengthen the paper."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: this paper mostly delivers what it claims. The design idea is simple and new: use dibenzotriangulene instead of triangulene so that the benzo extension leaves a gulf region that blocks the pentagonal-ring cyclodehydrogenation, allowing direct majority-minority coupling without spin quenching. The result is a dimer with a quintet ground state and an intermolecular exchange of 7 meV, an order of magnitude larger than the spacer-based ferromagnetic dimers. The trimer data support a septet ground state. That is a real step for organic ferromagnetism.\n\nThe strongest part is the combination of experiment and theory. The spin excitation energies are measured independently of the CASSCF-NEVPT2 calculations, which are parameter-free many-body benchmarks rather than fits to the data. The S=2 Kondo feature and the dI/dV maps matching the mean-field Hubbard LDOS make the high-spin assignment credible. The authors also check that the Heisenberg dimer model is a valid description of 3k before extracting J_eff = 7 meV; that is the right way to do it.\n\nThe soft spots are real but modest. There are no error bars or per-molecule statistics on the reported excitation energies; the numbers come from single spectra. The trimer assignment of the 23 meV excitation to a calculated 17 meV quintet relies on a generic surface-renormalization argument, and a 6 meV shift is not independently backed. A second trimer or a surface-including calculation would help. The structural assignment is load-bearing, though I disagree with the stress-test note that this is a serious weakness: CO-tip AFM resolves bond structure, and the authors show consistent assignments across multiple dimers. Asking for AFM simulation of the hypothetical cyclodehydrogenated isomer is fair polish, but not a fatal flaw.\n\nWho this is for: anyone working on nanographene magnetism, on-surface synthesis of open-shell molecules, or quantum spin chains. It deserves a serious referee. I would send it out.","headline":"A credible and useful demonstration of direct majority-minority ferromagnetic coupling in DBT dimers and trimers, with a 7 meV exchange; the main soft spots are missing error bars and the loosely anchored 23 meV trimer assignment.","tokens_in":17436,"tokens_out":2749,"would_cite":true,"duration_ms":25975,"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 paper establishes that direct majority–minority coupling of dibenzotriangulene units on Au(111) yields ferromagnetic dimers and trimers, with a quintet ground state and 7 meV intermolecular exchange.","keywords":["on-surface synthesis","dibenzotriangulene","ferromagnetic exchange","quantum spin chains","sublattice imbalance","spin excitations","scanning tunneling microscopy","Ovchinnikov rule"],"falsifier":"An atomically resolved AFM image of 3k or 5FM showing a pentagonal ring or a different connectivity at the inter-unit bond, or a spin-excitation spectrum whose multiplet ordering and energies are inconsistent with a quintet or septet ground state, would refute the assignment.","tokens_in":16296,"feed_emoji":"🧲","tokens_out":5342,"duration_ms":44666,"temperature":0.7,"pith_summary":"The paper reports a synthetic route to short ferromagnetic spin chains built from dibenzotriangulene (DBT), a carbon molecule whose two unpaired electrons occupy one of the two interpenetrating sublattices of its bipartite lattice. The key step is joining DBT units directly through a bond between the majority sublattice of one unit and the minority sublattice of the next, so the combined structure keeps an excess of one sublattice and Ovchinnikov's rule predicts a high-spin ground state. The benzo extension creates 'gulf' regions that suppress the pentagonal-ring formation which previously quenched the spins in direct couplings of triangulenes. Scanning probe measurements and many-body calculations identify a quintet ($S=2$) ground state for the ferromagnetic dimer, with an intermolecular exchange of $J_{\\mathrm{eff}}=7$ meV, and a septet ($S=3$) ground state for the ferromagnetic trimer. This matters because direct coupling gives an exchange roughly an order of magnitude stronger than the spacer-based ferromagnetic dimers reported earlier, which reached only about 1 meV.","feed_headline":"Direct coupling makes carbon spin chains ferromagnetic","feed_subtitle":"Dibenzotriangulene dimers show a quintet ground state and 7 meV exchange, opening a route to organic magnets.","key_machinery":"The load-bearing object is the gulf region of the dibenzotriangulene unit: the benzo extension at the coupling site that sterically prevents the cyclodehydrogenative formation of a pentagonal ring between coupled units, which had previously destroyed the bipartite lattice and quenched the spins. The design principle is a direct (spacer-free) carbon–carbon bond connecting the majority sublattice of one DBT to the minority sublattice of the next, which makes the global sublattice imbalance $|N_A-N_B|$ grow with chain length and therefore, by Ovchinnikov's rule $S=|N_A-N_B|/2$, forces a ferromagnetic ground state. The paper combines nearest-neighbour tight-binding and mean-field Hubbard calculations, CASSCF-NEVPT2 many-body calculations, and scanning probe spectroscopy of inelastic spin excitations to assign the ground states and extract exchange energies.","core_discovery":"On the paper's own terms, the central discovery is that dibenzotriangulene (DBT), because of its benzo extension, permits direct majority–minority coupling between neighbouring units without the formation of pentagonal rings that break bipartite symmetry and quench the spins. For the majority–minority dimer 3k the ground state is a quintet ($S=2$), consistent with a global sublattice imbalance $|N_A-N_B|=4$, and the spin excitation at 14 meV is assigned to a quintet–triplet transition, giving an effective Heisenberg exchange $J_{\\mathrm{eff}}=7$ meV. For the majority–minority trimer 5FM the ground state is a septet ($S=3$), as Ovchinnikov's rule demands. The two other coupling classes, majority–majority (2k) and minority–minority (4k), have no global sublattice imbalance and are antiferromagnetic singlets, with singlet–triplet excitations at 19 meV and 2.5 meV respectively, demonstrating that the sign of the exchange is controlled by the sublattice coupling motif.","pith_inferences":["If the gulf-region protection is a general design rule rather than specific to DBT, other pi-extended triangulene derivatives could be coupled directly to build ferromagnetic chains with different spin sizes and exchange strengths.","The paper's calculations are for gas-phase molecules, and it notes that surface effects may renormalize the spin excitation energies; the 7 meV value is a surface-measured effective exchange, so the intrinsic molecular exchange could differ.","A natural next test is to grow longer DBT chains and look for the spin-wave-like excitations expected of a ferromagnetic Heisenberg chain, rather than only dimer and trimer transitions.","The same precursor chemistry might be combined with insulating surfaces to reduce Kondo screening and reveal the unscreened high-spin multiplets."],"forward_implications":["Ferromagnetic coupling between open-shell polycyclic hydrocarbons can be obtained without a spacer, giving an exchange of about 7 meV, roughly an order of magnitude larger than spacer-based ferromagnetic dimers with exchange below 1 meV.","The same majority–minority bond motif should extend to longer DBT chains, with the total spin growing by one per added unit, offering a route to organic ferromagnetic quantum spin chains.","The sublattice coupling motif (majority–majority, majority–minority, minority–minority) deterministically sets the sign of the exchange, giving a design rule for magnetic coupling in nanographenes.","The S=2 Kondo resonance observed in the ferromagnetic dimer and the spectral-weight distribution in the trimer show that these chains can be probed as spin systems at the single-molecule level."],"supporting_citations":[{"why":"Ovchinnikov's rule predicts ground-state spin from sublattice imbalance, the principle the synthesis is built on.","marker":"[2]"},{"why":"Theory of intermolecular exchange in coupled nanographenes is used to rationalize kinetic superexchange scaling with hybridization.","marker":"[10]"},{"why":"Prior triangulene dimers with spacers establish antiferromagnetic exchange and the Heisenberg-dimer extraction of J_eff.","marker":"[11]"},{"why":"Methods for spin excitation spectroscopy and spectral weights in nanographene spin chains, including fractional edge excitations, are extended here.","marker":"[12]"},{"why":"Previous ferromagnetic triangulene trimers formed with spacers provide the baseline the present work aims to improve on.","marker":"[18]"},{"why":"Recent ferromagnetically coupled spin-1 nanographenes with weak exchange supply the S=2 Kondo comparison.","marker":"[20]"},{"why":"Shows that pentagonal-ring formation quenches spins in direct triangulene coupling, the failure mode the gulf region avoids.","marker":"[21]"},{"why":"Another report of pentagonal ring incorporation destroying bipartite magnetism in cove-edged ribbons.","marker":"[22]"},{"why":"Spin selection rule ΔS=0,±1 is used to assign the observed inelastic spin excitations.","marker":"[24]"},{"why":"Theory of surface renormalization of spin excitations in open-shell nanographenes is invoked to explain experiment-theory discrepancies.","marker":"[26]"}],"fun_headline_variants":["Benzo twist yields organic ferromagnetic spin chains","Direct coupling makes carbon spins ferromagnetic","Sublattice mismatch turns carbon chains ferromagnetic","No spacer, no spin quenching: organic ferromagnets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claim rests on the structural assignment of the ferromagnetic dimers and trimer as majority–minority coupled DBT units with intact gulf regions and no pentagonal ring; if cyclodehydrogenation or a different coupling motif occurred, the sublattice imbalance would disappear and the ferromagnetic ground state would collapse.","fun_headline_variants_meta":{"raw":{"variants":["Benzo twist yields organic ferromagnetic spin chains","Direct coupling makes carbon spins ferromagnetic","Sublattice mismatch turns carbon chains ferromagnetic","No spacer, no spin quenching: organic ferromagnets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000723,"raw_usage":{"total_tokens":3330,"prompt_tokens":1117,"completion_tokens":2213,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":733,"completion_tokens_details":{"reasoning_tokens":2153}},"tokens_in":733,"tokens_out":2213,"duration_ms":16229,"temperature":1.0,"reasoning_tokens":2153,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T14:28:46.744155+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"An atomically resolved AFM image of 3k or 5FM showing a pentagonal ring or a different connectivity at the inter-unit bond, or a spin-excitation spectrum whose multiplet ordering and energies are inconsistent with a quintet or septet ground state, would refute the assignment.","supporting_citations":[],"review_version":1}