Lattice Boltzmann model for non-ideal compressible fluid dynamics

Basic Information

• Paper ID: 2510.14712
• Title: Lattice Boltzmann model for non-ideal compressible fluid dynamics
• Authors: S. A. Hosseini, M. Feinberg, I. V. Karlin (ETH Zurich)
• Classification: physics.flu-dyn
• Published Journal: Journal of Fluid Mechanics
• Paper Link: https://arxiv.org/abs/2510.14712

Abstract

This paper proposes a lattice Boltzmann method for simulating compressible non-ideal fluid flows. The method employs a first-nearest-neighbor lattice and introduces a consistent set of correction terms through a quasi-equilibrium attractor, ensuring positive-definite and Galilean-invariant Navier-Stokes dissipation rates. This construction avoids the need for extended templates or temporary regularization while maintaining numerical stability and thermodynamic consistency across a wide range of flow states. The resulting model accurately reproduces fluid mechanics at both the Euler and Navier-Stokes hierarchical levels. As rigorous validation, the authors demonstrate for the first time quantitatively accurate simulation of droplet-shock interactions with Mach numbers up to 1.47 within the lattice Boltzmann framework.

Research Background and Motivation

Problem Definition

Non-ideal compressible fluid dynamics is an emerging and rapidly developing branch of fluid mechanics, primarily driven by the emergence of methods and technologies operating at near-critical, transcritical, and supercritical conditions. Such fluids include:

1. Supercritical fluids: Fluids at pressures and temperatures above the critical point
2. Dense vapors: Gases with compressibility factors deviating from ideal gas behavior
3. BZT fluids: Complex polymeric fluids with negative fundamental gas dynamic derivatives

Significance

1. Industrial application demands: Energy technologies such as organic Rankine cycles and supercritical CO₂ turbines
2. Scarcity of experimental data: Complex experimental setups and difficulty in data acquisition
3. Need for numerical tools: Requirement for consistent and efficient numerical tools to understand complex physical phenomena

Limitations of Existing Methods

1. Traditional CFD methods: In thermodynamically unstable spinodal regions, the square of the adiabatic sound speed becomes negative, causing evolution equations to transition from hyperbolic to parabolic type
2. Existing LBM models: Primarily designed for boiling applications, lacking systematic models for compressible non-ideal flows
3. Numerical stability issues: Stability challenges at high Mach numbers and strong non-ideal effects

Core Contributions

1. Proposed a novel two-distribution-function LBM model: Implements non-ideal compressible flow simulation using the classical first-nearest-neighbor lattice (D3Q27)
2. Introduced quasi-equilibrium attractor mechanism: Ensures positive-definite and Galilean-invariant dissipation rates through correction terms fᵢ and gᵢ
3. Achieved independent bulk viscosity control: Addresses the problem of potentially negative bulk viscosity in BGK structures
4. First implementation of high-Mach-number droplet-shock interaction simulation: Achieves quantitatively accurate simulation at Ma=1.47 within the LBM framework
5. Maintained thermodynamic consistency: Preserves numerical stability across a wide range of flow states

Methodology Details

Problem Formulation

Simulation of compressible non-ideal fluids satisfying the Navier-Stokes-Korteweg (NSK) equation system:

• Mass conservation: ∂ₜρ + ∇·(ρu) = 0
• Momentum conservation: ∂ₜ(ρu) + ∇·(ρu⊗u) + ∇P + ∇·T^NS + ∇·T^K = 0
• Energy conservation: ∂ₜ(ρE) + ∇·((ρE+P)u) + u·(∇·T^K) + ∇·(u·T^NS) + ∇·q = 0

where T^K is the Korteweg surface tension tensor and T^NS is the viscous stress tensor.

Model Architecture

1. Two-Distribution-Function Design

Two sets of distribution functions {fᵢ, gᵢ}, i=1,...,27:

• fᵢ: Handles mass and momentum conservation
• gᵢ: Handles energy conservation

2. Lattice Boltzmann Equation

fᵢ(x+cᵢδt, t+δt) = fᵢ + 2β(fᵢᵉᵠ - fᵢ) + (1-β)(fᵢ* - fᵢᵉᵠ)
gᵢ(x+cᵢδt, t+δt) = gᵢ + 2β(gᵢᵉᵠ - gᵢ) + (1-β)(gᵢ* - gᵢᵉᵠ)

where β = Pδt/(2μ + Pδt) is the relaxation parameter.

3. Equilibrium Distribution Functions

Adopts a product form:

fᵢᵉᵠ = ρ ∏_α Ψᵢα(uα, P/ρ + uα²)

where Ψᵢα is the basis function:

Ψᵢα(ξα, ζαα) = 1 - cᵢα² + ½[(3cᵢα² - 2)ζαα + cᵢαξα]

4. Quasi-Equilibrium Correction Terms

f-distribution function corrections:

• Force term: Fα = -κρ∂α∇²ρ (Korteweg force)
• Stress correction: Φαα includes diagonal stress tensor corrections and independent bulk viscosity control

g-distribution function corrections:

• Shifted velocity: u* = u + δt F/ρ
• Shifted temperature: T* = T - δt²F·F/(2ρ²cᵥ)
• Heat flux correction: Accounts for enthalpy gradient effects in non-ideal gases

Technical Innovations

1. Quasi-Equilibrium Attractor Mechanism

By introducing fᵢ and gᵢ, the system does not simply relax toward local equilibrium but toward a corrected quasi-equilibrium state, ensuring:

• Correct macroscopic dynamics
• Numerical stability
• Galilean invariance

2. Independent Bulk Viscosity Control

In traditional BGK models, bulk viscosity η' = ((D+2)/D - ρcs²/P)μ can be negative. Independent control is achieved through correction term Φ':

Φ' = P((D+2)/D - ρcs²/P - η/μ)(∇·u)

3. Thermodynamic Consistency

Carefully designed shifted temperature T* and energy corrections ensure thermodynamic consistency under nonlinear equations of state.

Experimental Setup

Validation Cases

1. Fundamental verification:
   • Sound speed measurement
   • Shear viscosity verification
   • Thermal Couette flow
   • Normal mode dissipation
2. Multiphase flow verification:
   • Liquid-vapor coexistence density
   • Interface consistency and convergence
3. Compressible flow verification:
   • One-dimensional non-ideal shock tube
   • Shock-liquid column interaction

Material Parameters

Primarily uses nitrogen (N₂) for verification:

• Critical pressure: Pc = 3.4×10⁶ Pa
• Critical density: ρc = 241.96 kg/m³
• Critical temperature: Tc = 126.2 K
• R/cv = 0.4

Equation of State

Employs van der Waals equation of state:

P(ρ,T) = ρRT/(1-bρ) - aρ²

Experimental Results

Main Results

1. Fundamental Physical Quantity Verification

• Sound speed: Perfect agreement with analytical solutions on saturated liquid and vapor branches for Tr∈0.7,1
• Shear viscosity: Maintains Galilean invariance at different Mach numbers with error <1%
• Thermal Couette flow: Consistent with analytical solutions for Pr∈{0.6,1.2,4.9} and Ma∈{0.8,1.2,1.6}

2. Multiphase Flow Properties

• Coexistence density: Complete agreement with Maxwell equal-area construction for Tr∈0.3,0.99
• Interface convergence: Exhibits good convergence characteristics at different grid resolutions

3. Non-Ideal Shock Tube

Successfully simulates three different non-ideal shock tube configurations:

• Case I: Rarefaction shock in Γ<0 region
• Case II: Classical compression wave behavior
• Case III: Mixed rarefaction wave dynamically crossing the Γ=0 line

All cases show excellent agreement with reference data.

4. Shock-Droplet Interaction

• First implementation of high-Mach-number simulation at Ma=1.47 within the LBM framework
• Accurately captures complex wave structures including incident shock, transmitted wave, reflected shock, and Mach stem
• Droplet deformation evolution shows high consistency with experimental data

Ablation Studies

Convergence verification through interface testing at different grid resolutions:

• Four resolution levels from δx = 5μm to 0.1μm
• Exhibits good grid convergence characteristics

Key Findings

1. Numerical stability: Maintains stability at high Mach numbers and strong non-ideal effects
2. Computational efficiency: Uses standard first-nearest-neighbor lattice with controllable computational cost
3. Physical fidelity: Accurately reproduces complex physical phenomena of non-ideal fluids

Related Work

Traditional LBM Multiphase Flow Models

1. Color-gradient model (Gunstensen et al. 1991)
2. Pseudopotential model (Shan & Chen 1993)
3. Free-energy model (Swift et al. 1996)

Development of Compressible LBM

• He & Doolen (2002): First attempt at compressible non-ideal flows
• Existing models primarily designed for boiling applications
• Lack of systematic models for compressible non-ideal flows

Advantages of This Work

1. Systematic approach: Complete two-distribution-function framework
2. Theoretical foundation: Rigorous multiscale analysis
3. Complete verification: Comprehensive validation from fundamental physical quantities to complex flow phenomena

Conclusions and Discussion

Main Conclusions

1. Successfully developed an LBM model applicable to compressible non-ideal fluids
2. Achieved numerical stability and physical consistency through quasi-equilibrium attractor mechanism
3. First implementation of high-Mach-number droplet-shock interaction simulation within the LBM framework
4. Model maintains thermodynamic consistency across a wide range of flow states

Limitations

1. Single-relaxation-time model: Currently uses simple BGK collision model, potentially insufficient for turbulent flow simulation
2. Equation of state restrictions: Primarily verified van der Waals equation; other complex equations of state require further verification
3. Limited three-dimensional verification: Most verification concentrated on one- and two-dimensional configurations

Future Directions

1. Extension to more robust collision models: For turbulent flow simulations
2. More complex equations of state: Such as Peng-Robinson
3. Large-scale three-dimensional applications: Application to practical engineering problems

In-Depth Evaluation

Strengths

1. Theoretical rigor: Complete multiscale analysis and mathematical derivation
2. Methodological innovation: Quasi-equilibrium attractor mechanism represents important innovation
3. Comprehensive verification: Systematic validation from fundamental physics to complex applications
4. Practical value: Addresses important engineering application problems
5. Numerical efficiency: Uses standard lattice avoiding computational complexity

Weaknesses

1. Simple collision model: BGK model has limited applicability to complex flows
2. Parameter dependency: Lack of systematic guidance for selection of certain correction parameters
3. Insufficient three-dimensional verification: Requires more verification on three-dimensional complex geometries

Impact

1. Academic contribution: Opens new directions for LBM in compressible non-ideal flow field
2. Engineering value: Provides important numerical tools for supercritical fluid technology
3. Methodological significance: Quasi-equilibrium attractor concept may influence other numerical methods

Applicable Scenarios

1. Supercritical fluid technology: CO₂ turbines, organic Rankine cycles
2. Multiphase flows: Droplet breakup, phase change processes
3. Shock dynamics: Complex wave interactions in high-speed flows
4. Direct numerical simulation: Fundamental physics research requiring high precision

References

1. Guardone, A. et al. (2024). Nonideal compressible fluid dynamics of dense vapors and supercritical fluids. Annual Review of Fluid Mechanics, 56, 241-269.
2. Hosseini, S.A. & Karlin, I.V. (2023). Lattice Boltzmann for non-ideal fluids: Fundamentals and practice. Physics Reports, 1030, 1-137.
3. Shan, X. & Chen, H. (1993). Lattice Boltzmann model for simulating flows with multiple phases and components. Physical Review E, 47(3), 1815.