Computational Rheology via LAMMPS
(a
two-day course) |
|
October 12 and 13, 2013 (Saturday and Sunday) |
Using Large Amplitude Oscillatory
Shear (LAOS) (a 1½-day course:
1 day of lecture and ½ day hands-on experience) |
|
October 12 and 13, 2013 (Saturday and Sunday) |
The short courses are held in conjunction with the
85th Annual Meeting of The
Society of Rheology (October 13- 17, 2013)
Return to Contents
Computational Rheology via LAMMPS |
|
Jeremy Lechman Reactive and Nanoscale Processes Department Sandia National Laboratories |
|
Matt
Lane Materials Science and Engineering department
Sandia National Laboratories |
|
Steve Plimpton
Computational Sciences Center Sandia National
Laboratories |
|
|
|
Using
Large Amplitude Oscillatory Shear (LAOS) |
|
Randy H. Ewoldt, Assistant Professor
Department of Mechanical Science and Engineering
University of Illinois at Urbana-Champaign
|
|
A. Jeffrey Giacomin, Professor
Department of
Chemical Engineering
Queen’s University |
Instructor Biosketches
Jeremy Lechman is a staff member in the Reactive and Nanoscale
Processes department at Sandia National Laboratories. His technical work
includes coarse-grained and meso-scale computational mechanics with
applications to multiphase, particulate, and composite materials.
His current interests are in developing macro-scale, continuum
governing equations and constitutive models for transport processes
in reactive heterogeneous materials.
Matt Lane
is a staff member in the Materials Science and Engineering
department at Sandia National Laboratories. He earned his doctorate in
Physics at the University of Texas at Austin, but was educated at
the University of Virginia. His computational research interests are
in atomistic study of nanoscale soft condensed matter and
far-from-equilibrium systems and his approaches are influenced
heavily by training in nonlinear dynamics.
Steve
Plimpton is a staff member in the computational sciences center
at Sandia. He is one of the principal developers of LAMMPS, as well
as other open-source scientific simulation codes which he supports.
His background is in solid-state physics and materials science.
Professor
Randy H. Ewoldt is Assistant Professor of Mechanical
Science and Engineering at the University of Illinois at
Urbana-Champaign. With a combination of experiment and theory, his
research group studies nonlinear rheology, non-Newtonian fluid
mechanics, mathematical modeling, and engineering design involving
soft materials and complex fluids. Prior to joining
Urbana-Champaign, he performed his doctoral research at MIT and
postdoctoral work at Minnesota.
Professor
A. Jeffrey Giacomin is President of The Society of
Rheology, Professor of Chemical Engineering at Queen’s University,
Kingston, Ontario, Canada.
He teaches polymer processing and rheology to senior undergraduate
and graduate students every semester, and his research focuses on
the role of rheology in plastics processing. Professor Giacomin is
an expert on the measurement of the nonlinear viscoelastic
properties of molten plastics, and on interpreting these
measurements.
Return to Contents
Computational Rheology
via LAMMPS /
Using
Large Amplitude Oscillatory Shear (LAOS)
Instructors: Jeremy Lechman, Matt Lane, and Steve Plimpton
Particle-based simulation has, in recent years, made great
strides toward closing the gaps in understanding the rheological
behavior of complex, multi-constituent and multiphase systems. Due
to advances in both computational infrastructure and techniques,
macroscopic rheological models can now be extracted from numerical
simulations of individual interacting components at ever-finer
length and time resolutions (e.g. grains, colloids , macromolecule,
and atoms). Obtaining and analyzing this detailed information is
crucial for prediction and designed control of the rheological
behavior of complex fluids in countless manufacturing/industrial,
biological and environmental processes.
In this course we will introduce participants to the relevant
methods and techniques for numerical simulation of complex fluids
for rheological applications using
LAMMPS, an
open-source molecular dynamics package. Mathematical and numerical
tools for gaining physical understanding will be emphasized.
Particle based simulations, both classical molecular dynamics and
coarse-grained methods, will be discussed. Hence, the participant
will receive an introduction to the basics of particle-based
numerical simulation using LAMMPS. This will include writing and
running LAMMPS input scripts and using the code to compute
rheological properties of both atomic-scale and coarse-grained
systems. Lectures on the technical basics will be interleaved with
hands-on tutorial sessions. Rheological applications will be
emphasized with a number of case studies and group discussions.
Short-course participants will have an opportunity to discuss their
particular research interests and applications.
Although some coding experience would be helpful, it is not
required to use the LAMMPS package. Some familiarity with molecular
modeling would also be helpful, though not required. Students should
come with personal laptop, running Windows, MacOS, or Linux. A
pre-built Windows executable version is available on the
LAMMPS download
site. Mac and Linux users should download and install LAMMPS
prior to the course. Detailed instructions are given at
http://lammps.sandia.gov in the 2nd chapter of the manual.
Course Outline
Session 1: LAMMPS overview |
|
- Molecular Dynamics basics
- Newton’s Equations of motion for many-particle
systems
- The pair potential
- Numerical integration – the Velocity Verlet
algorithm
- Advanced techniques
- How LAMMPS is structured
- Atoms
- Pairs
- Fixes
- Computes and Thermo
- Dumps and restarts
- Movies, images of published work done in LAMMPS
|
Session 2: Getting started with LAMMPS |
|
- Website and resources
- Downloading and installing/compiling
- LAMMPS out of the box – building a simple script
- Visualization of results – VMD?
- Other pre and post processing issues
- Examples from the LAMMPS distribution
|
Session 3: Hands-on applications lab I |
Session 4: Atomistic representations |
|
- Background/Foundations
- Strengths and weaknesses of atomistic study
- Representations and relevant potentials/theory
- Fully atomistic (Lennard-Jones, water,
polymer)
- Coarse-grained (bead-spring, implicit
solvent, united atom)
- Thermostats and ensembles – some statistical
mechanics background
- Creating Initial configurations
|
Session 5: Coarse-grain representations |
|
- Overview of coarse graining concepts and techniques
- Effective potentials for large particles
- DLVO models for colloidal particles
- van der Waals and Yukawa potentials
- Integrated Lennard-Jones model for colloids
- Numerically calculated complex interactions
- Granular potentials – noncolloidal particles
- Models based on Hertz theory for contact of
elastic solids
- Models based on JKR/DMT adhesive contact
theory for “sticky” particles
- Dissipative Particle Dynamics
- Coarse-grained solvent models
- Navier-Stokes (“Implicit Solvent”) methods
- Classical NS methods
- Stokesian Dynamics
- Lattice-Boltzmann, etc.
- Stochastic techniques
- Brownian Dynamics
- Dissipative Particle Dynamics
- Multi-particle Collision Dynamics
|
Session 6: Hands-on application lab II |
Session 7: Common calculations and
measurements |
|
- Simple thermodynamic quantities - pressure,
temperature, etc.
- Microstructure - radial distribution functions and
structure factors
- Diffusion, displacement, velocity distributions and
correlation functions
- Bulk viscosity (Green-Kubo)
- Issues and validation
- Non-equilibrium quantities – stress tensor
- Mueller-Plathe viscosity method
- Nonequilibrium MD viscosity method
- Issues and validation
|
Session 8: Post-processing, visualization
and data handling |
Session 9: Hands-on applications Lab III |
|
- Set up and run simple example problem to calculate
distributions, correlations, and equilibrium viscosity
- Set up and run simple example problem to calculate
shear viscosity
- Stress/viscosity for water in nanoconfinement
- Forces between nanoparticles in suspension
- Set up and run simple example problem to calculate
g(r), and MSD of suspension of hard-sphere colloids
- Set up and run simple example problem to calculate
shear viscosity of hard-sphere colloids
|
Session 10: Recent development and future
additions |
|
- Microrheological Approach
- Generalized Langevin Equation
- Memory kernel for colloids in
incompressible, Newtonian fluid
- Macrorheology of Non-Newtonian materials
- Peri-dynamics of visco-elastic solid
|
Course Schedule
Day 1 |
8:30 |
|
Introduction and course agenda |
8:45-10:15 |
|
Overview of Molecular/Particulate Dynamics via LAMMPS (Plimpton) |
10:15-10:30 |
|
Break |
10:30-12:00 |
|
Getting started with LAMMPS (Lane) |
12:00-1:30 |
|
Lunch Break |
1:30-3:00 |
|
Hands-on applications lab I (Plimpton, Lane, Lechman) |
3:00-3:45 |
|
Atomistic Applications (Lane) |
3:45-4:00 |
|
Break |
4:00-4:45 |
|
Coarse-grain Applications (Lechman) |
|
|
|
Day 2 |
8:30-9:15 |
|
Common equilibrium calculations (Lechman) |
9:15-10:15 |
|
Hands-on applications lab II (Plimpton, Lane, Lechman) |
10:15-10:30 |
|
Break |
10:30-11:15 |
|
Common non-equilibrium calculations (Lechman) |
11:15-12:15 |
|
Hands-on applications lab III (Plimpton, Lane,
Lechman) |
12:15-1:45 |
|
Lunch Break |
1:45-2:30 |
|
Getting under the hood – Making Changes to LAMMPS (Lane) |
2:30-3:15 |
|
Recent development and future additions (Plimpton) |
3:15-3:30 |
|
Break |
3:30-4:30 |
|
Hands-on applications lab IV, Discussion/Consultations (Plimpton,
Lane, Lechman) |
4:30-4:45 |
|
Wrap-up discussion and feedback |
|
|
|
Instructors: Prof. Randy H. Ewoldt and
Prof. A. Jeffrey Giacomin
A
popular use for commercial rheometers is the large amplitude
oscillatory shear (LAOS) experiment, or oscillatory strain sweep.
Rheologists have always used this to prepare for a linear
viscoelastic frequency sweep. Beyond linear viscoelasticity, this
nonlinear LAOS experiment generates rich data related to material
structure, processing, in-use conditions, and function; nonlinear
rheology is also a strong guide to the development of constitutive
models. This course addresses the unique challenges of
experimentally generating, analyzing, and interpreting data from
LAOS, and other large amplitude oscillatory techniques. Of the
various nonlinear deformation protocols, oscillatory techniques are
powerful in that they systematically survey the broad range of
deformation timescales and amplitudes (the Pipkin
space), and quantify both elastic and viscous components
simultaneously. Oscillatory techniques can also probe nonlinearity
more gradually than step inputs, which is valuable for probing shear
sensitive materials such as biological gels. This course will offer
a comprehensive introduction to those industrial researchers,
graduate students, and faculty that seek to probe microstructure,
develop constitutive models, or simply fingerprint the nonlinear
behavior of viscoelastic materials. Enrolled students will receive
software for analyzing raw oscillatory data, along with a playbook
for acquiring and analyzing data from LAOS.
Course Outline
- Rheological Material Functions Review
- Load input (compliance, fluidity)
- Deformation input (modulus, viscosity)
- Shear stress and normal stress difference responses
- Pipkin diagram
- Oscillatory deformation possibilities – not just LAOS
- Small Amplitude Oscillatory Shear (SAOS) Material
Functions
- Complex viscosities and complex moduli
- Complex normal stress difference coefficients
- Linear viscoelasticity and discrete spectra
- Predictions from complex material functions (Cox-Merz
relation, Laun relation, etc.)
- Large Amplitude Oscillatory Shear (LAOS) Material
Functions
- The standard Strain Sweep analysis (first-harmonic
moduli)
- Lissajous curves – the visualization playground
- Time domain analysis – Fourier-transform framework
- Deformation domain analysis – Chebyshev coefficients &
nonlinear moduli
- Relaxing the typical symmetry assumptions (even
harmonics, local moduli, sequence of physical processes,…)
- Rheometers
- Rectilinear flow
- Sliding plate
- Sandwich geometry
- Rotational
- Cone-plate
- Concentric cylinder
- Parallel plate
- Experimental Errors
- Total force versus shear stress
- Slip
- Edge failure
- Viscous heating
- Fluid inertia
- Start up
- Electrical interference
- Noise
- Signal Processing
- How to calculate G’ and G” ? (sine regression, Fourier,
and other options)
- Fourier transforms (DFT vs. FFT, underlying assumptions,
performance sensitivity)
- Noise removal
- Software distribution and tutorial
- Interpretation of Results
- Common LAOS signatures (simple analysis)
- Common LAOS signatures (advanced analysis)
- Identifying bad data (edge failure, shear-banding, …)
- Constitutive model fingerprints
- Examples of Using LAOS
- Probing microstructure
- Constitutive model development
- Nonlinear viscoelastic model parameters for POLYFLOW
- Design of viscous dampers
- Beyond LAOS: Examples of Oscillatory Deformation to Probe
Nonlinearities
- Superposed steady flow + oscillations (fluid
limit)
- Superposed stress or strain + oscillations (elastic
limit)
- LAO-Extension, LAO-Bending, LAO-Microrheology, LAO-Flow
- Other reciprocating material functions
Return to Contents
Short course registration includes a complete set of
course notes. Payment can be made on line with MasterCard, Visa,
Discover, or American Express.
Registration Fee for |
Through 9/6/2013 |
After 9/6/2013 |
Computational Rheology via LAMMPS (Two-day
Course) |
|
|
|
Member* |
$750 |
$850 |
|
Non-Member** (includes
membership for 2014) |
$810 |
$910 |
|
Student Member* |
$425 |
$525 |
|
Student Non-Member** (includes student membership for 2014) |
$455 |
$555 |
Using Large Amplitude Oscillatory Shear
(LAOS) (1½-day Course) |
|
|
|
Member* |
$625 |
$725 |
|
Non-Member** (includes
membership for 2014) |
$685 |
$785 |
|
Student Member* |
$350 |
$450 |
|
Student Non-Member** (includes student membership for 2014) |
$380 |
$480 |
|
|
|
|
|
* Member rates are available only to registrants who are members in
good standing as of June 30,
2013 or who are registered to attend the 85th Annual Meeting. |
|
** Non-members who are
registered to attend the 85th Annual Meeting may register
for the short course at the member rates. |
Cancellations for the short course received by
electronic mail (c/o The Local Arrangements Chair, Marie-Claude
Heuzey,
marie-claude.heuzey@polymtl.ca) by
September 6, 2013
will be refunded minus a $50 administrative charge. No refunds
will be granted after that date. Each class is limited to 40 students.
Return to Contents
Questions can be directed to Anne Mary Grillet,
Sandia National Laboratories, current chair of the SOR
Education Committee, at
amgrill@sandia.gov. |