Session W40: DFD IX

Quantifying non-gaussian character of motion of polymer fragments during degradation of hydrogels

Via mesoscale simulations, we characterize the process of photo-controlled degradation of spherical nanogels in a good solvent. We show that degradation can be used to dynamically tailor size, shape, and transport properties of these soft particles. We use Dissipative Particle Dynamics (DPD) approach with an adapted form of the modified Segmental Repulsive Potential (mSRP). To characterize degradation process, we track the structural and dynamic characteristics of the remnant nanogel and distribution of the broken-off fragments. To quantify spatial fluctuations in local dynamic behavior, we calculate the self-part of van-Hove correlation function for the network junction points and that for the reactive beads forming degradable bonds. We demonstrate large deviations from the Gaussian behavior during the degradation process. Further, we quantify variations in non-Gaussian character with the extent of degradation reaction for nanogel particles of various sizes and crosslink densities. Our study provides insights into using controlled degradation to dynamically tune shapes and transport properties of hydrogel nanocarriers.   

Nanoparticle dynamics in fully synthetic biomimetic analogues

Adding end-functionalized polymers to an emulsion produces hierarchical structures that percolate the material, significantly augmenting the material’s elasticity and introducing a finite yield stress. As a result, we find that this class of materials successfully replicate the structure, mechanics, and dynamics of soft, cellularized tissues. Here, we explore how nanoparticles transport through these polymer-linked emulsions with the goal of understanding transport in complex, biological systems. We vary the nanoparticle size, polymer concentration, and polymer molecular weight and evaluate how these properties affect transport properties, including long-time diffusivity, short-time localizations, and non-Gaussian distributions of displacements. Small particles exhibit faster-than-expected diffusion whereas large particles couple to the material viscoelasticity. Furthermore, nanoparticles readily explore space for samples prepared with a high molecular weight linker but are preferentially located in low permeability zones for emulsions containing a low molecular weight linker. We attribute these differences to the bridging probability of the polymers. Our findings elucidate how nanoparticle transport depends on structure, dynamics, and mechanics in dense suspensions and biological environments.   

Interaction between swarming active matter and flow: the impact on Lagrangian coherent structures

In recent years, research topics concerning active matter have attracted interest from diverse communities. It has been suggested that active matter-as represented by organisms such as zooplankton-has great potential in ocean mixing due to its intrinsic mobility and the sheer amount of biomass. However, prior investigations have predominantly overlooked the influence of external background flow, despite the ubiquity of flow driven by various sources in nature. The interaction between active matter and external flow structures has long been neglected. Here, we conducted experiments using a typical centimeter swimmer, A. salina, and an electromagnetically driven quasi-two-dimensional flow to study the interaction between active matter and flow. We focused on the impact of swarming active matter on hyperbolic Lagrangian coherent structures (LCSs) that mark the most straining regions in the flow. There is one decade of scale separation between active matter agents and the length scale of LCSs. We illustrated that the impact of active matter on LCSs was much more significant compared to localized random noise with similar energy input. In addition, we revealed that the perturbation generated by active matter could couple with the background flow and further deform the LCSs. In addition to the impact on the most straining hyperbolic regions of flow, we also revealed that the rotational elliptical region of the flow was much more susceptible to active matter perturbation. We further described how the influence of active matter changed with their number densities and background flow intensities. We revealed that the LCSs could be decently altered even at a small number density of active matter. Through this work, we aim to provide valuable insights and draw attention to the problem regarding the interaction between active matter and external flow structures.   

Diffusioosmotic dispersion of solute in a long, narrow channel

Solute-surface interactions have garnered considerable interest in recent years as a novel control mechanism for driving unique fluid dynamics and particle transport with potential applications in fields such as biomedicine, the development of microfluidic devices, and enhanced oil recovery. In this study, we will discuss dispersion induced by the diffusioosmotic motion near a charged wall in the presence of a solute concentration gradient. Here, we introduce a plug of salt with a Gaussian distribution at the center of a channel with no background flow. As the solute diffuses, the concentration gradient drives a diffusioosmotic slip flow at the walls, which results in a circulating flow in the channel; this, in turn, drives an advective flux of the solute concentration and alters the effective diffusivity of the solute as it diffuses along the channel. We will present theoretical predictions for the solute dynamics using a multiple-timescale analysis. Furthermore, we will show a cross-sectionally averaged concentration equation with an effective diffusivity and comment on its application in diffusioosmotic transport.   

Microfluidic Experiments Demonstrate the Effects of Flow Fluctuations on Biofilm Growth

Biofilms play critical roles in various environmental processes within porous media, influencing nutrient cycling, bioremediation, and hydrogeological transport. Understanding the dynamics of biofilm formation and growth is essential for predicting and controlling biofilm properties in soil, yet such process has not been fully understood. The majority of current studies have focused on the impact of steady flows on biofilm growth, while flow fluctuations are commonly encountered in porous media, especially soil. In this study, we investigated the effects of flow fluctuations on the growth of biofilms of a soil bacterium Pseudomonas putida through a combination of microfluidic experiments and theoretical models. Our experimental results revealed that biofilm growth under fluctuating flow conditions followed three distinct phases: lag, exponential, and fluctuation phases, in contrast to the four phases under steady flow conditions (lag, exponential, stationary, and decline phases). We demonstrated that low-frequency fluctuations promoted biofilm growth, while high-frequency fluctuations inhibited its development. We attributed the contradictory impacts of flow fluctuations on biofilm growth to the adjustment time (<!--[if gte msEquation 12]> style='font-family:"Cambria Math","serif";mso-ascii-font-family:"Cambria Math";
mso-hansi-font-family:"Cambria Math";mso-bidi-font-family:Times;mso-fareast-language:
ZH-CN;mso-bidi-font-weight:bold'> style='mso-bidi-font-style:normal'>T _{style='mso-bidi-font-style:normal'>0} T₀) needed for biofilm to grow after the shear stress changed from high to low. Furthermore, we developed a theoretical model that explains the observed biofilm growth under fluctuating flow conditions. Our insights into the mechanisms underlying biofilm development under fluctuating flows can inform the design of strategies to control biofilm formation in soil and other porous media.   

Simulations of Sphere Sedimentation in Viscoelastic Fluids using openFOAM: Notes on Boundary Conditions

Sphere sedimentation is an important benchmark problem in fluid dynamics and (for certain viscoelastic fluids) an active topic of ongoing research. In particular, sphere sedimentation in Wormlike micelles (WLM) often shows instabilities with dramatic and irregular fluctuations in the sedimentation velocity. These experiments have motivated several computational fluid dynamics (CFD) studies in the past few years, but in this talk we will argue that the selection of boundary conditions should be taken more seriously. Paradoxically, it has been common to study unsteady sphere sedimentation by enforcing a constant sedimentation velocity - this is partly a choice of convenience, as the alternative (enforcing momentum balance on the sphere) is more difficult to set up in a general purpose CFD solver like openFOAM or rheoTool. In this talk, we discuss a workaround in the openFOAM framework to create boundary conditions with more general control schemes. Time permitting, we will also compare predictions for slip vs no slip along the bounding cylinder (though which the sphere is falling), where the no-slip boundary condition provides a means of mitigating confinement effects.   

Decay and propagation of an isolated turbulent blob

We create and sustain an isolated blob of turbulence by repeatedly firing together vortex loops. In the steady state, our PIV and 3D PTV measurements reveal that the blob consists of a turbulent core ($Re_{lambda}=50-300$) surrounded by comparatively quiescent fluid. The properties of the vortex loops determine the turbulent intensity and the scales of motion within the blob. When the injection of vortex rings stops, a spherical front that separates the turbulent core from the quiescent surroundings, begins to propagate within the chamber, and the turbulence decays. This turbulence endures throughout the decay process, lasting more than fifteen minutes, as evidenced by the energy spectrum. Through experimental comparison of turbulence induced by different methods within the same chamber, we demonstrate that the large-scale turbulence motion dictates the decay law of energy. Using a simple low-order closure model, we create a spatially-extended description of turbulence propagation and decay. We then compare its energy profile predictions and decay law with the data. Crucially, the observed turbulent front shows qualitative features of the non-diffusive transport captured by this mean field theory, as opposed to simple diffusion. This model also provides a consistent explanation for the different decay laws observed in the experiments.   

Dynamical balances in transitional and self-similar Rayleigh-Taylor mixing layers

We study the dynamical balances in the pre-virtual origin and the self-similar Rayleigh-Taylor mixing layer to contrast the pre-transition and the self-similar layer. We employ two narrow-band initial conditions for the density interface, with low and high non-dimensional amplitudes or equivalently the interface slope. A large interface slope skips the linear growth phase and meaningfully reduces the virtual origin as the perturbations take a rapid nonlinear and alternate speedier route to transition; via ballistic ejecta as opposed to bubble and spikes. In this regime, the early growth is linear in nature as opposed to the early quadratic growth prevalent in the conventional bubble-spike growth picture. This pushes forward the virtual origin for the self-similar development for an identical perturbation length with a smaller interface slope. We show that the virtual origin of the self-similar growth is fundamentally connected to the transition metrics. With the two transition routes, we investigate how the memory of the perturbations considered in the study fades past the virtual origin and if any aspects persist in the late time large-scale structures. Self-similar mass flux dynamics in the variable-density turbulence is also contrasted with the dynamics in Boussinesq turbulence.   

Enabling a flow agnostic LES approach using deep learning

Despite the sustained growth in computing power, direct numerical simulations (DNS) of turbulence--where the entire range of scales are resolved--remain prohibitively expensive for natural and engineering flows. An attractive alternative is large eddy simulation (LES), where the large scales are resolved, while the entire spectrum of small scales is modeled, significantly boosting computational performance by nominally sacrificing on accuracy. Recent advancements in machine learning techniques have further bolstered LES closures. Nevertheless, considerable variability and deficiencies persists in LES models, particularly in applications such as scalar mixing, particle transport, and combustion, where small scales play a pivotal role. Here, we present a novel modeling approach for LES, where instead of modeling the entire range of small scales, a multilevel approach is adapted. Using tensor representation theory, a general function closure is obtained at each level in terms of filtered velocity gradients. A reduced order generalization is then obtained by employing autoencoders and training based on DNS data from a variety of turbulent flows. The approach outlined here results in a versatile, flow-agnostic LES closure that can be trained and applied to progressively more complex turbulent flows using deep neural networks.   

Physics Informed Contrastive Learning for Homogeneous Partial Differential Equation Surrogate Modeling

Neural operators have recently grown in popularity as Partial Differential Equation (PDEs) surrogate models. Learning mappings between function spaces, rather than individual functions, has proven to be a powerful approach to calculate fast, accurate solutions to complex PDEs. While much work has been done evaluating neural operator performance on a wide variety of surrogate modeling tasks, these works normally evaluate performance on a single equation at a time. In this work, we develop a novel contrastive pretraining framework utilizing Generalized Contrastive Loss that improves neural operator generalization across multiple governing equations simultaneously. Governing equation coefficients are used to measure ground-truth similarity between systems that adds weighting to positive and negative samples. A combination of physics-informed system evolution and latent-space model output are anchored to input data and used in our distance function. We find that physics-informed contrastive pretraining improves both accuracy and generalization for the Fourier Neural Operator in fixed-future, and autoregressive rollout tasks for the 1D and 2D Heat, Burgers, and linear advection equations.   

Matrix Product State Simulation of Reactive Flows

The matrix product state (MPS) representation, developed for approximating the state of quantum many-body systems, exploits their correlation structure to accurately capture the underlying physics in a low-rank form (i.e., in a massively reduced state space). Here, the methodology is employed for simulating chemically reacting turbulent flows. In doing so, the governing differential operators representing compressible, reacting turbulent flows are recast in the context of MPS, and their dynamics is simulated with various degrees of truncation. Simulations are performed to assess the effects of the Reynolds number, the Mach number and chemical heat release ratios on the compositional structure of the flow. The results via MPS-reduced order solutions are appraised against those generated via direct numerical simulation of the same flows.   

Reconstruction of porous media geometry from sparse velocity measurements using convolution neural networks

The study of transport phenomena, using non-destructive imaging technologies and simulations of fluid flow through complex geometries, has become increasingly vital, particularly in the biomedical domain. Such approaches provide valuable insight into critical processes, including blood flow in tissues, drug delivery, and extracellular transport, contributing to our understanding of physiology and medical interventions. However, significant challenges persist in extracting geometric features from imaging and reconstructing the flow domain for direct simulations. Recent advancements in machine learning techniques offer a promising solution by regenerating domain geometry based on relatively sparse sampling of flow velocity. Motivated by this, we apply deep learning techniques to predict porous media geometry, utilizing convolutional neural networks (CNNs). A CNN is trained on the velocity field or particle trajectories acquired from 2D Lattice Boltzmann simulations. We conduct an analysis in which we vary the complexities of porous media geometry and attain promising estimations of the flow domain using sparse velocity measurements. Extending this methodology to experimental data will notably aid in the development of quantitatively accurate predictive models with biomedical applications.   

Monte Carlo Lattice Gas for Fluctuating Fluids

For small fluid systems (or close to a critical point) fluctuations are important in fluids. The reason for fluctuations can be traced back to the fact that nature is discrete. However, the current mainstream fluid simulation models do not have an ideal answer for fluctuations. The lattice Boltzmann, for example, needs fluctuations to be added externally using a Langevin approach. 

Lattice gas methods are inherently noisy, and a lot of work analyzing fluctuating systems was pe rformed with them. Their non-noisy limit, the lattice Boltzmann method, eventually replaced these hydrodynamic lattice gas methods because certain artifacts of the original Boolean lattice gases could be overcome. We have been able to show that integer lattice gases can be constructed that are equivalent to the lattice Boltzmann method. These lattice gases offer both inherent fluctuations and unconditional stability that lattice Boltzmann lacks.

We present here an efficient integer lattice gas method based on distribution sampling for the hydrodynamic regime, allowing hydrodynamic systems to now be simulated with inherent fluctuations at a numerical cost comparable to lattice Boltzmann methods.