Computations in Science Seminars

Previous Talks: 2018

Jan 2018
10
Wed 12:15
Heinrich Jaeger, University of Chicago
e-mail:
The plot thickens: From discontinuous shear thickening to shear jamming

Over the last few years dense suspensions of hard particles in a simple liquid have become a model system in the soft condensed matter, granular materials, and rheology communities for the investigation of strongly non-Newtonian behavior. A key aspect underlying the recent surge of activity has been the realization that in addition to hydrodynamic interactions direct frictional contact between particles can occur. In fact, friction forces were found to be essential in order to explain some of the most striking phenomena observed in dense suspensions, such as an abrupt, essentially discontinuous onset of shear thickening, whereby the viscosity can jump up by over an order of magnitude as a critical shear rate is exceeded. So far, however, practically all theoretical models and simulations that include friction have treated it as a phenomenological parameter without considering its molecular origin. Furthermore, most models treated a situation in which shear is applied continuously, under steady-state conditions. This prevented these approaches from capturing the remarkable dynamic phenomena observed in dense suspensions, most notably the propagating jamming fronts associated with the transition from a merely shear-thickened to a solid-like jammed state. Thus, despite much recent progress, there remain fundamental questions both at the nano-scale, about the nature of the frictional interactions, and at the macro-scale, about the relation between steady-state and transient dynamic phenomena.

I will discuss recent experiments from our group that address these questions, focusing on the differences between discontinuous shear thickening (DST) and shear jamming (SJ). These experiments show how particle surface chemistry can play a central role in creating conditions that allow for SJ. We find the system’s ability to form interparticle hydrogen bonds when sheared into contact elicits SJ. We demonstrate this with charge stabilized polymer microspheres and non-spherical cornstarch particles, controlling hydrogen bond formation with solvents. The propensity for SJ is quantified by tensile tests and can be linked directly to an enhancement of the effective frictional interactions between particles, as measured by AFM and also observed by mapping out the steady-state rheology as a function of packing fraction.

Jan 2018
17
Wed 12:15
James Evans, University of Chicago
e-mail:
Host: William Irvine ()
Organizer: Delphine Coursault ()
Social Limits to Understanding

I provide an overview of my research tracing several ways in which social connection between scientists, engineers and citizens shape the limits of what a population can collectively know. This includes empirical demonstrations of how centralized networks decrease the truth value of collective certainty in biomedicine, how large teams shrink the search space of science and technology, and how flocking correlates investigations and limits the size of future understanding. I then explore how the complex system of science, technology and society generates productive social disconnection to accelerate advance through maintaining crossable boundaries between disciplines, ideologies, and the ways in which recombination are valued. To explore this last point, I model scientific discovery and technological invention as involved in the complex combination of contents including problems, methods and physical entities, which bridge contexts such as journals, subfields and conferences from which scientists and inventors drawn them. We can model the normal growth of ideas in articles and inventions by representing them as complex combinations of scientific and technical contents and contexts with a high-dimensional stochastic block model, which predicts more than 95% of new patents and articles in biomedicine and physics. The inverse probability of published papers and patents under this model--unlikely combinations of contents across contexts--predicts nearly 50% of the likelihood of revolutionary success, measured by outsized citations and major awards. I discuss the implication of these findings for science policy and practice.

Jan 2018
24
Wed 12:15
Alberto Fernandez-Nieves, Georgia Institute of Technology
e-mail:
Host: William Irvine ()
Organizer: Delphine Coursault ()
Fire-ant fluids: expansion, mechanics and waves

Motivated by classic thermodynamic experiments with dilute fluids, we explore the free and constrained expansion of fire-ant aggregations. In the latter case, we confine the ants to 2D vertical columns; hence, as the ants expand, they do work against the gravitational field. Surprisingly, we often observe the spontaneous generation of density waves; these propagate at a speed that depends on both the width and the amplitude of the wave, and occur cyclically. We also perform experiments in horizontal cells and find that the ants exhibit activity cycles, where the density homogeneity and mechanical properties of the aggregation change with activity. We believe that these cycles together with the large ant densities in our vertical columns are responsible for the generation of the observed waves. Finally, since the average ant density is larger at the bottom of the vertical column than a the top, we follow our temptation and attempt at interpreting the results in lieu of sedimentation equilibrium to seek for an equation of state. Despite our results are still highly preliminary, they provide interesting phenomenology that could perhaps be seen in active systems other than fire-ant aggregations.

Jan 2018
31
Wed 12:15
Jens Eggers , University of Bristol
e-mail:
Host: William Irvine ()
Organizer: Glen Hocky ()
Pair creation, motion, and annihilation of topological defects in 2D nematics

We present a novel framework for the study of disclinations in two-dimensional active nematic liquid crystals, and topological defects in general. The order tensor formalism is used to calculate exact multi-particle solutions of the linearized static equations inside a planar uniformly aligned state, so that the total charge has to vanish. Topological charge conservation then requires that there is always an equal number of $q=1/2$ and $q=-1/2$ charges. Starting from a set of hydrodynamic equations, we derive a low-dimensional dynamical system for the parameters of the static solutions, which describes the motion of a half-disclination pair, or of several pairs. Within this formalism, we model defect production and annihilation, as observed in experiments. Our dynamics also provide an estimate for the critical density at which production and annihilation rates are balanced.

Feb 2018
7
Wed 12:15
Haim Diamant, Tel Aviv University
e-mail:
Host: Tom Witten
Organizer: Yuval Yifat ()
Hyperuniform dynamic structures of forced colloids

Several materials have been found in recent years to exhibit "hyperuniformity", their constituent particles being distributed in a disordered but correlated configurations such that density fluctuations are strongly suppressed. Examples are sheared suspensions and emulsions, and random jammed packings of particles. In those systems the hyperuniform structure appears as a static absorbing state. Suppression of density fluctuations appears also in thermodynamic and dynamic systems which sample many spatial configurations. Examples are electrolytes and suspensions of forced colloids. We relate the statistical hyperuniformity in these systems to conservation laws and long-range interactions. We demonstrate in detail the suppression of density and velocity fluctuations in forced suspensions of asymmetric objects. In this case we have identified the basic mechanism leading to hyperuniformity. We argue that for certain object shapes the same mechanism can act in the opposite direction, destabilizing the dynamics.

Feb 2018
14
Wed 12:15
Paul McEuen, Cornell University
e-mail:
Host: William Irvine ()
Organizer: Glen Hocky ()
Listening to a thermal guitar: Hearing (and seeing) the vibrations of a carbon nanotube

Carbon nanotubes reside at the boundary of macroscopic mechanics and polymer physics. They can be made into guitar-string resonators and operated in vacuum, yet they are flexible enough that thermal fluctuations completely alter their properties. Here we show that individual carbon nanotubes can be picked up and strained with micron-sized tweezers, allowing us to study these novel fluctuating strings in a variety of new ways. Experiments include recording the “sound” of the thermal vibrations of a nanotube and “seeing” its real-time motion with light. We find many surprising results, including strong coupling between vibrational modes and very long intrinsic dissipation times, as well as many unresolved mysteries. These results open the door to the science of thermally-fluctuating strings in the underdamped limit, with many opportunities for theory, computation, and experiment.

Feb 2018
21
Wed 12:15
Ivan I Smalyukh, University of Colorado Boulder
e-mail:
Host: William Irvine ()
Organizer: Delphine Coursault ()
Topological Solitons and Other Knots in Soft Matter

Topologically nontrivial fields and vortices frequently arise in classical and quantum field theories, plasmas, optics, cosmology, hard condensed matter and atomic systems. On the other hand, soft matter systems, such as colloids and liquid crystals, offer the complexity in degrees of freedom and symmetries that allow for probing topologically analogous phenomena on experimentally accessible scales. In my lecture, I will discuss how surfaces of colloidal knots and handlebodies interact with the liquid crystalline molecular alignment fields and how topological knot solitons can emerge as static field configurations within the chiral colloidal ferromagnets [1-5]. I will show how such synergistic combinations of topology and self-assembly paradigms can emerge as an exciting scientific frontier of topological soft matter.

[1] A. Martinez, L. Hermosillo, M. Tasinkevych and I.I. Smalyukh. Proc. Natl. Acad. Sci. U.S.A. 112, 4546-4551 (2015). [2] A. Martinez, M. Ravnik, B. Lucero, R. Visvanathan, S. Žumer and I. I. Smalyukh. Nature Materials 13, 258–263 (2014).[3] B. Senyuk, Q. Liu, S. He, R. D. Kamien, R. B. Kusner, T. C. Lubensky and I. I. Smalyukh. Nature 493, 200-205 (2013). [4] P. J. Ackerman and I. I. Smalyukh. Nature Mater 16, 426-432 (2017). [5] J.-S.B. Tai, P.J. Ackerman and I.I. Smalyukh. PNAS. doi:10.1073/pnas.1716887115 (2018).

Feb 2018
28
Wed 12:15
Michelle Driscoll, Northwestern University
e-mail:
Organizer: Glen Hocky ()
Mind the gap: a cascade of instabilities created by rotating beads near a floor

Does a rotating bead always spin in place? Not if that bead is near a surface: rolling leads to translational motion, as well as very strong flows around the bead, even quite far away. These flows strongly couple the motion of nearby microrollers (rotating beads), which leads to a rich variety of collective effects. Using experiments in tandem with large-scale 3D simulations, we have shown that driving a compact group of microrollers leads to a new kind of flow instability, whose wavelength is controlled not by the driving torque or the fluid viscosity, but a geometric parameter: the microroller's distance above the container floor. Furthermore, under the right conditions, stable, compact clusters we term "critters" can emerge from the unstable interface. Our simulations and experiments suggest that these critters are a stable state of the system, move much faster than individual rollers, and quickly respond to a changing drive. We believe that critters are unique in that they are clusters which form only with hydrodynamic interactions; no interparticle potentials are needed to create these structures. Furthermore, as compact, self-assembled structures which can easily be remotely guided, critters may offer a promising tool for microscopic transport.

Mar 2018
14
Wed 12:15
Roy Beck-Barkai , Tel-Aviv University
e-mail:
Host: Tom Witten
Organizer: Peter Chung ()
From kB to kB: Universal and efficient entropy estimation using a compression algorithm

Entropy and free-energy estimation are key in thermodynamic characterization of simulated systems ranging from spin models through polymers, colloids, protein structure, and drug-design. Current techniques suffer from being model specific, requiring abundant computation resources and simulation at conditions far from the studied realization. In this talk, I will present a novel universal scheme to calculate entropy using lossless compression algorithms and validate it on simulated systems of increasing complexity. Our results show accurate entropy values compared to benchmark calculations while being computationally effective. In molecular-dynamics simulations of protein folding, we exhibit unmatched detection capability of the folded states by measuring previously undetectable entropy fluctuations along the simulation timeline. Such entropy evaluation opens a new window onto the dynamics of complex systems and allows efficient free-energy calculations.

Mar 2018
28
Wed 12:15
Ken Kamrin, MIT
e-mail:
Host: William Irvine ()
Organizer: Yuval Yifat ()
Modeling flowing granular material as a continuum: Surprising complexity meets surprising simplicity

Granular materials are common in everyday life but are historically difficult to model. This has direct real-world ramifications owing to the prominent role granular media play in multiple industries and in terrain dynamics. One can attempt to track every grain with discrete particle methods, but realistic systems are often too large for this approach and a continuum model is desired. However, granular media display unusual behaviors that complicate the continuum treatment: they can behave like solid, flow like liquid, or separate into a “gas”, and the rheology of the flowing state displays remarkable subtleties.

To address these challenges, in this talk we develop a family of continuum models and numerical solvers, which permit quantitative modeling capabilities for general problems and certain reduced-order approaches for problems of intrusion, impact, driving, and locomotion in grains. To calculate flows in general cases, a rather significant nonlocal effect is evident, which is well-described with our recent nonlocal model accounting for grain cooperativity within the rheology. On the other hand, to model just intrusion forces on submerged objects, we will show, and explain why, many of the experimentally observed results can be captured from a much simpler tension-free frictional plasticity model. This approach gives way to some surprisingly simple general tools, including the granular Resistive Force Theory, and a broad set of scaling laws inherent to the problem of granular locomotion. These scalings are validated experimentally and in discrete particle simulations suggesting a new down-scaled paradigm for granular locomotive design, on earth and beyond, to be used much like scaling laws in fluid mechanics. We close with ongoing efforts expanding into wet granular flows, multi-scale approaches, and self-optimizing wheels for off-road traction.

Apr 2018
4
Wed 12:15
Benjamin B. Machta, Yale University
e-mail:
Host: Arvind Murugan ()
Organizer: Peter Chung ()
When and why is a simpler model better?

Science is filled with toy models: abstractions of complicated systems that ignore microscopic details even when they are known. For a special class of models in physics, the renormalization group rigorously justifies the use of effective theories containing just a small number of relevant parameters. This philosophy seems to apply more broadly, even when the renormalization group cannot be used. But why? In this talk I will discuss an information theory approach to answering this question, or at least towards quantifying it. I will first review that typical models are sloppy, defined by a hierarchy in parameter importance. I will argue that sloppiness is both necessary and sufficient for a microscopic system to be amenable to description by a simpler effective theory. I will then show how renormalizable models become sloppy as their data is coarse-grained. Finally I will discuss our recent efforts to use the structure of these models to choose simpler effective theories automatically.

Apr 2018
11
Wed 12:15
Michael Shelley, Flatiron Institute, Simons Foundation,
and Courant Institute (NYU)
e-mail:
Host: William Irvine ()
Organizer: Glen Hocky ()
Active Mechanics in the Cell

Many fundamental phenomena in eukaryotic cells -- nuclear migration, spindle positioning, chromosome segregation -- involve the interaction of often transitory structures with boundaries and fluids. I will discuss the interaction of theory and simulation with experimental measurements of active processes within the cell. This includes understanding the force transduction mechanisms underlying nuclear migration, spindle positioning and oscillations, as well as how active displacement domains of chromatin might be forming in the interphase nucleus.

Apr 2018
18
Wed 12:15
Alfred Crosby, University of Massachusetts Amherst
e-mail:
Host: Heinrich Jaeger
Organizer: Steven Strong ()
Materials Mechanics for Impulsive Movement

Nature provides amazing examples of high velocity, high acceleration, impulsive movements that can be repeated numerous times over the course of an organism’s lifetime. Synthetic, or engineered, devices, on the other hand, are often challenged to achieve comparable performance across a wide range of size scales. Common to nature’s examples, including mantis shrimp and trap-jaw ants, is the integration of three essential components for elasticity-assisted movement: an actuator, spring, and latch. Elasticity-assisted motion has been utilized for thousands of years to amplify the power of natural or synthetic actuators; however, the scaling physics of these multi-component systems, especially in light of materials design, have not been widely considered. Here, we discuss our group’s efforts, within a multi-university collaborative team, to lay a foundation for understanding the role that materials properties and structure play in the performance of impulsive systems in nature with an eye toward aiding the development of engineered devices that can overcome current limitations. We first discuss the mechanics of elastic recoil and a set of systematic experiments on a resilin-like synthetic material. The results from this study leads to a common framework for describing the roles of geometry and materials properties for controlling duration, velocity, and acceleration. We then introduce recent advances of using mesoscale polymers, which build upon previously introduced concepts from our group, to develop high rate, large strain, microscale actuators. Collectively, these examples highlight the integrative approach of our group and how we use bio-inspired materials mechanics to inspire new technologies and provide fundamental insight.

Apr 2018
25
Wed 12:15
Martin van Hecke, Leiden Institute of Physics
e-mail:
Host: Arvind Murugan ()
Organizer: Delphine Coursault ()
Sequential Mechanical Metamaterials

Ordered sequences of motions govern the morphological transitions of a wide variety of natural and man-made systems, while the ability to interpret time-ordered signals underlies future smart materials that can be (re)programmed and process information. After a short introduction to mechanical metamaterials, we introduce here two novel classes of mechanical metamaterials, that can (1) exhibit sequential output and (2) are sensitive to sequential input. To obtain metamaterials that translate a global uniform compression into a precise multistep pathway of reconfigurations, we combine strongly nonlinear mechanical elements with a multimodal hierarchical structure, and demonstrate multi-step reconfigurations of digitally manufactured metamaterials. To obtain metamaterials that are sensitive to a sequence of mechanical inputs, we introduce the notion of non-commuting metamaterials. Our work aims to establish generic principles for infusing metamaterials with sequential input and output.

May 2018
2
Wed 12:15
Ned Wingreen, Princeton University
e-mail:
Host: Arvind Murugan ()
Organizer: Glen Hocky ()
Magic numbers in protein phase transitions

Biologists have recently come to appreciate that eukaryotic cells are home to a multiplicity of non-membrane bound compartments, many of which form and dissolve as needed for the cell to function. These dynamical "condensates" enable many central cellular functions – from ribosome assembly, to RNA regulation and storage, to signaling and metabolism. While it is clear that these compartments represent a type of separated phase, what controls their formation, how specific biological components are included or excluded, and how these structures influence physiological and biochemical processes remain largely mysterious. I will discuss recent experiments on phase separated condensates both in vitro and in vivo, and will present theoretical results that highlight a novel “magic number” effect relevant to the formation and control of two-component phase separated condensates.

May 2018
9
Wed 12:15
Leigh Orf , University of Wisconsin
e-mail:
Host: William Irvine (), Morgan O'Neill ()
Organizer: Peter Chung ()
Simulating and visualizing the most devastating thunderstorms

Tornadoes are among nature's most destructive forces. The most violent, long-lived tornadoes form within supercell thunderstorms. In this seminar, results from simulations of tornado-producing supercell thunderstorms at ultra-high resolution will be presented, as well as the technical challenges involved in conducting, analyzing and visualizing such simulations.

In a control simulation, tornado formation occurs in concert with processes not clearly seen in previous supercell simulations. Visualizations of model fields presented at very high temporal resolution (up to 1/6 of a second, the model time step) will be presented. These animations reveal a fascinating combination of processes that lead to the formation of a long lived very violent tornado that persists for well over an hour. In order to facilitate the saving and visualizing of large amounts of model data, a file system was developed that utilizes the HDF file format and ZFP lossy file compression, and this file system and associated middleware, dubbed LOFS, will be also described.

May 2018
16
Wed 12:15
Ariel Amir, Harvard
e-mail:
Host: Arvind Murugan ()
Organizer: Steven Strong ()
From single-cell variability and correlations across lineages to the population growth

Cells of all domains of life must coordinate their cell cycle meticulously in order for protein levels, cell size and DNA replication to be appropriately regulated and to be able to prevent stochastic fluctuations from accumulating over time. These control mechanisms will lead to correlations in various cellular traits across the lineage tree (notably, size and generation times). I will present a recent model we developed for understanding cellular homeostasis and characterizing these correlations and fluctuations. I will discuss the implications of these correlated fluctuations on the population growth. In contrast to the dogma, we find that variability may be detrimental to the population growth, suggesting that evolution would tend to suppress it.

May 2018
23
Wed 12:15
Niall M. Mangan, Northwestern University
e-mail:
Host: Arvind Murugan ()
Organizer: Yuval Yifat ()
Identification of Hybrid Dynamical Systems via Clustering and Sparse Regression

Inferring the structure and dynamical interactions of complex systems is critical to understanding and controlling their behavior. Hybrid systems are challenging to identify because the parameters and equation structure may change across multiple dynamical regimes. Key examples include varying transmission rates in epidemiological problems and legged locomotion. Many current methods focus on inferring a model for the system and detecting switching points in a time-centric framework. One can reframe the problem by clustering in data-driven coordinates, such that similar dynamical behavior is close together, and then use the sparse identification of nonlinear dynamics (SINDy) method to identify different dynamical regimes. I will discuss model selection using SINDy and information criteria. I will then demonstrate the success of the method hybrid-SINDy on a spring-mass model and a simple infectious disease model with time-dependent transmission rates. I will also investigate robustness to noise and cluster-size.

Jun 2018
6
Wed 12:15
OPEN