Scribd
Upload
91 views

Lec 1 (2/6/02) Overview of Modeling Concepts

1. This document provides an overview of modeling concepts and introduces modeling and simulation as a way to understand complex systems. 2. It discusses why modeling and simulation should be taught at the undergraduate level and what the goals are, including gaining appreciation of modeling applications and hands-on simulation experience. 3. Several examples are given to illustrate modeling applications in materials science, including modeling the behavior of dislocations to understand material strength, modeling a shape charge using multiscale techniques, and modeling a small crack using a combination of quantum, atomistic, and continuum models.

Uploaded by

rebson
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
91 views

Lec 1 (2/6/02) Overview of Modeling Concepts

1. This document provides an overview of modeling concepts and introduces modeling and simulation as a way to understand complex systems. 2. It discusses why modeling and simulation should be taught at the undergraduate level and what the goals are, including gaining appreciation of modeling applications and hands-on simulation experience. 3. Several examples are given to illustrate modeling applications in materials science, including modeling the behavior of dislocations to understand material strength, modeling a shape charge using multiscale techniques, and modeling a small crack using a combination of quantum, atomistic, and continuum models.

Uploaded by

rebson
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Lec 1 (2/6/02) Overview of Modeling Concepts

We begin by considering several simple questions.

1. Why Modeling and Simulation ?


This is a fundamental and quantitative way to understand complex
systems/phenomena which is complementary to the traditional approaches of theory and
experiment. Modeling and Simulation (Mod/Sim) are concerned with powerful methods
of analysis designed to exploit high performance computing. This approach is becoming
increasingly widespread in basic research and advanced technological applications, cross
cutting the fields of physics, chemistry, mechanics, engineering, and biology.

2. Why teach Mod/Sim at the UG level ?


Mod/Sim studies are mostly being carried out at the graduate and postgraduate
levels. But there is no reason why the undergraduates cannot participate in a meaningful
way and benefit from the physical insights and technical know-how that it can provide.
We believe that engaging the undergrads broadly across the Institute through a team of
multidisciplinary faculty, as instructors and mentors, can succeed at MIT, providing the
students with a broader academic exposure than what they would normally encounter
within their own departments.

3. What are the goals of Mod/Sim ?


The students should gain a significant appreciation of the broad use of modeling
in several fields of science and engineering, acquire hands-on experience with simulation,
ranging from basic use of computers to advanced techniques, and develop communication
skills by working with practicing professionals. Additional benefits could come from
further interactions with the faculty afterwards, such as mentoring, UROPs, thesis
supervision, etc.
4. Subject is taught in two parts (see Lecture Schedule)
Part One: simulation methods lectures, HW
Part Two: term project, applications lectures/demos, presentation

In both parts students have an opportunity to learn concepts in thermodynamics,


statistical mechanics, quantum mechanics, solid and fluid mechanics, materials science,
condensed matter physics, chemical physics, etc. through a modeling and simulation
applications context.

In this, our first meeting, we will discuss briefly what we mean by modeling and
give several examples.

Model ≡ a simplified description of an actual system, useful for studying the system
behavior. Models are used in every branch of science; they are tools for dealing with
complex systems and the interactions of their constituent parts. Scientists and engineers
have long used models to better understand the systems they are studying, for analysis
and quantification, performance prediction and design.

Types of models -- a physical system can be described at different levels:


(a) a collection of nuclei and electrons (electronic-structure level)
(b) a collection of atoms and molecules (atomistic or molecular)
(c) a set of coupled structural elements (finite-element)
(d) a medium described by fields and distributions (continuum)

In terms of details, (a) has the most details (note we are not considering the
protons and neutrons that make up a nucleus, that would take us into the domain of
nuclear physics). At each level one can construct a model that can be used to analyze the
system properties and behavior.

When is a model useful? When a complex system can be simplified (reduced) to a


tractable level that makes clear the essential structure of system and how the internal
constituents interact with each other.

Key to a good model lies in what and how simplification is introduced; it is very
important to understand what aspects of the system the model is intended to describe, and
what are the model's limitations as a result of the simplification.

Suppose we are interested studying a Cu crystal at the atomistic level. This means
our model is a collection of N Cu atoms which interact with each other through a
potential energy U(r 1 , r2 ,...r N ) , where r 1 ,r 2 ,...r N denote the positions of atoms 1, 2, ...,
N respectively. At this level U is a given function. Knowing U and the system structure
we can use the model to calculate the various properties of Cu crystal. How? The
process that allows us to do this is simulation.

We continue our introductory discussion by showing a few brief examples of


materials problems that can be addressed by modeling and simulation. Fig. 1-1 shows a
list of topics compiled by the Materials Processing Center at MIT, an organization linking
the research activities at the Institute with industrial companies with an interest in
materials technology. This list shows the variety of problems that involve one form of
modeling research or another in producing materials all kinds of technologies. Notice
that atomistic modeling is at the top of the list - this is a technique we will come back to
discuss in some detail. While Fig. 1-1 may be impressive in its technological variety, it
may be easier to see what is materials modeling by thinking about only one kind of
materials property such as the strength of material.

Fig. 1-2 shows 3 high-speed photographs of a shape charge where a metal is


pushed through an orifice by an explosive, the molten metal then expands as in a plume
once it clears the orifice in about 12 microseconds. This process is of interest to the
scientists at the Lawrence Livermore National Laboratory in connection with its program
on (nuclear) stockpile stewardship. This is work for the Department of Energy to ensure
the safety and reliability of our existing nuclear weapons stockpile. What this figure
conveys is that a fundamental process of metal being deformed at very high pressure and
strain rate actually has important practical applications, in this case a matter of national
security and defense. The relevance of the shape charge to modeling of strength of
materials is indicated in Fig. 1-3, showing a sequence of models which takes the basic
understanding gained from modeling the behavior of dislocations (a line defect in
crystals) at the atomistic scale through first the behavior of a single crystal under stress
(on the micron scale, then the behavior of a polycrystal, a collection of single crystals (on
the mm scale), and finally to the actual shape charge specimen. This is an example of
multiscale modeling about which we will say more in the next lecture.

Fig. 1-3 shows another aspect of the strength of materials problem, that of stress-
strain behavior of a Cu crystal. The plot show the variation of stress with strain; this kind
of information is very useful in analyzing the strength and deformation of a material. If
one wants to understand the various features of plastic deformation, starting a strain of
0.01 (stage I) and extending to larger strains, then one has to consider how the underlying
defects, particulrly dislocations, are behaving. This is a current challenge to the research
community. On the right of Fig. 1-3 two TEM pictures show the patterns of dislocation
structures. It is not yet possible to explain these features fully from basic theory or
simulation.

We conclude this short discussion with Fig. 1-4 showing how one can study a
small crack, with dimension of a few nanometers, using a combination of three models,
TB refers to a quantum mechanical calculation which is very computationally intensive
and is therefore applied only at the most critical region (the crack tip), MD refers to
atomistic calculation which is applied to the region surrounding the crack tip, and finally
FE (FEM in the next lecture) is applied sufficiently far from the crack where the material
is not subjected to any stress concentration and therefore can be treated more
approximately.

Remainder of class discussion is a short talk by Professor Gerd Ceder on his


current research on modeling materials for batteries using first-principles (quantum
mechanical) calculations.
Fig. 1-1 MPC Modeling List

Fig. 1-2 Shape Charge (LLNL)


Fig. 1-3 Stress-Strain Curve (Cu)

Fig. 1-4 Crack Simulation Multiscale

You might also like