{"id":"8782b3df-0b7a-43e7-b62e-fdfb1119635f","arxiv_id":"2504.16810","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Dimer complexation energies from DFT, comparing parallel and antiparallel molecular pairs, correlate with the presence or absence of polar order in several closely related liquid crystal families.","lead":"This paper uses DFT calculations of pairs of molecules to explain why some liquid crystals form polar (ferroelectric nematic) phases and others do not. It argues that specific directional intermolecular interactions, not dipole strength alone, determine whether molecules align with parallel or antiparallel dipoles.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"DIO's computed antiparallel global minimum conflicts with the paper's own criterion; without a specified energy-gap threshold the framework rationalizes rather than predicts polar order.","rationale":"The reader's weakest assumption is that gas-phase rigid-monomer dimer energies transfer to condensed-phase ordering. I agree that this is fragile, especially because the decisive DIO gap is 0.1 kcal/mol. My stress-test identifies a more direct and internal problem: even taking the computed energies at face value, DIO's antiparallel global minimum does not match the stated criterion, so the framework's predictive content is underspecified. The RM734/RM734-CN comparison does show a clean reversal of global preference, which is the strongest evidence in the paper. For the DIO family, however, the only difference between polar and apolar materials is the size of the antiparallel stabilization, and no threshold is supplied. This is not an attack on the chemistry; the qualitative interaction motifs are plausible. It is a request for an explicit decision rule and a blind test. Because the reader already returned CONDITIONAL with a request for benchmarks, a blind prediction, and released inputs, my concern reinforces that verdict rather than changing it. A higher-level single-point benchmark on DIO is the most direct way to see whether the 0.1 kcal/mol result is even real; if it is, the paper needs a threshold or a different observable to be predictive.","tokens_in":9645,"tokens_out":4926,"duration_ms":46630,"concrete_test":"Perform single-point DLPNO-CCSD(T)/CBS or CCSD(T)-F12/aug-cc-pVDZ-F12 calculations on the optimized DIO parallel and antiparallel geometries from Figure 8, and on the CIO and DIO(-F) minima. If the DIO antiparallel preference remains larger than about 0.3 kcal/mol, then the paper's binary criterion would classify DIO as apolar, so the framework needs an explicit threshold or a different condensed-phase observable before it can predict polar order. If the gap reverses, the B97-D3 result is below method accuracy and the workflow must be rebenchmarked before use.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that relative complexation energies of parallel versus antiparallel dimer configurations determine whether a material exhibits polar order. The DIO result in Figure 8a contradicts this criterion as stated: the antiparallel dimer is the global minimum, lower by only 0.1 kcal/mol, yet DIO is an archetypal ferroelectric nematic. CIO and DIO(-F) are apolar and their antiparallel forms are 'far lower' in energy. The only distinction offered is a magnitude difference ('only 0.1' versus 'far lower'), but no threshold is defined or justified. Consequently, the workflow cannot currently predict polar order; it can rationalize known phase behavior after the fact. This internal tension is independent of the additional numerical problem that 0.1 kcal/mol is below the expected accuracy of B97-D3/cc-pVTZ. If the intended predictor is instead the stabilization of parallel configurations relative to a chemical analogue, that criterion is not stated and would need to be tested independently.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript introduces a computational workflow for probing the bimolecular potential energy surface (bPES) of liquid crystal-forming molecules, using rigid scans at the B97-D3/cc-pVTZ level with counterpoise correction followed by unconstrained optimization of selected minima. The authors apply this workflow to pairs of closely related materials: RM734 vs RM734-CN, DIO vs CIO vs DIO(-F), and the mixed RM734/DIO pair. Their central claim is that the presence or absence of the ferroelectric nematic (NF) phase can be accounted for by the relative complexation energies of parallel versus antiparallel dimer configurations, with polar order arising from specific directional non-covalent interactions rather than from generic dipole-dipole forces. They find that RM734 favors parallel geometries while RM734-CN favors antiparallel, that DIO favors antiparallel only marginally while CIO and DIO(-F) favor antiparallel strongly, and that RM734/DIO mixes ideally because their parallel interactions are complementary.","tokens_in":9859,"tokens_out":3025,"duration_ms":30094,"significance":"If validated, the proposed framework could provide a practically useful, chemistry-specific route to design new polar liquid crystals by computing dimer interaction energies, a field that currently lacks first-principles design rules. The work is commendable for using counterpoise-corrected DFT, for explicitly analyzing electrostatic potential and quadrupole anisotropies, and for addressing a set of experimentally well-characterized comparisons that present a challenging test of any model. The identification of specific interaction motifs (e.g., nitro-nitro and nitro-ester contacts in RM734, dioxane-mediated contacts in DIO) is insightful and may guide synthetic efforts. However, the paper's own DIO result contradicts the stated parallel-versus-antiparallel criterion unless an energy-gap threshold is introduced, and the decisive energy difference in that case (0.1 kcal/mol) is below typical DFT accuracy. As written, the workflow rationalizes known phase behavior rather than providing a validated prediction scheme. The strengths of the approach and the clarity of the presentation justify serious consideration, but the central predictive claim requires substantial additional support.","major_comments":[{"comment":"The paper's central criterion is that polar order is associated with a preference for parallel dimer configurations. Yet for DIO, the text states that the antiparallel form is the global energy minimum by only 0.1 kcal/mol, despite DIO being an archetypal NF material. This is a direct counterexample to the criterion as stated unless an energy-gap threshold is specified. The distinction between 'only 0.1 kcal/mol' for DIO and 'far lower' for CIO and DIO(-F) is qualitative; no numerical threshold, statistical confidence interval, or sensitivity analysis is provided. Consequently, the framework cannot classify DIO a priori, and the claim of prediction in the title and abstract is not supported. The authors should either define and justify a threshold, restate the criterion in a falsifiable form (e.g., relative to a chemical analogue), or explicitly reframe the work as providing post hoc rationalization rather than prediction.","section":"Results, DIO paragraph and Figure 8a"},{"comment":"The decisive DIO result rests on a 0.1 kcal/mol energy difference between antiparallel and parallel optimized dimers. This is below the expected accuracy of the B97-D3/cc-pVTZ method for noncovalent interactions, which is typically at least 0.5 kcal/mol for such systems. The manuscript provides no error bars, no benchmarks against higher-level wavefunction methods (e.g., CCSD(T)/CBS) or a reasonable DFT reference set, and no assessment of numerical noise (grid, convergence tolerances). Since the sign of the energy difference is the entire basis for classifying DIO versus CIO/DIO(-F), the conclusion is not robust. A higher-level single-point benchmark on the optimized minima, or at least a clear discussion of expected uncertainty, is needed.","section":"Methods and Figure 8a"},{"comment":"The workflow computes dimer complexation energies for rigid monomers in the gas phase at zero temperature, then equates these energy differences with the thermodynamic preference for parallel vs antiparallel order in the condensed phase. This neglects conformational flexibility, vibrational and configurational entropy, and many-body/polarization effects of the surrounding medium. These effects can plausibly alter the ordering of free-energy differences at the 0.1-1 kcal/mol scale that is decisive here. The authors should either test these assumptions (e.g., by harmonic free-energy corrections, ab initio molecular dynamics on selected dimers, or comparison with empirical force-field simulations of the bulk) or explicitly acknowledge that the framework is an approximate energetic surrogate and discuss how the neglected terms might affect the conclusions.","section":"Methods, rigid-monomer and gas-phase approximations"},{"comment":"The bPES is constructed using arbitrary numerical parameters: a 1 Å grid resolution, minimum and maximum interatomic separation cutoffs of 3.0 Å and 5.0 Å, selection of only 10 discrete minima for optimization, and an RMSD duplicate threshold of 0.2 Å. No sensitivity analysis is presented to show that the identified global minima and the relative stabilities of parallel versus antiparallel forms are robust to these choices. Because the entire argument depends on which minimum is the global one, the authors should test at least a subset of these parameters (e.g., finer grid, different cutoffs, more minima) and report how the main conclusions change.","section":"Methods, bPES construction and sampling"}],"minor_comments":[{"comment":"The similarity metric is referred to as 'Roger-Stanimo' in the caption; this appears to be a typo for 'Rogers-Tanimoto'.","section":"Figure 1 caption"},{"comment":"There are several typographical errors, e.g., 'antiparalell' in the Figure 8 caption and 'HCNF in 14 or NTBF in 15' in the introduction, where the citation formatting is unclear. A careful proofread is needed.","section":"Throughout"},{"comment":"The description of the transition temperature sources is terse; it would be helpful to list the exact phases and temperatures in a table for each material, since the experimental phase behavior is the ground truth against which the calculations are compared.","section":"Methods, first paragraph"},{"comment":"The phrase 'ΔG_complex' is used in the early part of the Results section, while the rest of the paper reports ΔE_cplx. The distinction between free energy and electronic energy should be clarified, as the calculations appear to report electronic energies only.","section":"Results, RM734/RM734-CN section"},{"comment":"The figure shows DIO(+F) and discusses its expected behavior, but DIO(+F) is not part of the computational study. The text should make clear whether these are literature results or predictions, and if predictions, they should be flagged as such.","section":"Figure 7"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a candidate for publication in a soft-matter journal if the authors address the predictive gap. The central problem is not the chemistry or the calculations themselves, but the mismatch between the title/abstract claim of 'predicting' and the actual evidence, which includes a direct counterexample (DIO) and a decisive energy difference below numerical reliability. I recommend requiring a revised version that either introduces a justified quantitative criterion with error bars and sensitivity analyses, or explicitly limits the claims to rationalization of known materials. I also note that parts of the framework were already described in the authors' JACS 2025 paper (ref. 47), so the novelty relative to that prior work should be clarified in the revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, here's my read of Hobbs et al.\n\nThe work applies the bPES workflow from their JACS paper to a handful of known NF-forming and non-forming materials. The new content is the application to RM734-CN, CIO, DIO(-F), and the RM734/DIO heterodimer, and the interaction-level stories are chemically plausible: nitro groups engaging in staggered electrostatics, dioxane's electrostatic complementarity, fluorination patterning that tunes ring quadrupoles. I think those qualitative motifs will be useful to people designing new materials.\n\nThe soft spot is the central claim. The paper's criterion is that the sign of the parallel/antiparallel complexation-energy difference predicts polar order. But DIO, a canonical ferroelectric nematic, has an antiparallel global minimum, lower by only 0.1 kcal/mol. The paper tries to rescue this by contrasting that with CIO/DIO(-F), where antiparallel is 'far lower,' but no threshold is specified. As stated, the criterion doesn't distinguish DIO from CIO; it only rationalizes after the fact. That's a real internal tension, independent of the numerical issue that 0.1 kcal/mol is below the expected accuracy of B97-D3/cc-pVTZ for noncovalent interactions. The paper lacks error bars or higher-level benchmarks. The rigid-monomer, gas-phase approximation also ignores entropy and condensed-phase effects—an approximation that's fine for hypothesis generation but too fragile for the predictive title.\n\nOn the positive side, the authors are frank in the conclusions that they are rationalizing, not predicting, and they don't oversell the dipole argument. The citation pattern looks fine; the workflow is self-cited but that's legitimate since they're building on their own method.\n\nWho is this for? Computational chemists and LC experimentalists working on polar nematics. It will be a useful reference for the interaction motifs, but I would not cite it as evidence for a predictive framework until the threshold question is answered and a genuine blind prediction is made.\n\nMy recommendation: send it to peer review. A good referee can push them to define the criterion more carefully, add benchmarks, and ideally test on a new compound. The qualitative chemistry is worth preserving, but the current version's claim is too strong for its evidence.","headline":"Plausible interaction motifs for polar order, but the predictive claim is undercut by the DIO anomaly and missing error bars; deserves a rigorous revision.","tokens_in":74,"tokens_out":2243,"would_cite":false,"duration_ms":40868,"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":"The presence or absence of the ferroelectric nematic phase in closely related liquid crystals is decided by the relative stability of parallel versus antiparallel molecular pairs, not by dipole-moment magnitude alone.","keywords":["polar liquid crystals","ferroelectric nematic","bimolecular potential energy surface","intermolecular interactions","complexation energy","density functional theory","electrostatic potential","molecular design"],"falsifier":"Recompute the decisive dimers with a higher-level electronic-structure method or an explicit condensed-phase free-energy simulation: if the antiparallel form of DIO remains clearly more stable while DIO still shows the $N_F$ phase, or if the parallel form of RM734-CN becomes clearly more stable while RM734-CN still lacks it, the central claim is contradicted. A synthetic test would be to make a DIO analogue whose electrostatic potential the framework predicts to restore the parallel preference and show that it still fails to form a polar phase.","tokens_in":9465,"feed_emoji":"🧲","tokens_out":8987,"duration_ms":77312,"temperature":0.7,"pith_summary":"This paper argues that whether a liquid crystal develops polar order—its molecular dipoles all pointing the same way—can be read off from how two of its molecules prefer to pack together. In the closely related pair RM734, which shows the ferroelectric nematic phase, and RM734-CN, which does not, the framework traces the difference to specific lateral contacts: the nitro group stabilizes parallel pairs, while the nitrile group opens a deep antiparallel minimum. The same dimer-energy comparison accounts for why DIO is polar while its near relatives CIO and DIO(-F) are not. If this picture is right, dipole-dipole forces are not the primary cause of spontaneous polar order; instead, subtle directional non-covalent interactions between specific chemical groups decide, and candidate materials could be screened from gas-phase dimer calculations before synthesis.","feed_headline":"Two-molecule geometry decides which liquid crystals go polar","feed_subtitle":"Parallel-versus-antiparallel dimer energies separate polar materials from non-polar lookalikes.","key_machinery":"The central object is the bimolecular potential energy surface (bPES): a second copy of the optimized monomer is translated over a three-dimensional grid in both parallel and antiparallel orientations, and every physically valid configuration is assigned a counterpoise-corrected DFT single-point energy (B97-D3/cc-pVTZ). Discrete minima are then freely optimized, duplicate geometries are removed by RMSD comparison, and surviving minima are classified into pairing modes such as twisted parallel, slipped parallel, and twisted antiparallel. The argument is carried by comparing the complexation energies $\\Delta E_{\\rm cplx}$ of parallel versus antiparallel minima, with electrostatic potential maps and traceless quadrupole tensor components used to explain why specific contact geometries are stabilizing or repulsive. The $N_F$ phase is treated not as a property of one molecule but as a competition between these lateral interaction motifs.","core_discovery":"On the paper's own terms, the central discovery is that a rigid bimolecular potential energy surface—scanned in parallel and antiparallel orientations and then relaxed at the minima—reproduces the experimentally known presence or absence of the $N_F$ phase in three closely related families. For RM734 the global minimum is a twisted parallel pair stabilized by contacts between nitro and ester groups; for RM734-CN the global minimum is a twisted antiparallel pair that the nitro-terminated material cannot adopt for electrostatic reasons. For DIO the parallel and antiparallel forms are almost exactly balanced, with the antiparallel form lower by only 0.1 kcal/mol, and the paper argues that this near-degeneracy, rather than a clear parallel win, is the signature of a material that still becomes polar. In CIO and DIO(-F), by contrast, the antiparallel form is far lower in energy, matching the absence of polar order. The authors conclude that specific molecular features and strong directional intermolecular interactions, not dipole-dipole forces, establish polar order.","pith_inferences":["Beyond the paper, the near-degeneracy of DIO suggests that the framework's real predictive content may be the balance of contact energies rather than a hard threshold; a natural next test is whether higher-level wavefunction calculations preserve the 0.1 kcal/mol ordering.","The gas-phase dimer picture omits entropy and many-body packing; extending it to finite-temperature dimer free energies or polarizable molecular dynamics would test whether the predicted parallel preference survives in the actual fluid.","If the contact-motif picture holds, the same logic could be exported to other polar soft matter—organic ferroelectrics, polar molecular crystals, or surface monolayers—where local pairwise contacts rather than isolated dipole moments decide the emergent polarity.","One testable extension is to use the dimer preference to predict $T(N_F)$ itself: systems with a larger parallel-favouring $\\Delta E_{\\rm cplx}$ should show higher onset temperatures, as the DIO(+F) trend already hints."],"forward_implications":["A candidate material's tendency to form a polar phase can be screened from dimer calculations before synthesis, using the sign and size of $\\Delta E_{\\rm cplx}$ for parallel versus antiparallel pairs.","Chemical groups with spatially extended charge distributions—nitro, dioxane, optimally placed fluorines—act as polarity-promoting contact motifs; replacing them with nitrile, cyclohexane, or a removed fluorine should usually suppress the $N_F$ phase, matching experiment.","Fluorination pattern controls polar order through electrostatic potential anisotropy and quadrupole sign changes, giving a concrete recipe for tuning $T(N_F)$ by editing ring fluorination.","Mixtures lose polar order when the components lack complementary lateral interactions, explaining the rapid suppression of $N_F$ observed in blends such as 5CB with DIO or RM734."],"supporting_citations":[{"why":"Supplies the rigid bimolecular PES workflow and software tool that this study applies to polar liquid crystals.","marker":"ref. 47"},{"why":"Provide the experimental phase behavior of RM734 and RM734-CN that the framework must reproduce.","marker":"refs. 20, 21"},{"why":"Reports DIO and its ferroelectric nematic phase, the central DIO-family target for prediction.","marker":"ref. 22"},{"why":"Reports DIO(-F) and DIO(+F) phase behavior and the fluorination/ESP trends used in the quadrupole analysis.","marker":"ref. 42"},{"why":"Reports CIO and related dioxane analogues whose apolar phases are explained by weakened parallel interactions.","marker":"ref. 43"},{"why":"Provides the experimental ideal-mixing result for RM734 and DIO that motivates the heterodimer analysis.","marker":"ref. 59"},{"why":"Earlier demonstration that spatially uniform electrostatic potential promotes lateral interactions, used to interpret fluorination effects.","marker":"ref. 16"},{"why":"Identify the electronic-structure package in which all DFT calculations were performed.","marker":"refs. 55, 56"},{"why":"Defines the B97-D3 dispersion-corrected functional used for all energies.","marker":"ref. 57"},{"why":"Defines the cc-pVTZ basis set used for all optimizations and single-point calculations.","marker":"ref. 58"}],"fun_headline_variants":["Dimer geometry predicts polar liquid crystal order","Parallel vs antiparallel pairs: key to polar phases","Nitro-ester contacts drive polar liquid crystals","Near-equal dimer energies still allow polarity","Molecular pair scan reveals polar order rules"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The controlling premise is that gas-phase dimer complexation energies, computed for rigid optimized monomers with B97-D3/cc-pVTZ, reliably predict which condensed-phase liquid-crystal ordering wins—even when the deciding energy gap is 0.1 kcal/mol.","fun_headline_variants_meta":{"raw":{"variants":["Dimer geometry predicts polar liquid crystal order","Parallel vs antiparallel pairs: key to polar phases","Nitro-ester contacts drive polar liquid crystals","Near-equal dimer energies still allow polarity","Molecular pair scan reveals polar order rules"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00024,"raw_usage":{"total_tokens":1507,"prompt_tokens":925,"completion_tokens":582,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":541,"completion_tokens_details":{"reasoning_tokens":513}},"tokens_in":541,"tokens_out":582,"duration_ms":5624,"temperature":1.0,"reasoning_tokens":513,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T10:54:15.236062+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the decisive dimers with a higher-level electronic-structure method or an explicit condensed-phase free-energy simulation: if the antiparallel form of DIO remains clearly more stable while DIO still shows the $N_F$ phase, or if the parallel form of RM734-CN becomes clearly more stable while RM734-CN still lacks it, the central claim is contradicted. A synthetic test would be to make a DIO analogue whose electrostatic potential the framework predicts to restore the parallel preference and show that it still fails to form a polar phase.","supporting_citations":[],"review_version":1}