{"id":"865963f7-c673-44e9-9e6b-843984d5c375","arxiv_id":"2505.12164","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Dilute polymer chains in a narrow tube show single-file diffusion that ends either in collective Rouse motion or in a chain-switching regime, depending on tube width and chain length.","lead":"Simulations of flexible polymer chains in narrow tubes show that their diffusion follows a sequence of regimes, ending either in collective motion or in a faster switching regime when chains can pass each other. The authors map these regimes into a phase diagram and build a fast lattice model that reproduces the same behavior.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Switching-regime timescale is taken from the first switch anywhere among N chains, with no-switch runs censored; this biases the quantitative Ds(h) prediction in Fig. 6 and shifts the phase boundary.","rationale":"The paper's central claim is the existence of two routes to normal long-time diffusion: the cooperative Rouse regime (DR = Do/N) and the switching regime (Ds > DR). The qualitative content of this claim is well supported by the LD MSD curves, the collapse in Fig. 4, and the independent MC results in Fig. 11, which directly show D scaling as sqrt(PQ) with a corresponding ts ~ 1/(PQ). The load-bearing weakness I find is in the quantitative prediction of the switching regime in the LD part: the theory's ts is a per-particle inter-switch time, but the quantity measured and inserted into Eq. 26 is the first system-wide switch time, averaged only over runs that exhibited a switch. This is not merely a fitting detail: the system-first-event time is an order-statistic that decreases with N, and the censoring of no-switch runs biases it downward exactly in the regime where switching is rare. The result is an overestimate of Ds(h) and an uncertain location of the Rouse-to-switching boundary in the M-h phase diagram. The reader's weakest_assumption identified the same core proxy problem; I sharpen it with the explicit N-scaling and censoring mechanisms and propose a direct steady-state measurement to settle it. Because the existence of the switching pathway is corroborated by the MC model and the paper is presented as a phase diagram with acknowledged approximations, the appropriate verdict remains CONDITIONAL rather than ACCEPT or REJECT. A targeted correction to how ts is estimated would strengthen the quantitative claim.","tokens_in":15106,"tokens_out":11576,"duration_ms":119850,"concrete_test":"Rerun the LD switching-time analysis for N = 3, 5, and 10 at a few [M, h] points on both sides of the switching boundary (e.g., M = 8 and M = 10, h = 5.5, 6, 7.5), recording every switch per tagged chain rather than only the first system-wide event, and treating runs without any switch as right-censored in a survival analysis. Compute the steady-state per-chain switch rate lambda from the slope of cumulative switches versus time and set ts = 1/lambda. Re-evaluate Eq. 26 and the dashed line in Fig. 6 with this corrected ts, and compare it to the measured asymptotic diffusion coefficients. If the corrected Ds differs from the published curve by more than ~30%, or if <t1> from the original protocol scales roughly as 1/N when N is varied while keeping local density fixed, the proxy is responsible for a quantitative bias in the switching-regime prediction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The theory in Sec. II A defines ts (Eq. 10) as the mean time between switching events involving a given particle, but Sec. IV C measures t1 as the time of the first chain-switch anywhere in the system of N=5 chains, and Eq. 26 then replaces ts with this system-wide first-switching time. Two biases follow. First, t1 is an extreme statistic: if each of the N gaps switches as a Poisson process with rate lambda, the system-first-event time has mean 1/(N*lambda), while a tagged particle is involved in two gaps and has mean per-particle switching time 1/(2*lambda). Thus <t1> ~ (2/N)*ts = 0.4*ts for N=5, so Eq. 12/26 overestimates Ds by sqrt(N/2) ~ 1.58 and artificially introduces an N-dependence that the paper argues should not exist. Second, the average is conditional on observing at least one switch; runs without switching are discarded (explicitly noted in Sec. IV C). Near the Rouse-to-switching boundary, most runs may be censored, so <t1> is biased downward, inflating the apparent switching rate and moving the predicted switching regime to smaller h. Because Eq. 13 and Eq. 26 quantify the second pathway to normal long-time diffusion (the switching regime), this proxy directly affects the quantitative content of the phase diagram in Fig. 6 and the placement of the corresponding regime boundary, even though the MC results in Fig. 11 independently support the scaling D ~ sqrt(PQ).","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the crossover from single-file to normal diffusion for dilute flexible chains in a narrow toroidal tube using Langevin dynamics (LD) and a two-parameter lattice Monte Carlo (LMC) model. The theory predicts the sequence Do -> F t^{1/2} -> DR = Do/N for hard-core particles, and additionally a switching regime with Ds > DR when neighboring chains exchange positions. LD simulations confirm the Do -> F -> DR sequence through the rescaling collapses in Fig. 4 and measure an SFD exponent 0.52(1); the MC model reproduces the same phenomenology and shows D/Do proportional to sqrt(PQ) in the switching regime. The paper proposes M-h and P-Q phase diagrams and a heuristic mapping between the two models.","tokens_in":15590,"tokens_out":13690,"duration_ms":129386,"significance":"The paper addresses a timely and experimentally relevant question, and its core scaling scenario is well supported by internal evidence: the predicted transition times in Eqs. (21)-(22) produce good collapses in Fig. 4, the measured SFD exponent is consistent with 1/2, the displacement distributions are Gaussian, and the MC result D/Do proportional to sqrt(PQ) with no N-dependence (Fig. 11) is a clear testable prediction. The two-parameter MC model is economical, and the phase diagrams in Fig. 9 are informative. If the switching-regime analysis is made quantitative, the paper would provide a useful framework for nanochannel experiments. The main weakness is that the LD switching-regime prediction currently rests on a biased first-passage statistic.","major_comments":[{"comment":"The predicted switching-regime diffusion coefficient is built from a censored, system-wide first-passage statistic rather than the per-particle switching time defined in Sec. II A. In Sec. II A, ts is the mean time between switching events involving a given particle (text after Eq. 10), but Sec. IV C records t1, the time of the first chain-switch anywhere among N=5 chains, and runs with no switch are discarded (as noted in the text). Replacing ts by <t1> in Eq. 26 is biased in two ways: for Poisson switching with rate lambda per gap, <t1> is approximately 1/(N lambda) = (2/N) ts, so for N=5 Eq. 26 overestimates Ds by sqrt(N/2) ~ 1.58 and introduces an N-dependence that Eq. 12 asserts should not exist; and censoring no-switch runs biases <t1> downward close to the Rouse-to-switching boundary, shifting the onset of the switching regime to smaller h. Because <t1> is also fitted by Eq. 25, the dashed curve in Fig. 6 is not an independent quantitative prediction. The existence of a switching branch remains plausible from the MC data, but a direct per-particle switching-rate measurement or an extreme-statistics correction is required for the LD prediction.","section":"Sec. IV C, Eqs. (25)-(26), Fig. 6"},{"comment":"The wide-tube boundary of the M-h phase diagram is set by h = 5 Rg(M), with the factor of 5 stated to be arbitrary. This line is used to separate the intermediate regime from the free-diffusion regime, so the quantitative placement of the switching region on the phase diagram is not derived. If the phase diagram is to be a predictive map, this criterion should be replaced by a measured or derived condition (for example, to ~ ts) or explicitly labelled as illustrative rather than theoretical.","section":"Sec. IV D, Fig. 7(a)"},{"comment":"The proposed mapping between the LD parameters (M,h) and the MC parameters (P,Q) is not validated. Equation (28) is assembled from the postulated proportionalities td ~ 1/Q, M h^{-1/nu} ~ 1/(PQ), and tau_c ~ M; the text acknowledges that the first two are assumptions. The visual similarity of Fig. 12 is not by itself evidence for the mapping, since both phase diagrams are organized by similarly monotone axes. A direct comparison of LD and MC at matched parameter values, or at least a sensitivity analysis of the assumed proportionalities, is needed before the mapping can be presented as an established connection.","section":"Sec. V, Eq. (28), Fig. 12"}],"minor_comments":[{"comment":"The caption of Fig. 6 states ell_o = 11 for M = 10, while the text and Eq. (26) use ell_o = 13 (= M + 3). Please reconcile; if the data were taken at ell_o = 11, the dashed line is evaluated at the wrong spacing.","section":"Fig. 6 caption vs Sec. IV C"},{"comment":"The estimate <t1>_infinity approximately ell_o^2/(2 Do) should be reconciled with Eq. (2), which gives t_o ~ ell^2/(8 Do) for the same closing process; the factor of four and the use of ell_o versus ell = ell_o - Sx should be clarified.","section":"Sec. IV C, Eq. (23)"},{"comment":"Please report the fraction of runs with no switching event for each (M,h) condition; this quantifies the censoring of <t1> and lets the reader judge where the average is reliable.","section":"Sec. IV C, Fig. 5(c)"},{"comment":"The dashed lines in Fig. 9 are derived from the linear fit in Fig. 11; the text should state explicitly that these are empirical boundaries fitted to the same data rather than independent theoretical predictions.","section":"Sec. IV E, Fig. 9"},{"comment":"Writing M ~ 1/QP^{1/6} is ambiguous; M ~ 1/(Q P^{1/6}) would avoid misreading, and the choice nu = 3/5 here versus nu ~ 0.68 elsewhere should be justified.","section":"Sec. V, Eq. (28)"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope and the core results are publishable after the switching-regime analysis is fixed. I would not reject on the basis of the time-scale proxy because the MC scaling and the Do-F-DR sequence are independently supported. The main concern is that the authors not oversell the LD-MC mapping, since Eq. (28) rests on untested proportionalities."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper gives a clean Langevin-dynamics demonstration of the Do → t^{1/2} → DR sequence for Rouse chains in a toroidal tube, and the phase-diagram idea is genuinely useful. The new content is the claim that there are two routes to normal long-time diffusion: cooperative Rouse motion when chains cannot pass, and a switching regime when they can. That distinction is physically sensible, and the lattice Monte Carlo model backs it with an independent D ~ sqrt(PQ) scaling that does not depend on the LD fitting.\n\nTheMC model is the most valuable part. Two parameters, P and Q, control overlap and disentanglement, and it reproduces the three regimes with a clean collapse in Fig. 8 and a linear D/Do = 1.50√PQ fit in Fig. 11. The central sequence is well supported by the LD data: the collapse in Fig. 4, the exponent 0.52(1), and the Gaussian displacement distributions. Credit is due for that.\n\nThe soft spots are real. In Sec. IV C, the switching-regime prediction replaces the per-particle switching time ts with <t1>, the time of the first switch anywhere among N=5 chains. That is an extreme statistic: the first of several gaps will switch roughly N times faster than a given particle's next event, so <t1> is a biased proxy. The censoring of no-switch runs makes it worse: near the Rouse-to-switching boundary, runs without any switch are discarded, so the average is biased downward and the apparent switching rate is inflated. The paper acknowledges approximations, but this is not a minor correction—it shifts the quantitative Ds(h) curve in Fig. 6 and the location of the switching boundary. The MC result D ~ sqrt(PQ) is independent and stands. The phase boundary in Fig. 7 also uses an arbitrary factor of 5, which the authors admit. So the phase diagram is a good map of regimes, but the switching-regime boundary should be read as qualitative, not quantitative.\n\nThe LD-to-MC mapping in Eq. 28 is heuristic, but the authors say so explicitly; that is acceptable if presented as a guide rather than a derivation. The citation pattern looks fine, including the authors' own prior work on Rgx scaling, which is relevant and not self-serving.\n\nThis paper is for people working on nanochannel DNA, polymer single-file diffusion, and coarse-grained MC methods for confined polymers. It deserves a serious referee. The switching-time statistic should be fixed in revision: use a tagged-particle switching rate, or at least correct for N and censoring, and report the fraction of runs with no switch. Even without that correction, the central physics and the MC results are solid enough to merit publication after revision.","headline":"A useful polymer phase diagram with a clean LD confirmation of Do→t^1/2→DR, but the switching-regime prediction is biased by a first-switch-time proxy and censored runs.","tokens_in":16022,"tokens_out":2482,"would_cite":true,"duration_ms":25589,"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":"Dilute polymer chains in a narrow tube move freely, then in single file, then either cooperatively or by swapping, and the paper maps which route wins as chain length and tube width change.","keywords":["single-file diffusion","polymer dynamics","Rouse model","Langevin dynamics","lattice Monte Carlo","nanochannel confinement","phase diagram","entropic barrier"],"falsifier":"Measure the full waiting-time distribution between successive chain-swap events in the Langevin-dynamics model over times much longer than the current runs, without discarding no-swap trajectories; if the steady-state mean inter-swap time differs from $\\langle t_1\\rangle$ beyond statistical error, the estimate of $D_s(h)$ and the switching-regime boundaries need revision. A complementary check would track labeled DNA molecules in nanochannels and test whether the asymptotic diffusivity is independent of the number of chains $N$ when swapping dominates.","tokens_in":1995,"feed_emoji":"🧪","tokens_out":6326,"duration_ms":110660,"temperature":0.7,"pith_summary":"This paper tries to establish that a dilute solution of flexible polymer chains confined in a narrow toroidal tube passes through a universal sequence of diffusion regimes, and that the long-time outcome has two distinct routes. First the chains diffuse freely, then they enter single-file diffusion with mean-square displacement growing as $t^{1/2}$; at long times either the whole chain population moves cooperatively with diffusion coefficient $D_o/N$, or, if the chains can swap positions, a switching regime appears with a larger coefficient $D_s$. The authors identify molecular weight $M$, tube width $h$, and chain spacing as the control parameters, and summarize the behavior in a phase diagram based on the intermediate-time exponent $\\alpha$. They also build a one-dimensional lattice Monte Carlo model that reproduces the same physics with two parameters, one for chain overlap and one for disentanglement. Knowing which route dominates matters for nanochannel experiments, DNA mapping, and one-dimensional electrophoresis, where molecular weight and channel width are tunable.","feed_headline":"Two escape routes from single-file diffusion in narrow tubes","feed_subtitle":"Map shows when chain-swapping beats cooperative motion as tube width grows.","key_machinery":"The central object is the switching time $t_s$, the mean interval between chain-exchange events, and its ratio to the cooperative time $t_\\lambda$. The load-bearing identity is Eq. 12, $D_s = \\sqrt{8t_o/(\\pi t_s)}\\,D_o = \\sqrt{t_\\lambda/t_s}\\,D_R$, which converts a local event rate into a macroscopic diffusion coefficient. The paper estimates $t_s$ from the mean first-switching time $\\langle t_1\\rangle$, whose empirical fit grows exponentially as the effective tube width $h-2$ shrinks, because chain exchange requires crossing an entropic barrier set by the number of blobs $M/g(h) \\sim M/(h-2)^{3/2}$. The lattice Monte Carlo model carries the same mechanism through two parameters, $P$ for overlap and $Q$ for disentanglement, giving $t_s \\sim 1/(PQ)$ and hence $D \\propto \\sqrt{PQ}$, which matches the long-time data.","core_discovery":"The central claim is that polymer chains in narrow channels behave differently from hard particles: they can exchange order even when the channel is narrower than twice their size, because two chains can entangle, co-diffuse, and then disentangle. This single fact creates two different pathways to normal long-time diffusion. If the time $t_s$ between switching events exceeds the cooperative time $t_\\lambda$, the system follows the hard-particle sequence $D_o \\to F t^{1/2} \\to D_R = D_o/N$. If $t_o < t_s < t_\\lambda$, the chains enter a switching regime with $D_s = F/t_s^{1/2} = \\sqrt{8t_o/(\\pi t_s)}\\,D_o = \\sqrt{t_\\lambda/t_s}\\,D_R$, so that $D_s > D_R$ and the Rouse regime disappears. Using Langevin dynamics of Rouse chains, the paper measures the mean first-switching time $\\langle t_1\\rangle$, fits its exponential divergence with $(h-2)^{-1.4}$, and uses it to estimate $D_s(h)$. A lattice Monte Carlo model with overlap probability $P$ and disentanglement probability $Q$ reproduces the phase diagram and shows $D/D_o \\propto \\sqrt{PQ}$, independent of the number of chains $N$, exactly as the switching-time formula predicts.","pith_inferences":["Because the paper measures the first-switching time only over runs where a swap occurs, a renewal-theory estimate using the full steady-state waiting-time distribution between swaps could replace the empirical fit and would be a natural follow-up.","The two-pathway picture suggests a design rule for nanochannel separations: in the switching window, $\\langle t_1\\rangle \\sim M^3$ makes the long-time mobility strongly molecular-weight dependent, which could be exploited to sort chains by size.","For polydisperse or bimodal mixtures, chains of different lengths have different switching times, so the model points toward single-file-triggered segregation or local ordering along the tube, an extension the paper mentions but does not develop.","The mapping between the two simulation models relies on the scaling $t_d \\sim M^2 h^{2-1/\\nu}$ for the disentanglement time; direct Langevin-dynamics measurements of disentanglement times would test whether that scaling actually holds for the bead-spring model."],"forward_implications":["The intermediate-time exponent $\\alpha$ in $\\langle x^2\\rangle \\sim t^\\alpha$ serves as a single scalar order parameter for the phase diagram: values near $1/2$ mark strong single-file behavior, values near $1$ mark free or wide-tube diffusion, and intermediate values mark the coexistence zone.","When chain switching dominates, the long-time diffusion coefficient $D_s$ is predicted to be independent of the number of chains $N$, unlike the cooperative Rouse value $D_R = D_o/N$; the Monte Carlo data confirm this $N$-independence.","The switching time diverges exponentially as the effective tube width $h-2$ decreases, so single-file behavior is robust for chains that form many blobs, and the switching regime appears only in a relatively narrow range of tube widths.","The lattice Monte Carlo model with parameters $P$ and $Q$ reproduces the Langevin-dynamics phase diagram, and the proposed mappings $h \\sim \\sqrt{P}$ and $M \\sim 1/(Q P^{1/6})$ suggest the two simulation approaches describe the same physical regimes.","For a fixed molecular weight, increasing the tube width moves the system from the cooperative Rouse regime through the switching regime to free diffusion, so the asymptotic diffusion coefficient rises monotonically from $D_o/N$ to $D_o$."],"supporting_citations":[{"why":"Establishes the crossover from single-file to Fickian diffusion when particles can switch positions, the basic scenario the polymer case generalizes.","marker":"[6]"},{"why":"Provides the time scales and scaling laws for single-file dynamics of colloids in circular channels, supporting the cooperative Rouse regime and $t_\\lambda$.","marker":"[8]"},{"why":"Gives the mathematical single-file result $\\langle x^2\\rangle \\sim t^{1/2}$ that underlies the intermediate SFD regime.","marker":"[11]"},{"why":"Supports the non-Fickian single-file behavior and the form of the SFD mobility $F$.","marker":"[12]"},{"why":"Supplies the Rouse-chain result $D_o(M) = D_1/M$, used for free diffusion and for the cooperative limit $D_R = D_o/N$.","marker":"[13]"},{"why":"Provides the blob model for chains confined in small pores, used to estimate the number of blobs and the entropic barrier controlling chain switching.","marker":"[21]"},{"why":"Gives the axial radius-of-gyration scaling and the finite-size Flory exponent for this specific Langevin-dynamics model, used in the span and switching-time estimates.","marker":"[22]"},{"why":"Provides the disentanglement-time scaling $t_d \\sim M^2 h^{2-1/\\nu}$ used to connect the Langevin-dynamics and Monte Carlo parameters.","marker":"[23]"}],"fun_headline_variants":["Polymers swap order to break single-file diffusion","Two routes to normal diffusion in polymer tubes","Phase diagram maps two escapes from single-file diffusion","Chain swapping creates a second diffusion regime"],"cache_read_input_tokens":18048,"weakest_assumption_plain":"The load-bearing premise is that the average time to the first chain-swap event, measured only over simulation runs where a swap actually occurred, equals the steady-state switching timescale that controls long-time diffusion; if runs without swaps are censored or rare first events skew the mean, the predicted switching diffusivity and phase boundaries shift.","fun_headline_variants_meta":{"raw":{"variants":["Polymers swap order to break single-file diffusion","Two routes to normal diffusion in polymer tubes","Phase diagram maps two escapes from single-file diffusion","Chain swapping creates a second diffusion regime"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000759,"raw_usage":{"total_tokens":3388,"prompt_tokens":979,"completion_tokens":2409,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":595,"completion_tokens_details":{"reasoning_tokens":2353}},"tokens_in":595,"tokens_out":2409,"duration_ms":21403,"temperature":1.0,"reasoning_tokens":2353,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T20:39:27.242778+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the full waiting-time distribution between successive chain-swap events in the Langevin-dynamics model over times much longer than the current runs, without discarding no-swap trajectories; if the steady-state mean inter-swap time differs from $\\langle t_1\\rangle$ beyond statistical error, the estimate of $D_s(h)$ and the switching-regime boundaries need revision. A complementary check would track labeled DNA molecules in nanochannels and test whether the asymptotic diffusivity is independent of the number of chains $N$ when swapping dominates.","supporting_citations":[{"cited_title":"Sané , author J","cited_arxiv_id":null,"evidence_quote":"Establishes the crossover from single-file to Fickian diffusion when particles can switch positions, the basic scenario the polymer case generalizes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supports the non-Fickian single-file behavior and the form of the SFD mobility $F$."},{"cited_title":"Doi \\ and\\ author S","cited_arxiv_id":null,"evidence_quote":"Supplies the Rouse-chain result $D_o(M) = D_1/M$, used for free diffusion and for the cooperative limit $D_R = D_o/N$."},{"cited_title":"Daoud \\ and\\ author P","cited_arxiv_id":null,"evidence_quote":"Provides the blob model for chains confined in small pores, used to estimate the number of blobs and the entropic barrier controlling chain switching."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the axial radius-of-gyration scaling and the finite-size Flory exponent for this specific Langevin-dynamics model, used in the span and switching-time estimates."},{"cited_title":"Arnold \\ and\\ author S","cited_arxiv_id":null,"evidence_quote":"Provides the disentanglement-time scaling $t_d \\sim M^2 h^{2-1/\\nu}$ used to connect the Langevin-dynamics and Monte Carlo parameters."}],"review_version":1}