Computing bulk and shear viscosities from simulations of fluids with dissipative and stochastic interactions
summary
The gist
Exact values for bulk and shear viscosity are important to characterize a fluid and they are a necessary input for a continuum description Here we present two novel methods to compute bulk…
In short
The paper introduces two new non-equilibrium molecular dynamics (NEMD) techniques to calculate bulk viscosity from steady-state fluid simulations using periodic boundary conditions: a 'particle transfer method' and a 'force driven method.' These methods are then tested against Green-Kubo relations using Dissipative Particle Dynamics (DPD) fluid simulations. Results show good agreement, with discrepancies resolved by neglecting kinetic contributions in theoretical predictions.
Key concepts
- Bulk Viscosity ($Φ$)
- A measure of a fluid's resistance to uniform compression or expansion. It is a crucial property for describing fluids in continuum mechanics and requires exact values from simulations for accurate modeling.
- Non-Equilibrium Molecular Dynamics (NEMD)
- A simulation technique used to study fluids by applying external forces or flows to the particles, allowing researchers to measure transport properties like viscosity. It helps characterize how a fluid responds when it is not in thermal equilibrium.
- Green-Kubo Relations
- Mathematical relations used to calculate transport coefficients, such as viscosity, by integrating the time correlation of microscopic fluctuations. The paper tests generalized forms of these relations to better separate random (stochastic) and dissipative effects.
- Dissipative Particle Dynamics (DPD)
- A type of molecular dynamics simulation that models fluid behavior by including both conservative forces and dissipative forces, which introduce friction. This method is used here to test the calculation methods for bulk viscosity in a realistic, dissipative system.
Terminology used across episodes
This episode discusses
- Computing bulk and shear viscosities from simulations of fluids with dissipative and stochastic interactions · Paper Radio
The paper
Computing bulk and shear viscosities from simulations of fluids with dissipative and stochastic interactions · Read on arXiv
Gerhard Jung, Friederike Schmid
Institut für Physik, Johannes Gutenberg-Universität Mainz
DOI: 10.1063/1.4950760
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: "Computing bulk and shear viscosities from simulations of fluids with dissipative and stochastic interactions".
Mira: Here's the gist of the paper. Exact values for bulk and shear viscosity are important to characterize a fluid and they are a necessary input for a continuum description Here we present two novel methods to compute bulk viscosities by non-equilibrium molecular dynamics (NEMD) simulations of steady-state systems with periodic boundary conditions The paper proposes two novel NEMD techniques to measure the bulk viscosity ζ from fluid simulations of steady-state systems with periodic boundary conditions Second, we test the Green-Kubo relations (1) and (2) using the example of a dissipative particle dynamics (DPD) fluid and compare the resulting values with the results from NEMD simulations.
Kai: Novel NEMD Techniques The paper introduces two novel NEMD techniques to measure the bulk viscosity ζ from fluid simulations of steady-state systems with periodic boundary conditions One, denoted ”particle transfer method”, creates a divergent flow field by manually displacing particles This method is most efficient in systems of particles interacting by soft potentials.
Mira: The other, denoted ”force driven method”, makes use of a spatially varying body force and a non-zero center of mass velocity and can be used in a wider range of molecular dynamics simulations.
Kai: Green-Kubo Relations The paper discusses different expressions proposed in the literature for calculating transport coefficients using Green-Kubo relations It focuses on generalized Green-Kubo expressions that have the useful advantage of separating the stochastic contribution and should therefore give better statistics.
Title and authors: Kai: So we're talking about this paper, "Computing bulk and shear viscosities from simulations of fluids with dissipative and stochastic interactions." It’s by Gerhard Jung and Friederike Schmid. The title tells you exactly what it’s about: calculating the bulk viscosity zeta and the shear viscosity eta using molecular dynamics simulations that have both dissipative forces and random forces.
Mira: I think that title is pretty dense, but basically, they are trying to figure out how to get those exact values for bulk and shear viscosity because those numbers are absolutely essential for describing any fluid in a continuum level description. It sets up the problem right there: getting these inputs from complex molecular dynamics simulations.
Lev: From an engineering standpoint, getting the exact viscosity values is crucial because if you're designing something that needs a continuum model, those inputs have to be right, especially when dealing with things like real fluids or materials under stress.
Kai: Exactly. It’s not just about getting an estimate; it’s about characterizing the fluid accurately enough to use it in a bigger model. The authors are proposing two specific new methods to do this, which is what they call non-equilibrium molecular dynamics simulations of steady-state systems with periodic boundary conditions.
Mira: And the implication there is that standard equilibrium simulations using Green-Kubo relations just don't work cleanly when you have those random and dissipative forces mixed in; so they need a new way to look at it.
The paper's summary: Kai: So what the paper actually does is introduce two novel NEMD techniques for measuring bulk viscosity zeta from these steady-state systems. One method is called the particle transfer method, and the other is the force driven method. These methods create a divergent flow field by manually displacing particles or by applying an external bulk force with an inhomogeneous profile.
Mira: That sounds like they are solving the problem of creating that steady state flow that you need to measure viscosity from, which is a big hurdle in these types of simulations. The particle transfer method is most efficient when you're dealing with particles interacting via soft potentials.
Lev: Creating a steady-state divergent flow is tough; usually, the momentum transport and the direction of flux become parallel, so you need that extra mass transport to keep things going, which this paper addresses by suggesting these two ways to achieve it.
Kai: Right. And then they test these methods against the Green-Kubo relations using a dissipative particle dynamics fluid to see how well they actually work in practice. They compare their new NEMD results with the theoretical expressions derived from the literature.
Mira: That comparison is key because it lets them see where those different theoretical formulas for transport coefficients actually agree, which is what they call generalized Green-Kubo expressions that try to separate the stochastic contribution and should give better statistics.
The paper's improvements: Kai: The paper suggests two main improvements. First, these new NEMD techniques are proposed to compute bulk viscosity zeta from steady-state systems with periodic boundary conditions, which is a big step for getting those values reliably.
Mira: And the second improvement is testing the Green-Kubo relations using a dissipative particle dynamics fluid. This allows them to compare their simulation results against theoretical predictions, and they found that the bulk viscosity results obtained with the formula number twenty-three are in excellent agreement with what they measured via NEMD.
Lev: That agreement is good, but we have to look at the discrepancy; they note a constant offset of about zero point three when comparing their simulation results for shear viscosity to the theoretical prediction. That’s a detail that engineers have to account for if you're building hardware based on these simulations.
Kai: And then they explain why there is a gap in bulk viscosity theory—they say it can be resolved by neglecting the kinetic contribution in the theoretical prediction because that part only reflects energy fluctuations to pressure, not the viscous stress tensor.
Mira: That makes sense from a theoretical viewpoint, but it also means they found a way to get closer to agreement by focusing on those generalized Green-Kubo expressions that explicitly account for the fact that the dissipative force is not invariant under time reversal symmetry.
Conclusion: Kai: So, to wrap up this paper, they’ve shown two new ways—the particle transfer method and the force driven method—to compute bulk viscosity from steady-state systems with periodic boundary conditions. They also verified that their results for bulk viscosity using the formula number twenty-three match their NEMD simulations really well.
Mira: The big implication is that they can now get a better statistical separation of the stochastic and dissipative contributions when testing those generalized Green-Kubo expressions, which is something previous attempts haven't managed yet.
Lev: For us on the hardware side, it means these methods could potentially be used to compute bulk viscosities both in Hamiltonian systems and in dissipative systems that have local momentum conservation, which broadens the applicability of these simulation tools significantly.
Kai: It will be interesting to test this stuff on other systems, like an isothermal Lennard-Jones system with high densities and hard-core potentials. And these NEMD methods are flexible enough for both Hamiltonian and dissipative setups now.
Mira: Indeed, the work supports the validity of that generic expression number twenty-two, which includes the contribution of stochastic forces and accounts for the lack of time reversal symmetry in dissipative force dynamics. It’s a solid piece of work for getting those transport coefficients right.
Lev: I just think it opens up new avenues for testing these simulation methods on different physical systems that might be more relevant to real-world applications.
More episodes
- 2610.32678-Probing the limits of materials design via a comprehensive crystal database of 14.6 million structures with 140 properties
- 2610.33595-Native-oxide-driven layer asymmetry, two-step switching and exchange bias in Co/Ir/Co trilayers
- 2503.19572-Spin models from nonlinear cellular automata
- 2606.23350-Irreversibility Enhances Quantum-Enhanced Markov-Chain Monte Carlo
- 2608.02744-Two-Parameter Ansatz for the Violation of Eigenstate Thermalization
- 2210.00806-Low- and high-energy localization landscapes for tight-binding Hamiltonians in two-dimensional lattices
- 2409.09625-Selective switch for band-edge alignments in pentagonal CdSeTe monolayer: Atom-valley locking
- 2606.20168-Norms, overlaps and Yangian descendants for the Haldane-Shastry spin chain
- 2512.19503-Strain-induced splitting of the CCDW-NCCDW phase transition in 1T-TaS 2
- 2307.12638-Strong localization of microwaves beyond two dimensions in aperiodic Vogel spirals