{"id":"70f74adb-a066-4d7b-9c41-afd47e21ca55","arxiv_id":"1908.01972","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Using renewal theory inside a CTRW description, the authors extend the MCT+RFOT Unified theory so that slow activated relaxation and fast MCT diffusion decouple, predicting the Stokes-Einstein breakdown.","lead":"This paper shows why a popular theory of supercooled liquids fails to predict the Stokes-Einstein breakdown, and then modifies the theory using random-walk mathematics to make the breakdown appear. The modified theory says structural relaxation is controlled by slow activated jumps while diffusion remains fast, which matches experiments on the liquid Salol.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The SE breakdown in Fig. 2 rests on an unsupported substitution of a single activated exponential for the persistence function of the two-exponential renewal process; the correct survival function is bi-exponential and the diffusive part already contains activated dynamics.","rationale":"The paper's central claim is that the extended Unified theory (Eq.24) predicts the Stokes-Einstein breakdown in Salol with the right magnitude. The numerical result in Fig.2 is real given the equations, but the equations themselves are where the argument is weakest. The reader's verdict correctly identified the mapping in Sec. II D as load-bearing. My stress-test sharpens this: the mapping is not merely an unproven identification; it is inconsistent with the renewal theory used to motivate it. The two-exponential mixture in Eq.20 has a survival function that is a weighted sum of exponentials, and the long-time tail carries a prefactor that is not reproduced by Eq.23. More importantly, the renewal equation's diffusive part φ_diff is set equal to the full Unified self-correlation φ_s, which contains the same activated hopping process that P(t) is supposed to represent. This double-counting means Eq.24 is not a CTRW/renewal model with two waiting-time distributions; it is an ad hoc construct that guarantees a slow structural relaxation by putting an activated exponential in the persistence slot and leaves the MSD unchanged. The equal-timescale limit τ1=τ2 shows the construction does not reduce to the original process, confirming that the slow process is inserted rather than derived. The paper does provide a correct statement that for two exponentials the mean residual time exceeds the mean interarrival time, and the observation that the original Unified theory fails to give SE breakdown is a useful negative result. But those do not establish the positive prediction. If the proposed concrete test shows that using the correct survival function preserves Fig.2, then the conditional acceptance would be justified; if not, the central claim should be considered unverified. Since the issue is precisely the one the reader flagged, I would keep the CONDITIONAL verdict but make the condition explicit: derive Eq.23-24 from a self-consistent CTRW model or test the bi-exponential persistence function.","tokens_in":16665,"tokens_out":13242,"duration_ms":143677,"concrete_test":"Reproduce Fig. 2 for Salol with the same parameters, but replace the persistence function in Eq.24 with the equilibrium survival function of the two-exponential waiting-time distribution Eq.20, P_R(t)=[τ1^2 e^{-t/τ1}+τ2^2 e^{-t/τ2}]/(τ1+τ2), taking τ1=1/P_hop(T) and τ2 from the MCT timescale, and keep φ_s(q,t) as in the paper. If Dτ_s/D0τ_s0 no longer grows with decreasing T (or grows much less than in Fig. 2), the predicted Stokes-Einstein breakdown is an artifact of the single-exponential persistence assumption. As a subsidiary check, run the equal-timescale case τ1=τ2 and verify that Eq.24 reduces to the original Unified result; if it does not, the renewal decomposition is internally inconsistent.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central prediction of the extended theory is produced by the mapping in Sec. II D, specifically Eq.23 and Eq.24, and that mapping is not a consequence of the CTRW/renewal formalism the paper invokes. The persistent-time argument (Eqs.20-22) is made for a stationary renewal process with a *mixture* waiting-time density P_waiting(t)=1/(2τ1)e^{-t/τ1}+1/(2τ2)e^{-t/τ2}; its equilibrium survival probability is P_R(t)=[τ1^2 e^{-t/τ1}+τ2^2 e^{-t/τ2}]/(τ1+τ2), a bi-exponential with a prefactor on the slow tail, not the single exponential exp(-P_hop t) inserted in Eq.23. Thus the slow 'persistent time' identified from the mixture is not the persistence function used in Eq.24. In addition, φ_diff in Eq.24 is taken to be the full Unified self-correlation φ_s=φ_s_hop φ_s_MCT (Appendix I), which already includes the activated hopping contribution φ_s_hop. Since P(t) is supposed to describe the first activated event, inserting φ_s_hop into the convolution double-counts the activated process: after the first jump the dynamics is again described by the activated channel, so the decomposition is not a renewal process with a single well-defined waiting-time distribution. The limit τ1=τ2 makes this explicit: Eq.22 gives τp=τx, but Eq.23-24 with identical rates yields φ=(1+P_hop t)e^{-P_hop t} rather than the single-exponential CTRW result e^{-P_hop t}, so the construction does not reduce to the process it claims to generalize. The SE breakdown in Fig. 2 therefore follows from the choice of P(t) and from leaving the MSD unchanged, not from a consistent two-waiting-time renewal model.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper examines the Unified MCT+RFOT theory for supercooled liquids and shows that this theory, in its original form (schemes 1 and 2), fails to reproduce the Stokes-Einstein (SE) breakdown for Salol. The authors first demonstrate that the Unified theory has the same structure as a continuous-time random walk with two exponential waiting-time distributions, where the total relaxation is dominated by the fast process. They then extend the theory using renewal-theory concepts: the first jump (persistence) is assigned to the slow activated dynamics, while subsequent jumps are described by the unchanged Unified two-channel correlation function. This gives Eq. (24), which is shown in Fig. 2 to reproduce the experimental SE breakdown with a decoupling ratio of about 10 at low temperature. The paper also computes a dynamic correlation length from the wavenumber dependence of the relaxation time and finds that it grows faster than the static RFOT length scale.","tokens_in":17103,"tokens_out":10314,"duration_ms":98168,"significance":"If the central mapping is accepted, the paper offers a microscopically motivated extension of MCT that captures the SE breakdown, a hallmark of supercooled liquids, and it makes a falsifiable prediction about a rapidly growing dynamic length scale. The formal observation that the Unified theory is equivalent to a two-exponential CTRW and is therefore dominated by the fast process is a useful and clearly presented result. The comparison with the Salol experiments is concrete, and the authors correctly identify that a full self-consistent extension of MCT should incorporate renewal statistics. However, the load-bearing identification of the persistence function with a single activated exponential (Eq. 23) is posited rather than derived, and the unchanged diffusion part (Eq. 24) is an additional assumption. The numerical predictions also rely on unspecified parameters tau0 and gamma. These issues limit the present form of the theory's predictive power, but they are addressable within the scope of a revision.","major_comments":[{"comment":"The identification P(t)=exp(-Phop(Delta F)t) is an assumption, not a consequence of the renewal formalism developed in Eqs. (20)-(22). For the two-exponential mixture of Eq. (20), the survival probability is 1/2 exp(-t/tau1)+1/2 exp(-t/tau2), which is not a single exponential. The persistent-time inequality of Eq. (22) only establishes that the mean first-jump time exceeds the mean exchange time; it does not fix the functional form of P(t). Since the strong decoupling between relaxation and diffusion shown in Fig. 2 follows directly from placing the slow activated rate in P(t) and the fast MCT product in phi_diff, this mapping is load-bearing and needs either a derivation from the CTRW statistics or an explicit statement that it is a phenomenological ansatz with supporting evidence.","section":"Sec. II D, Eq. (23)"},{"comment":"The identification phi_diff(t-t') = phi_s(t-t') = phi_s_hop phi_s_MCT means that the activated hopping process appears both in the persistence term P(t) and in the subsequent-dynamics term. In a renewal description, after the first jump the waiting time distribution for all subsequent jumps should be specified; here the same phi_s that already contains phi_s_hop is recycled. The statement that 'the diffusive part of the dynamics is not effected by the renewal theory and is same as given in the first part of this article' is asserted without derivation. This makes the decomposition in Eq. (24) ambiguous and prevents Eq. (24) from being a genuine renewal equation for the two-process system described in Sec. II D.","section":"Sec. II D, Eq. (24)"},{"comment":"The diffusion coefficient used in Fig. 2 is the long-time limit of the MSD of the original Unified theory (Eq. (10)), not a quantity derived from the renewal-modified dynamics of Eq. (24). The paper explicitly states that 'the MSD formalism remains unchanged' but does not justify why the first-jump slowdown leaves the long-time diffusion coefficient unaltered. This matters because the SE breakdown in Fig. 2 is obtained by combining the new slow relaxation time with the old, unchanged diffusion; a self-consistent extension of the Unified theory should derive D from the same renewal process. The authors should either provide a CTRW argument that D is indeed unchanged in the stationary state or acknowledge that the predicted breakdown is a hybrid construction.","section":"Sec. III A and Sec. II B"},{"comment":"The numerical values of the microscopic attempt time tau0 (Eq. (23)) and the binary friction gamma (Eq. (1)) are never specified. The Salol results in Figs. 2-5 depend on these parameters, so the calculations cannot be reproduced from the information given. The authors should list the values and, ideally, show the sensitivity of the SE-breakdown ratio to reasonable variations in tau0 and gamma.","section":"Appendix II and Sec. III A"}],"minor_comments":[{"comment":"The phrase 'continuous time random work' (used in the abstract and in Sec. I) should be 'continuous time random walk'.","section":"Abstract and Introduction"},{"comment":"The threshold Dq^2 tau_q^s = 5 used to define q*(T) is arbitrary; the dependence of l_dynamic on this threshold should be discussed.","section":"Sec. III B, Fig. 4"},{"comment":"Eq. (24) uses the self part phi_s for the diffusive part, while Eq. (18) uses phi_diff; the relationship between phi_diff and phi_s should be stated explicitly when Eq. (24) is introduced.","section":"Sec. II D, Eq. (24)"},{"comment":"Reference 40 is incomplete ('C. D. et al.'); the full author list and journal information should be provided.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and addresses a timely problem. The central issue is the ad hoc nature of the renewal mapping in Eqs. (23)-(24); the authors should be asked to either derive these equations from the CTRW mixture or clearly frame the extended theory as a phenomenological postulate with a discussion of its validity. The reproducibility gap concerning tau0 and gamma should also be fixed. I do not see grounds for rejection if these points are addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does something useful before it does something questionable. It shows clearly that the existing Unified theory (MCT+RFOT), in both schemes, fails to predict the SE breakdown, and it explains why: the CTRW analysis of two exponential waiting times gives a product form where the fast dynamics dominates both relaxation and diffusion. That part is solid and informative. The persistent-time inequality for two exponentials is also correct, and the derivation of the MSD in the appendices is careful. The extraction of a q-dependent dynamic lengthscale is a nice addition, though the threshold Dq^2tau=5 is ad hoc.\n\nThe trouble starts with Eq.23. The authors identify the persistence function P(t) with a single activated exponential, exp(-P_hop t). That is not a consequence of the renewal formalism they set up. For the two-exponential mixture in Eq.20, the survival probability is bi-exponential, not the single exponential they insert. More seriously, the diffusive part phi_diff is taken to be the full self-correlation phi_s = phi_s_hop * phi_s_MCT. Since phi_s_hop already contains activated hopping, the convolution in Eq.24 double-counts the activated channel: after the first jump, the dynamics is again described by the activated process, so this is not a renewal process with a well-defined waiting-time distribution. The tau1=tau2 limit makes the inconsistency visible: the construction reduces to (1+P_hop t) e^{-P_hop t}, not the single-exponential CTRW result. The stress-test note is right about these points. The SE breakdown in Fig.2 follows from the choice of P(t) and from leaving the MSD unchanged, not from the two-waiting-time renewal picture.\n\nThat said, this is not a dismissible paper. The physics motivation is reasonable, the numerical match to Salol is suggestive, and the proposed modification—slow first jump, faster subsequent jumps—is a clear and testable idea. The soft spots are the unproven mapping and the double-counting; they are serious but not fatal if the authors can provide a derivation or at least a careful justification of Eq.23, and if they address why phi_diff retains hopping. The overclaim about all extended MCT forms needs softening.\n\nFor a referee: this deserves peer review, because it is a serious attempt to fix a known deficiency in an influential theory, and it makes a concrete experimental prediction. But the referee should insist on a proper treatment of the persistence function and the renewal decomposition. I would not cite it in its current form, but I would watch for a revised version.","headline":"A serious attempt to extend MCT to the Stokes-Einstein breakdown, but the key renewal step is an assumption and the Fig. 2 prediction is a consequence of that assumption rather than a robust result; still worth refereeing.","tokens_in":17602,"tokens_out":3436,"would_cite":false,"duration_ms":40367,"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":"Adding renewal-theory time splitting to extended mode-coupling theory reproduces the Stokes-Einstein breakdown in supercooled Salol.","keywords":["Stokes-Einstein breakdown","mode coupling theory","random first order transition theory","continuous time random walk","renewal theory","persistent time","supercooled liquids","dynamic length scale"],"falsifier":"Track single-particle trajectories in a simulation of a supercooled liquid and extract the first-jump waiting time and the subsequent-jump waiting times separately. The extended theory predicts that the first-jump time should follow the RFOT activated rate $\\exp(-\\Delta F/k_BT)$ and should exceed the exchange time increasingly as $T$ drops, with $D\\tau_s/D_0\\tau_{s0}$ rising monotonically to roughly ten near the Salol $T_g$; a first-jump time no slower than later jumps, or a saturating or absent decoupling in $D\\tau_s$, would falsify the mechanism.","tokens_in":16417,"feed_emoji":"🧊","tokens_out":11235,"duration_ms":104563,"temperature":0.7,"pith_summary":"The paper aims to explain why supercooled liquids violate the Stokes-Einstein relation, in which diffusion and structural relaxation are locked together. It first shows that the existing Unified theory, which adds random-first-order-transition activated dynamics to mode-coupling theory, cannot produce the breakdown because its relaxation function is mathematically a continuous-time random walk with two waiting-time distributions, and such two-channel walks are controlled by the faster process. The paper then adds renewal theory, which distinguishes the time of the first jump (the persistent time) from the time of subsequent jumps (the exchange time), and shows that even two exponential waiting-time distributions make the persistent time the slower one. Identifying the persistent time with activated hopping and leaving the diffusive part unchanged gives Eq. (24), in which structural relaxation follows the activated barrier while diffusion follows the mode-coupling process. The result is a predicted decoupling factor of about ten for Salol that tracks the experimental trend, so the paper claims to have found the missing renewal ingredient for all extended MCT formulations.","feed_headline":"Supercooled liquids break Stokes-Einstein: first jumps are slow","feed_subtitle":"Adding a renewal-time split to extended mode-coupling theory reproduces the observed decoupling of diffusion and relaxation.","key_machinery":"The load-bearing object is the renewal-theory decomposition of a two-process random walk into a persistence function and a diffusive part. For a waiting-time distribution that is a mixture of two exponentials with timescales $\\tau_1,\\tau_2$, the persistent time is $\\langle \\tau_p\\rangle=(\\tau_1^2+\\tau_2^2)/(\\tau_1+\\tau_2)$ and the exchange time is $\\langle\\tau_x\\rangle=(\\tau_1+\\tau_2)/2$, so $\\langle\\tau_p\\rangle>\\langle\\tau_x\\rangle$ whenever the two channels differ; this is the identity that makes the first jump slower than later jumps. The paper maps $P(t)$ to the RFOT activated process and $\\phi_{\\rm diff}$ to the unchanged $\\phi_s=\\phi_s^{\\rm hop}\\phi_s^{\\rm MCT}$, converting the Unified theory into Eq. (24). This persistent-versus-exchange asymmetry is what transfers the slow timescale into structural relaxation while leaving diffusion fast.","core_discovery":"The central claim is that the earlier Unified theory fails at the Stokes-Einstein breakdown for a structural reason, not a parameter one. Because its relaxation function factorizes as $\\phi=\\phi_{\\rm MCT}\\phi_{\\rm hop}$, it is equivalent to a CTRW in which activated and MCT-like channels contribute two exponential waiting-time distributions; known CTRW results say such a mixture relaxes on the fast timescale, so both $\\tau_s$ and $D$ are MCT-controlled and never decouple. The paper's fix is the renewal-theory form (Eq. 24), $\\phi_{\\rm renewal}(q,t)=P(t)-\\int_0^t \\dot P(t')\\phi_s(q,t-t')\\,dt'$, with persistence function $P(t)=\\exp(-P_{\\rm hop}(\\Delta F)t)$ assigned to activated hopping and the diffusive part $\\phi_s=\\phi_s^{\\rm hop}\\phi_s^{\\rm MCT}$. In the extended theory, at low temperature structural relaxation is governed by the activated rate while diffusion remains MCT-like, so $D\\tau_s/D_0\\tau_{s0}$ rises to about ten as Salol is cooled toward $T_g$, in line with the experimental $D\\eta/T$ data. The paper further finds that the dynamic length scale extracted from the wavenumber dependence of $\\tau_s$ grows faster than the RFOT static length scale.","pith_inferences":["The persistent/exchange split could be measured directly from molecular-dynamics trajectories by classifying rearrangements into first and subsequent hops, giving a microscopic signature of the SE breakdown independent of the macroscopic $D\\tau_s$.","Because the barrier $\\Delta F$ is tied to configurational entropy through RFOT, the theory implies the magnitude of the SE breakdown should correlate with how steeply the entropy drops on cooling, a prediction that could be tested across glassformers with different fragilities.","The master-curve collapse of $D q^2\\tau_q^s$ against $q\\,l_{\\rm dynamic}$ suggests that the Fickian-to-non-Fickian crossover is a scaling feature that, if confirmed in other systems, would make $l_{\\rm dynamic}$ a well-defined operational measure of dynamical heterogeneity."],"forward_implications":["At low temperatures the structural relaxation time $\\tau_s$ is set by the activated barrier $\\Delta F(T)$, while the diffusion coefficient stays controlled by the MCT-like process, so the two decouple as $T$ decreases.","The failure of the original Unified theory is generic: any extended MCT that combines activated and MCT channels only through a product of correlators is equivalent to a two-channel CTRW and will be dominated by the fast process, so it cannot produce a strong SE breakdown.","In the extended theory, the ratio $D\\tau_s/D_0\\tau_{s0}$ for Salol grows to about ten near $T_g$ and does not saturate, matching the experimental $D\\eta/T$ trend.","The dynamic correlation length obtained from the Fickian-to-non-Fickian crossover in $D q^2 \\tau_q^s$ scales the relaxation data onto a master curve and grows faster than the static RFOT length.","At high temperatures, activated and MCT contributions to diffusion are comparable, so the theory recovers Stokes-Einstein behavior above the onset."],"supporting_citations":[{"why":"introduces the Unified theory whose relaxation function is the product the paper analyzes.","marker":"22"},{"why":"shows the Unified theory extends MCT relaxation to low temperatures, the framework being modified.","marker":"24"},{"why":"earlier extended MCT that reported only weak decoupling, the baseline the paper explains.","marker":"23"},{"why":"provides the CTRW result that two waiting-time distributions make dynamics fast-process dominated, diagnosing the original failure.","marker":"29,30"},{"why":"defines persistent and exchange times from waiting-time distribution moments, the renewal ingredient.","marker":"31"},{"why":"applies persistence/exchange decoupling in kinetically constrained models and links it to the SE breakdown.","marker":"32"},{"why":"supplies the RFOT free-energy barrier and static correlation length used to set the activated hopping rate.","marker":"13"},{"why":"provides the experimental Salol $D\\eta/T$ data used to compare the predicted decoupling.","marker":"40"}],"fun_headline_variants":["Renewal-time split fixes Stokes-Einstein breakdown in extended MCT","Why extended MCT missed SE breakdown: two waiting times, one fast","Slow first jumps reconcile mode-coupling theory with Stokes-Einstein","Persistent time unlocks SE breakdown in supercooled liquids","Unified theory's SE failure traced to fast MCT-dominated dynamics"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the mapping, introduced in Sec. II D, that at low temperatures the activated process is the slowest process, so the persistence function is $P(t)=\\exp(-P_{\\rm hop}(\\Delta F)t)$, and that the diffusive correlator $\\phi_s=\\phi_s^{\\rm hop}\\phi_s^{\\rm MCT}$ is not altered by renewal. If the first-jump time is not controlled by the activated barrier, or if renewal changes the diffusion channel, the predicted decoupling in Fig. 2 does not follow.","fun_headline_variants_meta":{"raw":{"variants":["Renewal-time split fixes Stokes-Einstein breakdown in extended MCT","Why extended MCT missed SE breakdown: two waiting times, one fast","Slow first jumps reconcile mode-coupling theory with Stokes-Einstein","Persistent time unlocks SE breakdown in supercooled liquids","Unified theory's SE failure traced to fast MCT-dominated dynamics"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000964,"raw_usage":{"total_tokens":4211,"prompt_tokens":1160,"completion_tokens":3051,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":776,"completion_tokens_details":{"reasoning_tokens":2959}},"tokens_in":776,"tokens_out":3051,"duration_ms":22213,"temperature":1.0,"reasoning_tokens":2959,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:58:43.154962+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Track single-particle trajectories in a simulation of a supercooled liquid and extract the first-jump waiting time and the subsequent-jump waiting times separately. The extended theory predicts that the first-jump time should follow the RFOT activated rate $\\exp(-\\Delta F/k_BT)$ and should exceed the exchange time increasingly as $T$ drops, with $D\\tau_s/D_0\\tau_{s0}$ rising monotonically to roughly ten near the Salol $T_g$; a first-jump time no slower than later jumps, or a saturating or absent decoupling in $D\\tau_s$, would falsify the mechanism.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"introduces the Unified theory whose relaxation function is the product the paper analyzes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"shows the Unified theory extends MCT relaxation to low temperatures, the framework being modified."},{"cited_title":"Grimmett and D","cited_arxiv_id":null,"evidence_quote":"defines persistent and exchange times from waiting-time distribution moments, the renewal ingredient."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"applies persistence/exchange decoupling in kinetically constrained models and links it to the SE breakdown."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the experimental Salol $D\\eta/T$ data used to compare the predicted decoupling."}],"review_version":1}