Nonlocal fractional Kardar-Parisi-Zhang dynamics of grain boundaries

arXiv:2606.21903 · cond-mat.stat-mech · Submitted 2026-06-20 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Nonlocal fractional Kardar-Parisi-Zhang dynamics of grain boundaries".

Mira: Here's the gist of the paper. The kinetic roughening of driven grain boundaries (GBs) is fundamentally mediated by the collective behavior of disconnections, whose long-range 1/r elastic interactions distinguish them from conventional growing surfaces Driven GBs undergo a sharp transition of morphology at the yield point, from the quenched Edwards-Wilkinson universality class (H ≈ 0.33) to an anomalous fKPZ regime (H ≈ 0.8), while the 1/r nonlocal elasticity manifests as the fractional relaxation of GB morphology The migration of grain boundaries (GBs) fundamentally governs the microstructural evolution, plastic deformation, and macroscopic thermomechanical properties of polycrystalline materials Far from the classical picture of continuous, curvature-driven flow, GB migration is now firmly established to be mediated by the nucleation and lateral glide of discrete topological line defects known as disconnections Each disconnection is characterized by a dual topological invariant, a Burgers vector b quantifying the interfacial shear displacement, and a step height h quantifying the normal migration, that together provide the mechanistic basis for shear-coupled GB migration Modeling Nonlocal Elasticity The internal force of disconnections is mediated by surrounding bulk elastically, as described by the long-range interactions Specifically, the internal force is given by The long-range interactions among disconnections are encoded into this N-body Langevin framework In Fourier space, this transformation yields the spectral representation Coarse-Graining to the fKPZ Equation The coarse-graining procedure culminates in the following fKPZ equation The total noise is thus decoupled as etatot = etath(x,t) + etaq(x, y(x,t)) This representation establishes that the macroscopic fKPZ mechanism here is not a local expansion, but rather the avalanches triggered by the collective behavior of disconnections Universality Classes and Scaling The fKPZ equation recovers three regimes observed in MD simulations In the pre-yield regime, the GB morphology is governed by the qEW universality class with a universal Hurst exponent of H ≈ 0.33 Upon yielding, the interfacial morphology converges to the fKPZ universality class, anchoring at a plateau of H ≈ 0.8 This anomalous scaling is dictated by the spatial correlation exponent of the spatially correlated noise, yielding an exact roughness exponent of H = 7/9 Temperature Dependence and Stress Anomaly The extracted driving stresses exhibit a non-Arrhenius dependence of temperature This anomalous behavior provides strong evidence for the interplay between disconnection avalanches and thermal fluctuations The transition from unhindered avalanche glide at low temperatures to the fluctuation-pinned super-roughening phase at high temperatures manifests as the observed stress anomaly The onset of severe yielding pushes the interface from a subdiffusive state (H ≈ 0.33) into a super-roughening paradigm (H ≈ 0.8) The emergence of spatially-correlated noise, originating from defect coalescence, serves as a dimension-agnostic signature of the catastrophic failure This morphological evolution also unveils a system-agnostic universality, mirroring the kinetic roughening signatures observed in dynamic fracture Temperature Effects on Glide At low temperatures (300 K), the structurally ordered interface facilitates facile, low-barrier glide of disconnections, necessitating minimal stress However, as temperature increases (500 K), the thermally roughened interface manifests as an array of dynamic pinning centers, retarding the disconnection motion and thereby demanding larger driving stress At even higher temperatures (750 K - 1000 K), abundant thermal activation (e.g., atomic shuffling) overrides the pinning landscape, inducing interfacial softening and recovering the classical Arrhenius-like temperature dependence Yield Point Dynamics The yield point emerges as a bifurcation where the stress-driven sliding amplifies the kinematic steepening to overwhelmingly outpace fractional relaxation This breakdown of fractional regularization inevitably drives compressive modes toward a gradient catastrophe, culminating in the formation of shocks The post-yield fKPZ dynamics is governed by the dual-noise field, i.e., etatot(x,t) = etath(x,t) + etadis(x,t) Morphological Fossilization The postyield fKPZ dynamics is governed by the dual-noise field, i.e., etatot(x,t) = etath(x,t) + etadis(x,t) The morphological correspondence circumvents this limitation, i.e., the 1D out-of-plane trace of the crack surface, measured parallel to the crack front, serves as a morphological 'fossil' of the instantaneous dynamic front During dynamic failure driven by the coalescence of damages, analogous to the disconnection avalanches, the 1D trace roughness of the fracture surface universally jumps to H ≈ 0.8 Universality Across Physics These findings unveil a fundamental universality that transcends disparate branches of physics While their elastodynamic kernels diverge due to spatial dimensions, both the GB and crack front are fundamentally mediated by the nonlocal elasticity and propelled by intermittent avalanches The onset of severe yielding pushes the interface from a subdiffusive state (H ≈ 0.33) into a super-roughening paradigm (H ≈ 0.8) The emergence of spatially-correlated noise, originating from defect coalescence, serves as a dimension-agnostic signature of the catastrophic failure Governing Equation Analysis In Fourier space, this spatial differentiation introduces a q2 multiplier to the noise field Since the avalanche exponent epsilon = 4/3 is less than 2, the transformed noise spectrum vanishes in the hydrodynamic limit The yield point emerges as a bifurcation where stress-driven sliding amplifies kinematic steepening to overwhelmingly outpace fractional relaxation This breakdown of fractional regularization inevitably drives compressive modes toward a gradient catastrophe, culminating in the formation of shocks Final Morphological State By dynamically arresting the steepening wave fronts, this avalanche noise field averts the shock singularity and anchors the system at the plateau of H ≈ 0.8 Ultimately, this morphological evolution also unveils a system-agnostic universality, mirroring the kinetic roughening signatures observed in dynamic fracture The onset of severe yielding pushes the interface from a subdiffusive state (H ≈ 0.33) into a super-roughening paradigm (H ≈ 0.8) The kinetic roughening of driven grain boundaries is governed by nonlocal fractional Kardar-Parisi-Zhang dynamics The morphology transitions from the quenched Edwards-Wilkinson universality class (H ≈ 0.33) to an anomalous fKPZ regime (H ≈ 0.8) at the yield point This transition is driven by disconnection avalanches acting as spatially correlated noise, which arrests the gradient catastrophe predicted by the fractional Burgers equation Temperature Influence on Dynamics The interplay between disconnection avalanches and thermal fluctuations is evident in the temperature dependence of driving stresses Low temperatures favor facile glide, while higher temperatures manifest as dynamic pinning centers that retard disconnection motion Theoretical Framework Derivations The theoretical framework involves mapping the Langevin dynamics onto a continuous interface height profile The essential step in capturing the nonlocal interactions is transforming the sum of discrete 1/r terms into the Cauchy principal value integral over the density of disconnections This procedure endows the elastic interactions with the spectral representation Numerical Validation By solving the fKPZ equation with a spatially-correlated noise featuring a correlation exponent of epsilon = 1.4, numerical results show that a plateau of H ≈ 0.8 is reproduced upon yielding This perfectly matches atomistic observations The ensemble averages of 50, 300, and 100 trajectories were obtained by solving Eq.(2) with the correlation exponent of the spatially-correlated noise epsilon = 1.4 Comparison to Fracture Universality The observed plateau of H ≈ 0.8 is not merely a coincidence, but the intrinsic signature of avalanche dynamics governed by the fKPZ universality class This places the driven GB within the universality class of dynamic fracture The 1D out-of-plane trace roughness of the fracture surface universally jumps to H ≈ 0.8 during dynamic failure The GB morphology undergoes a dynamic transition from a sub-diffusive elastic string to an anomalously roughened fKPZ interface The pre-yield regime is characterized by the qEW universality class with H ≈ 0.33 The post-yield regime is dominated by disconnection avalanches, leading to an anomalous scaling H ≈ 0.8 This signature is a dimension-agnostic signature of the catastrophic failure Final Conclusions The results reveal that the morphology of driven GBs undergoes a dynamic transition from a sub-diffusive elastic string to an anomalously roughened fKPZ interface The onset of severe yielding pushes the interface from a subdiffusive state (H ≈ 0.33) into a super-roughening paradigm (H ≈ 0.8) The emergence of spatially-correlated noise, originating from defect coalescence, serves as a dimension-agnostic signature of the catastrophic failure Acknowledgments This work was funded by the National Natural Science Foundation of China (Grant No.

Kai: 12102145), and Natural Science Foundation of Jiangsu Province (Grant No.

Paper summary: Kai: So we're looking at this paper now titled "Nonlocal fractional Kardar-Parisi-Zhang dynamics of grain boundaries." Basically, they’re showing that how these grain boundaries move when driven is controlled by a collective behavior called disconnections.

Mira: That’s right. The central idea here is that the long-range one over r elastic interactions between these disconnections are what really distinguish them from regular surfaces, which is what drives the kinetic roughening we see.

Lev: From my side, I’m thinking about how this maps onto real hardware. If we were building a system where these grain boundaries existed, the nonlocal elasticity means the internal force isn't just local; it depends on everything around it in a way that requires N body Langevin framework to describe it <ref:2606.21903#pg1>.

Kai: Right, so they use molecular dynamics simulations to see this happening and they find a sharp change in the shape of the boundary at the yield point.

Mira: They found a transition from the quenched Edwards-Wilkinson universality class with a Hurst exponent H equal to zero point three three to an anomalous fKPZ regime with H equal to zero point eight <ref:2606.21903#pg1>. That change is driven by that nonlocal elasticity manifesting as this fractional relaxation of the boundary shape itself.

Lev: And what’s interesting for me is that they show this avalanche noise actually arrests the gradient catastrophe, which is usually predicted by the standard KPZ nonlinearity <ref:2606.21903#pg2>. That suggests a mechanism for stabilizing the system when it gets really rough.

Kai: So what does this mean for us? It means that when things start yielding and getting rough, the way they get rough is dictated by these collective disconnections rather than just local forces <ref:2606.21903#pg1>.

Mira: And it connects this behavior to something else too; the resulting morphology parallels the universality class of dynamic fracture <ref:2606.21903#pg1>. It’s a dimension-agnostic signature of catastrophic failure, which is quite powerful.

Lev: From a hardware perspective, if we try to run this on real systems, the complexity introduced by those long-range interactions means the overhead for error correction would be significant <ref:2606.21903#pg2>.

Kai: So they also looked at how temperature affects things, and they found that low temperatures allow for easier movement of these disconnections, but higher temperatures introduce dynamic pinning centers that slow everything down <ref:2606.21903#pg3>.

Paper summary: Mira: And the paper shows this is linked to a stress anomaly; the way you pull on it changes with temperature in a non-Arrhenius way, which points to these interplay between thermal fluctuations and those disconnection avalanches <ref:2606.21903#pg1>.

Lev: That temperature dependence is something I'd need to verify on actual hardware, because if we can't control the pinning landscape at high temperatures, we won't be able to predict the stress behavior accurately <ref:2606.21903#pg3>.

Kai: What about the mathematical side of this? They establish a continuous field theory first to make sure it’s not just some phenomenological guess about how disconnections work <ref:2606.21903#pg3>.

Mira: They do that by reducing the evolution to the center of mass trajectories of N discrete disconnections, and then applying virtual work principles to find the driving force for a virtual glide <ref:2606.21903#pg3>.

Lev: That sounds like a lot of bookkeeping for simulation. I wonder if solving that full N body system is feasible when you have these nonlocal kernels involved <ref:2606.21903#pg1>.

Kai: But the whole point is that they ended up coarse-graining this down to an fKPZ equation where the total noise is split into a local part and a part dependent on the shape of the interface itself <ref:2606.21903#pg2>.

Mira: Because that representation shows that the macroscopic fKPZ mechanism isn't some simple local expansion, but rather these avalanches triggered by the collective behavior of all those disconnections <ref:2606.21903#pg2>.

Lev: So if we were trying to model this on a real processor, we’d have to deal with that dual noise field, the part from the shape and the part from the disconnections <ref:2606.21903#pg4>.

Kai: And they found that in Fourier space, this spatial differentiation introduces a squared multiplier to the noise field because of how fast things are evolving <ref:2606.21903#pg4>.

Mira: And because the avalanche exponent is four/three which is less than two, that transformed noise spectrum vanishes in the hydrodynamic limit <ref:2606.21903#pg4>. That’s a key scaling feature for them.

Lev: So if we think about real-world constraints, this vanishing spectrum in the limit means that for very large systems, this specific nonlocal effect gets washed out unless you're looking at specific microscopic details <ref:2606.21903#pg4>.

Paper summary: Kai: But they still found a plateau of H equal to zero point eight when they solve the fKPZ equation using a spatially correlated noise with a correlation exponent of one point four <ref:2606.21903#pg5>. That's what they got from their MD runs.

Mira: And that result, H equal to zero point eight, is what they claim is the universal value for this regime, and it matches what they see in other systems <ref:2606.21903#pg3>.

Lev: Matching results are good for theory, but I always ask about the limitations they state at the end of their work. Where does their model stop working? <ref:2606.21903#pg5>.

Kai: They mention that the morphological correspondence they use, like looking at the 1D trace roughness of a crack surface parallel to its front, is just a 'fossil' of the instantaneous dynamic front, not necessarily a perfect description of every single microscopic detail <ref:2606.21903#pg5>.

Mira: So they are linking these grain boundary dynamics to dynamic fracture in terms of universal scaling, which is a big connection for condensed matter theorists like us <ref:2606.21903#pg1>.

Lev: It means if we can understand the avalanche mechanism here, it might inform how we model interfaces in other complex systems where local rules break down <ref:2606.21903#pg4>.

Kai: And for anyone listening who isn't a materials scientist, the main thing is that severe yielding pushes the interface from a subdiffusive state with H equal to zero point three three into this super-roughening regime with H equal to zero point eight <ref:2606.21903#pg1>.

Mira: It highlights how the interplay between those long-range elastic interactions and thermal fluctuations creates a dynamic transition at the yield point <ref:2606.21903#pg1>.

Lev: And it also shows that temperature matters for glide, but when you get very hot, thermal activation overrides the pinning landscape and we start seeing that Arrhenius-like temperature dependence again <ref:2606.21903#pg3>.

Kai: So the title itself, "Nonlocal fractional Kardar-Parisi-Zhang dynamics of grain boundaries," really captures the essence of what they did here <ref:2606.21903#pg1>.

Mira: It’s a study that bridges the gap between how we see microscopic disconnections and the macroscopic non-equilibrium statistical physics of grain boundaries <ref:2606.21903#pg2>.

Lev: And ultimately, it tells us that these systems follow a universal scaling behavior when they hit severe yielding, regardless of the specific material details <ref:2606.21903#pg1>.

Conclusion: Kai: So, to wrap up, we're talking about grain boundaries moving under stress and how they transition from behaving like simple elastic strings to something much more complex when they start yielding <ref:2606.21903#pg1>.

Mira: Exactly. The title itself points to the core idea that what we see isn't just local physics anymore; it’s about nonlocal fractional Kardar-Parisi-Zhang dynamics <ref:2606.21903#pg1>.

Lev: From an error correction standpoint, this means any real hardware trying to model this will face a huge overhead because of that long range interaction you mentioned earlier <ref:2606.21903#pg4>.

Kai: Right, so when the authors say it's about collective behavior of disconnections, they are basically saying the driving force isn't one single line sliding but a whole crowd of defects acting together <ref:2606.21903#pg1>.

Mira: And that collective action is what creates this spatial correlation in the noise field, which drives that jump from the subdiffusive state to the super-roughening regime <ref:2606.21903#pg1>.

Lev: So if we look at the math, they found this scaling exponent of four thirds, which is less than two, so in a truly big system, that noise part essentially vanishes in the hydrodynamic limit <ref:2606.21903#pg4>.

Kai: That's a bit like saying if you have a huge crowd of things moving, the individual random jitters don't matter as much as the way they are all linked up <ref:2606.21903#pg4>.

Mira: Precisely. The key result here is that this noise pattern acts like a stabilizer; it stops the gradient catastrophe that we usually expect from these kinds of nonlinear dynamics <ref:2606.21903#pg4>.

Lev: That stabilization is what allows the system to settle at a specific rough state, H equal to point eight, which is what they found in their simulations with an alpha equal to one point four <ref:2606.21903#pg5>.

Kai: So for someone just listening who doesn't know about grain boundaries, the simple thing is that severe stress causes the interface to get rougher than you'd expect, and this roughness is controlled by how defects interact across long distances <ref:2606.21903#pg1>.

Mira: And what this implies for us in condensed matter theory is that we have a new way to look at these interfaces where the driving physics comes from the non-local structure of the defects themselves <ref:2606.21903#pg2>.

Lev: So, moving forward, if we want to actually build a model for this on real hardware, we need to figure out how much noise you can effectively filter out when you have that kind of long range coupling in the simulation <ref:2606.21903#pg4>.

School of Mechanical Engineering, Jiangnan University

cond-mat.stat-mech

Submitted: 2026-06-20

Updated: 2026-11-11

Comments: 39 pages, 8 figures

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 79/100

The gist: The kinetic roughening of driven grain boundaries (GBs) is fundamentally mediated by the collective behavior of disconnections…

Key concepts

Disconnections
These are discrete topological line defects on a grain boundary that govern migration. They have two key properties: a Burgers vector (quantifying shear displacement) and a step height (quantifying normal migration), which together explain how the boundary moves under shear.
fKPZ Equation
This is the governing equation describing the anomalous kinetic roughening of driven grain boundaries. It shows that macroscopic roughness isn't just local growth but results from avalanches triggered by the collective behavior of these disconnections, leading to a specific scaling exponent.
Universality Class Transition
The system shifts its behavior at the yield point from a quenched Edwards-Wilkinson class (H≈0.33) to an fKPZ class (H≈0.8). This transition is driven by the emergence of spatially correlated noise from defect coalescence, demonstrating a universal signature of catastrophic failure across different physical systems.

Terminology

Summary

The kinetic roughening of driven grain boundaries (GBs) is fundamentally mediated by the collective behavior of disconnections, whose long-range 1/r elastic interactions distinguish them from conventional growing surfaces

Driven GBs undergo a sharp transition of morphology at the yield point, from the quenched Edwards-Wilkinson universality class (H ≈ 0.33) to an anomalous fKPZ regime (H ≈ 0. p. 0

Where the paper says it

Through large-scale molecular dynamics simulations and a nonlocal fractional Kardar-Parisi-Zhang (fKPZ) theory, we demonstrate that driven GBs undergo a sharp transition of morphology at the yield point, from the quenched Edwards-Wilkinson universality class (H ≈ 0. 33) to an anomalous fKPZ regime (H ≈ 0.

p. 0
Where the paper says it

Through large-scale molecular dynamics simulations and a nonlocal fractional Kardar-Parisi-Zhang (fKPZ) theory, we demonstrate that driven GBs undergo a sharp transition of morphology at the yield point, from the quenched Edwards-Wilkinson universality class (H ≈ 0. 33) to an anomalous fKPZ regime (H ≈ 0.

8), while the 1/r nonlocal elasticity manifests as the fractional relaxation of GB morphology

Introduction to Grain Boundary Dynamics

The migration of grain boundaries (GBs) fundamentally governs the microstructural evolution, plastic deformation, and macroscopic thermomechanical properties of polycrystalline materials Far from the classical picture of continuous, curvature-driven flow, GB migration is now firmly established to be mediated by the nucleation and lateral glide of discrete topological line defects known as disconnections Each disconnection is characterized by a dual topological invariant, a Burgers vector b quantifying the interfacial shear displacement, and a step height h quantifying the normal migration, that together provide the mechanistic basis for shear-coupled GB migration

Modeling Nonlocal Elasticity

The internal force of disconnections is mediated by surrounding bulk elastically, as described by the long-range interactions Specifically, the internal force is given by The long-range interactions among disconnections are encoded into this N-body Langevin framework In Fourier space, this transformation yields the spectral representation

Coarse-Graining to the fKPZ Equation

The coarse-graining procedure culminates in the following fKPZ equation The total noise is thus decoupled as etatot = etath(x,t) + etaq(x, y(x,t)) This representation establishes that the macroscopic fKPZ mechanism here is not a local expansion, but rather the avalanches triggered by the collective behavior of disconnections

Universality Classes and Scaling

The fKPZ equation recovers three regimes observed in MD simulations In the pre-yield regime, the GB morphology is governed by the qEW universality class with a universal Hurst exponent of H ≈ 0.33 Upon yielding, the interfacial morphology converges to the fKPZ universality class, anchoring at a plateau of H ≈ 0. p. 5

Where the paper says it

33, which is in quantitative agreement with MD results [1,3,37]. However, upon yielding, the interfacial morphology converges to the fKPZ universality class.

8 This anomalous scaling is dictated by the spatial correlation exponent of the spatially correlated noise, yielding an exact roughness exponent of H = 7/9

Temperature Dependence and Stress Anomaly

The extracted driving stresses exhibit a non-Arrhenius dependence of temperature This anomalous behavior provides strong evidence for the interplay between disconnection avalanches and thermal fluctuations The transition from unhindered avalanche glide at low temperatures to the fluctuation-pinned super-roughening phase at high temperatures manifests as the observed stress anomaly

Conclusion on Universality

The onset of severe yielding pushes the interface from a subdiffusive state (H ≈ 0.33) into a super-roughening paradigm (H ≈ 0. p. 5

Where the paper says it

The onset of severe yielding pushes the interface from a subdiffusive state (𝐻 ≅ 0. 33) into a super-roughening paradigm (𝐻 ≅ 0.

p. 5
Where the paper says it

The onset of severe yielding pushes the interface from a subdiffusive state (𝐻 ≅ 0. 33) into a super-roughening paradigm (𝐻 ≅ 0.

p. 5
Where the paper says it

The onset of severe yielding pushes the interface from a subdiffusive state (𝐻 ≅ 0. 33) into a super-roughening paradigm (𝐻 ≅ 0.

p. 5
Where the paper says it

The onset of severe yielding pushes the interface from a subdiffusive state (𝐻 ≅ 0. 33) into a super-roughening paradigm (𝐻 ≅ 0.

8) The emergence of spatially-correlated noise, originating from defect coalescence, serves as a dimension-agnostic signature of the catastrophic failure This morphological evolution also unveils a system-agnostic universality, mirroring the kinetic roughening signatures observed in dynamic fracture

Temperature Effects on Glide

At low temperatures (300 K), the structurally ordered interface facilitates facile, low-barrier glide of disconnections, necessitating minimal stress However, as temperature increases (500 K), the thermally roughened interface manifests as an array of dynamic pinning centers, retarding the disconnection motion and thereby demanding larger driving stress At even higher temperatures (750 K - 1000 K), abundant thermal activation (e.g., atomic shuffling) overrides the pinning landscape, inducing interfacial softening and recovering the classical Arrhenius-like temperature dependence

Yield Point Dynamics

The yield point emerges as a bifurcation where the stress-driven sliding amplifies the kinematic steepening to overwhelmingly outpace fractional relaxation This breakdown of fractional regularization inevitably drives compressive modes toward a gradient catastrophe, culminating in the formation of shocks The post-yield fKPZ dynamics is governed by the dual-noise field, i.e., etatot(x,t) = etath(x,t) + etadis(x,t)

Morphological Fossilization

The postyield fKPZ dynamics is governed by the dual-noise field, i.e., etatot(x,t) = etath(x,t) + etadis(x,t) The morphological correspondence circumvents this limitation, i. p. 4

Where the paper says it

8 signature places the driven GB within the universality class of dynamic fracture. While the in-situ tracking of 1D crack fronts remains experimentally elusive, the morphological correspondence circumvents this limitation, i.

e., the 1D out-of-plane trace of the crack surface, measured parallel to the crack front, serves as a morphological 'fossil' of the instantaneous dynamic front During dynamic failure driven by the coalescence of damages, analogous to the disconnection avalanches, the 1D trace roughness of the fracture surface universally jumps to H ≈ 0.8

Universality Across Physics

These findings unveil a fundamental universality that transcends disparate branches of physics While their elastodynamic kernels diverge due to spatial dimensions, both the GB and crack front are fundamentally mediated by the nonlocal elasticity and propelled by intermittent avalanches The onset of severe yielding pushes the interface from a subdiffusive state (H ≈ 0.33) into a super-roughening paradigm (H ≈ 0.8) The emergence of spatially-correlated noise, originating from defect coalescence, serves as a dimension-agnostic signature of the catastrophic failure

Governing Equation Analysis

In Fourier space, this spatial differentiation introduces a q2 multiplier to the noise field Since the avalanche exponent epsilon = 4/3 is less than 2, the transformed noise spectrum vanishes in the hydrodynamic limit The yield point emerges as a bifurcation where stress-driven sliding amplifies kinematic steepening to overwhelmingly outpace fractional relaxation This breakdown of fractional regularization inevitably drives compressive modes toward a gradient catastrophe, culminating in the formation of shocks

Final Morphological State

By dynamically arresting the steepening wave fronts, this avalanche noise field averts the shock singularity and anchors the system at the plateau of H ≈ 0.8 Ultimately, this morphological evolution also unveils a system-agnostic universality, mirroring the kinetic roughening signatures observed in dynamic fracture The onset of severe yielding pushes the interface from a subdiffusive state (H ≈ 0. p. 5

Where the paper says it

By dynamically arresting the steepening wave fronts, this avalanche noise field averts the shock singularity and anchors the system at the plateau of 𝐻 ≅ 0. Ultimately, this morphological evolution also unveils a system-agnostic universality, mirroring the kinetic roughening signatures observed in dynamic fracture.

33) into a super-roughening paradigm (H ≈ 0.8)

Summary of Findings

The kinetic roughening of driven grain boundaries is governed by nonlocal fractional Kardar-Parisi-Zhang dynamics The morphology transitions from the quenched Edwards-Wilkinson universality class (H ≈ 0.33) to an anomalous fKPZ regime (H ≈ 0.8) at the yield point This transition is driven by disconnection avalanches acting as spatially correlated noise, which arrests the gradient catastrophe predicted by the fractional Burgers equation

Temperature Influence on Dynamics

The interplay between disconnection avalanches and thermal fluctuations is evident in the temperature dependence of driving stresses Low temperatures favor facile glide, while higher temperatures manifest as dynamic pinning centers that retard disconnection motion

Theoretical Framework Derivations

The theoretical framework involves mapping the Langevin dynamics onto a continuous interface height profile The essential step in capturing the nonlocal interactions is transforming the sum of discrete 1/r terms into the Cauchy principal value integral over the density of disconnections This procedure endows the elastic interactions with the spectral representation

Numerical Validation

By solving the fKPZ equation with a spatially-correlated noise featuring a correlation exponent of epsilon = 1.4, numerical results show that a plateau of H ≈ 0. p. 3

Where the paper says it

2, by solving the fKPZ equation with a spatially-correlated noise featuring a correlation exponent of 𝜖 = 1. 4 [33], the numerical results show that a plateau of 𝐻 ≅ 0.

8 is reproduced upon yielding This perfectly matches atomistic observations The ensemble averages of 50, 300, and 100 trajectories were obtained by solving Eq. p. 3
Where the paper says it

Morphological transition and the anomalous Hurst exponent 𝐻 across the yield point 𝑡𝑐 . Ensemble averages of 50, 300, and 1000 trajectories are obtained by solving Eq.

(2) with the correlation exponent of the spatially-correlated noise epsilon = 1. p. 3
Where the paper says it

As shown in Fig. 2, by solving the fKPZ equation with a spatially-correlated noise featuring a correlation exponent of 𝜖 = 1.

4

Comparison to Fracture Universality

The observed plateau of H ≈ 0.8 is not merely a coincidence, but the intrinsic signature of avalanche dynamics governed by the fKPZ universality class This places the driven GB within the universality class of dynamic fracture The 1D out-of-plane trace roughness of the fracture surface universally jumps to H ≈ 0. p. 4

Where the paper says it

Therefore, the observed plateau of 𝐻 ≅ 0. 8 is not merely a coincidence, but the intrinsic signature of avalanche dynamics governed by the fKPZ universality class.

8 during dynamic failure

Summary of Key Results

The GB morphology undergoes a dynamic transition from a sub-diffusive elastic string to an anomalously roughened fKPZ interface The pre-yield regime is characterized by the qEW universality class with H ≈ 0.33 The post-yield regime is dominated by disconnection avalanches, leading to an anomalous scaling H ≈ 0. p. 4

Where the paper says it

Physically, this emergent singularity signifies the onset of disconnection avalanches, driving the system far from the thermal equilibrium state and propelling it into an anomalous phase. anomalous scaling in post-yield regime Upon yielding, the GB enters into an anomalous phase dominated by disconnection avalanches.

8 This signature is a dimension-agnostic signature of the catastrophic failure

Final Conclusions

The results reveal that the morphology of driven GBs undergoes a dynamic transition from a sub-diffusive elastic string to an anomalously roughened fKPZ interface The onset of severe yielding pushes the interface from a subdiffusive state (H ≈ 0.33) into a super-roughening paradigm (H ≈ 0.8) The emergence of spatially-correlated noise, originating from defect coalescence, serves as a dimension-agnostic signature of the catastrophic failure

Acknowledgments

This work was funded by the National Natural Science Foundation of China (Grant No. 12102145), and Natural Science Foundation of Jiangsu Province (Grant No. p. 5

Where the paper says it

ACKNOWLEDGMENTS This work was funded by the National Natural Science Foundation of China (Grant No. 12102145), and Natural Science Foundation of Jiangsu Province (Grant No.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI systems can predict morphological transitions in driven grain boundaries by employing a nonlocal fractional Kardar-Parisi-Zhang (fKPZ) theory derived from large-scale molecular dynamics simulations, allowing them to predict shifts between quenched Edwards-Wilkinson and fKPZ universality classes based on applied strain.

  2. The improved system can model the yield point transition in grain boundaries by incorporating the disconnection avalanches as a source of spatially correlated noise, which dynamically arrests gradient catastrophe and anchors the interface at an anomalous scaling exponent H equal to 0.8.

  3. These systems can predict temperature-dependent flow stress anomalies by coupling disconnection avalanche dynamics with thermal fluctuations, enabling accurate modeling of how thermal activation overrides pinning landscapes at high temperatures compared to low temperatures.

Abstract

The kinetic roughening of driven grain boundaries (GBs) is mediated by the collective behavior of disconnections, whose long-range 1/r elastic interactions distinguish them from conventional growing surfaces. Large-scale molecular dynamics simulations show that the roughness of driven GBs undergoes a sharp crossover as the applied strain increases, with the Hurst exponent (H) increasing from approximately 0.33 to 0.8. We develop a nonlocal fractional Kardar-Parisi-Zhang (fKPZ) theory to describe this low-temperature behavior, where the 1/r nonlocal elasticity behaves as the fractional relaxation of the GB morphology. The results show that the spatially-correlated noise associated with disconnection avalanches interacts with the KPZ nonlinearity to produce the post-crossover roughening observed in atomistic simulations. The post-crossover roughness exponent (H 0.8) is comparable to values reported for dynamic fracture, suggesting the underlying physics shared by the collective behavior of defects (e.g., disconnections, dislocations) and interfacial roughening.

Related papers