Multiscale Engineering and Science Hub (MESHub) for Undergraduate Research and Education

skoptsytruculentAI and Robotics

Nov 15, 2013 (3 years and 9 months ago)

89 views

Multiscale Engineering and Scien
ce Hub (MESHub) for Undergraduate Research
and Education

Project Summary

We propose to
develop
a state
-
of
-
the
-
art

learning and research experience in
multiscale
engineering
and science

for undergraduate students, and
a multi
disciplinary systems
perspective for real world
applications
.

A multiuniversity

pilot

undergraduate course will be
developed
as

a collaborative effort between
ODU and RPI on the fundamentals and
applications of multiscale engineering

and science
.

C
ollabora
tive research project
s

(included as an integral part of the course) will be designed to
provide hands
-
on experience and
help

to
develop the team skills of the learners
.
T
o enhance the effectiveness of learning,
a
Multiscale Engineering and Science Hub (MES
Hub) will be developed
, with
complementary

features
to those
of the NSF
funded
NanoHub,

providing a personal and collaborative learning
and research environment. It will support pervasive learning in both the formal and informal
settings.

The hub will incl
ude: a) interactive, engaging learning modules
and tutorials
for introducing
diverse multiscale engineering and science applications,
as well as
modeling, and visual
simulations
strategies
to undergraduate students

that will draw on their background in
mat
hematics, science and engineering
; b) knowledge repository, with links to relevant digital
and multimedia information; c) comprehensive information retrieval, customization and
summarization tools; and d) a flexible wiki, to facilitate communication, colla
boration, and
knowledge generation by the learners.

Assessments will be made of the learning
and other
gains

resulting from the
pilot
course,
the MESHub and the collaborative project
s
.


Intellectual Merit

The learning and research experience provided by th
e
pilot
course, the hub and the
collaborative project
s
,

are expected to prepare the learners for the utilization of
the
multiscale
engineering principles and technologies to the
simulation and
design of complex engineering
systems
. Learners
completing

the
course
would be able to
further enhance their research
experience by participating in
either
the newly funded NSF program on multiscale engineering
at Washington State University
, or the
Multiscale Science
and
Engineering
C
enter at RPI.


Broader Impact

The

intellectual content of
the
multiscale engineering

learning modules
, along with the use of
emerging technologies and facilities in the MESHub
,

are expected to have
at least three positive
results
: (a)
h
elp attract
talented students with a
variety of backg
rounds and career interests to
the field of multiscale
engineering
and

science
;

(b)
e
nhance learning and concept retention
while sparking greater interest in multi
-
disciplinary engineering practices by providing stronger
visual cues and improved delivery m
odes
; and

(c)
b
enefit minority universities and community
colleges in particular

from visual

learning and other technologies incorporated into the hub.

1. Introduction


Many challenging problems associated with complex engineering and biological systems
e
xhibit phenomena occurring on a vast range of spatial and temporal length scales. Examples
include material synthesis and developments, modeling and control of fluid turbulence,
atmospheric modeling, design and fabrication of micro scale devices, image pro
cessing
,
bio
-
inspired sensor
s

and biomimetics,
and
a
multitude

of others. In recent years intense effort has
been devoted to the mathematical
foundation and computational strategies
of

multiscale
modeling and simulation

(Refs.1
-
4)
. This endeavor has prompt
ed the development of graduate
engineering
curriculum

as well as research centers on multiscale science and engineering.
T
he
multiscale

modeling science and engineering field
has matured to the level where

there
is a
n
urgent

need for

undergraduate educati
on and research training in this field.


1.1
Motivation for the Proposed Research


One of the
most

challenging aspects of any undergraduate curriculum is modeling and
simulation of various engineering problems represented by partial differential or discr
ete
equations. In an introductory continuum mechanics and thermal sciences courses, the concept
of boundary value problem, consisting of equilibrium, kinematical, and constitutive equations
together with essential and natural boundary conditions, is intro
duced. It is the compl
e
mentary
role
s

of physics and mathematics, i.e., the ability of mathematics to
quantify and model

complex
physical laws that is emphasized.

However,
t
he relation between material microstructure at
various scales and how it is related
to constitutive equations is mostly

of
f

limits


in most of the
engineering departments. In today’s curriculum,
these
relation
s

are
explained
(at best)

in terms
of
some equations

(
such as
those under
equilibrium

conditions)
,
that
directly follow from
physi
cal laws, but others, such as constitutive equations,
do

not follow directly from physical
laws
. While this so
-
called
phenomenological
approach is not pedagogically flawed, it has
significant shortcomings for complex history
-
dependent materials well into t
heir nonlinear
regime. For history
-
dependent materials
(guided by internal Markovian
processes
)
numerous
experiments
are required
to quantify material response
,

and even then
the approach
may fail if
the phemenological behavior is not well understood. In
practice, however, one defines a hand
-
full of constitutive law parameters that are believed to “capture” various failure mechanism
s

observed experimentally.

Furthermore, the microscopic behavior (both under equilibrium and
non
-
equilibrium stressed conditio
ns) is usually ignored, and neither is the connection between
such fundamental processes and the overall macroscopic responses clearly explained or
discussed.
Thus, t
here is a gap between the physics
-
based microscopic understanding and its
relevance to the

higher
-
level systems models.


To address some of the
aforementioned issues
, w
e propose to
develop
a pilot
undergraduate
course, and a multiscale engineering and science Hub (MESHub)
. The course is

based on a
physics
-
based

approach by which constitutive eq
uations (or directly field quantities)
are deduced from finer scale(s), at the
levels
where established laws of physics are better
understood

and can be directly applied. The phenomenological connection between different
time
-

and length
-
scales, and the re
levant boundary conditions will be emphasized.

The
enormous gains that can be accrued by this approach have been reported in numerous articles

(
Refs. 5
-
10
)
. Multiscale computations have been identified (see page 14 in
[
5
]
) as one of the
areas critical to f
uture nanotechnology advances. The FY2004 $3.7
-
billion
-
dollar National
Nanotechnology Bill (page 14 in
[
11
]
) states that: “approaches that integrate more than one
such technique (…molecular simulations, continuum
-
based models, etc.) will play an important
role in this effort.” Yet, the educational program emphasizing multiscale approach has not been
pursued so far.



The need for such a multiscale approach
for design of future complex engineering
systems
has recently been echoed at the highest level in majo
r U.S. industries. Michael
Idelchick, Vice President for Advanced Technologies at General Electric, stated that the “design
based on multiscale principles, would have a major impact on our productivity in water
purification and aerospace businesses. This a
ssessment is based on Rensselaer’s unique
depth and breadth of experience in basic science and engineering that have led to a history of
successful collaborations with GE.” John Kelly, senior Vice President of IBM remarked that the
“multiscale design prin
ciples in conjunction with our investment in CCNI (Rensselaer’s new
supercomputing center) will revolutionize engineering design in nanotechnology.”

2.
Proposed
Educational
and
Research

Program


2.1 Brief Overview


The
overall goal

of
our

proposed activ
ities is the
revision and enhancement
of
undergraduate science and engineering curriculum to ensure that the new vision of multiscale
modeling and simulation can be practiced at the bachelor’s degree level.
Initially, collaborative
research projects will b
e planned, involving students from both institutions working under the
direction of the principal investigators. Subsequently, w
e envision
the formation of a
cyber
enabled
learning and research community in Multiscale
Engineering
and
Science

(involving
sev
eral universities, research centers and industry)
,

and
a “hierarchical,” integrated, education
-
research approach that combines coaching
-
trained graduate students with team
-
experienced
undergraduates
.

Encouraging and facilitating interaction between the und
ergraduate and
graduate students is unique and would be fruitful in so far as: (a)
l
ending support, advice and
exposure to the undergraduates on more advanced concepts and engineering problems in a
collegial and student
-
friendly manner, and (b)
h
ighlightin
g opportunities at the graduate level,
including immersion in funded research projects that pay for education.


T
hree

other key aspects of integrating research into the undergraduate curriculum will be
pursued
, namely
:



1
.

A Multiscale Engineering and Sci
ence Hub (MESHub) will be developed to
facilitate
collaborative research
and
enhance the effectiveness of learning. The key components of the
hub are:



A flexible wiki, to facilitate communication, collaboration and knowledge generation by the
learners.



Int
eractive, engaging learning modules
and tutorials
for introducing diverse multiscale
engineering and science applications, modeling and visual simulation strategies to the
undergraduate engineering students.



Knowledge repository, with links to the diverse
relevant digital and multimedia information.



Comprehensive information retrieval, customization and summarization tools

The hub will incorporate some of the
tools in
the NSF supported NanoHub
, but will have
other complementary features to those in that hub

(e.g., information retrieval).


2.

A
ll undergraduate
course

project experiences will require modern computational
methods, bringing state
-
of
-
the
-
art
modeling and visual simulation

tools into undergraduate
education.
Initially

the tools
that are
taught in
undergraduate
engineering analysis courses

will
be used
.
In
the
future
,
additional
multiscale modeling and simulation modules supporting the
course

projects will be gradually incorporated into the core curriculum.


3.

A

workshop will be held at one of th
e two partner universities, which will allow the
undergraduate students to make formal presentation of their research. Project reports will also
be posted on the MESHub.


W
hile our overall view of “research experience for undergraduates” is embedded
throug
hout our educational vision, we will also apply for REU grants for summer projects. As an
extension of the undergraduate course team projects, several undergraduate students at the
two partner university will be selected at the beginning of their junior ye
ar.
In future years, t
he
students will be paired with the graduate students and post docs.
At ODU, the students will be
working at the Center for Advanced Engineering Environments
,

under the direction of Profs.
Kandil, Joshi, Nguyen (Co
-
PIs), and Prof. Noo
r (PI).
At Rensselaer they will be closely working
with graduate students involved in the Rensselaer Multiscale Science and Engineering Center
(MSEC) directed by Prof. Fish (proposal co
-
PI). In the summer, the students will be integrated
into REU activiti
es.


Also, l
earners
completing

the
pilot
course (and the project) will have the opportunity to
enhance their research experience by participating in either the newly
-
funded NSF program on
Multiscale Engineering at Washington State University

(
see
Suppleme
ntary Documentation
)
,
or the Multiscale Science and Engineering center at RPI.


Further integration
of
m
ultiscale engineering
will be made
into the

senior design course
required by ABET
. One of the ABET requirements is a Senior Design course for the Engine
ering
curriculum. Various departments satisfy this through a 2
-
semester offering (e.g., the ECE 485
-
486 sequence). This design course also forms the basis for many of our motivated students who
get accepted in the BS
-
MS program to pursue more intensive and

open
-
ended research tasks
and design problems. The BS
-
MS program at ODU was initiated to encourage high achieving
students to begin considering graduate school early on. This has proved to be very successful
and popular. In the present
context, the propos
ed
m
ultiscale
engineering

and science
course
would feed into such Senior Design courses and offer the pathway to a more comprehensive,
practically relevant and multi
-
disciplinary engineering education and research.


2.2 Major Objectives of the Project



The major objectives of the
proposed

one
-
year exploratory project are:


1) Plan and develop

(a)

A

pilot
undergraduate course on multiscale engineering and science
.

(b)

A MESHub, with
complementary
features
to those
of the NSF funded NanoHub

(in addition, to so
me of the current tools which are useful to the project).

(c)

Course projects to train the learners in developing the team and research skills
.
Th
e

project details and tasks will be inter
-
woven with the available tools and
technologies

at both ODU and RPI
.


2
) Assess the effectiveness of the course, MESHub and project
s

with regard to:

(a)

Enhancing the effectiveness of learning

multiscale engineering and science
fundamentals
, as well as student interest and participation

(b)

Developing the research skills
, multi
-
disc
iplinary concepts and “know
-
how”.

(c)

Attracting talented students from a wider variety of backgrounds and career
interests to the field of multiscale engineering and science
. We will also ass
ess if
this
experience
leads to an increase in our BS
-
MS program.



2.3 Overview of Research Activities


The primary goal of the proposed research and education program is to provide students
with the necessary multiscale systems
s
kills that would allow them to meet the future challenges
of engineering design, such for

instance called for
by

the Vice Presidents of GE and IBM. In
order to understand these challenges consider a multiscale design system developed at
Rensselaer as shown in Fig.
1
. It consists of the following modules and technologies:

a.
Mathematical upsca
ling
: derivation of coarse
-
scale
equations from fine
-
scale equations.
Mathematical
upscaling

are methods which take advantage of scale
separation and self
-
similarity to decompose a complex
multiscale problem into a sequence of much simpler
problems. Among
the mathematical upscaling methods
that will be considered are the
mathematical
homogenization

and its various variants.

b
.
Computational upscaling:

reducing the complexity
of solving a fine
-
scale problem to a manageable size that
can be adapted based on a
vailable computational
resources and error estimates in the quantities of
interest
, without sacrificing
on
overall accuracy
. Among
the mathematical upscaling methods that will be
considered are
various model reduction methods
.

c.
Model calibration:

solving

an inverse problem for
constitutive parameters (interfaces, fibers/tows, matrix)
F
F
i
i
n
n
e
e


s
s
c
c
a
a
l
l
e
e


m
m
o
o
d
d
e
e
l
l


A
A
B
B
A
A
Q
Q
U
U
S
S


C
C
o
o
a
a
r
r
s
s
e
e


s
s
c
c
a
a
l
l
e
e


m
m
o
o
d
d
e
e
l
l


A
A
B
B
A
A
Q
Q
U
U
S
S


Meso
-
scale

Model

I
I
n
n
v
v
e
e
r
r
s
s
e
e


S
S
o
o
l
l
u
u
t
t
i
i
o
o
n
n


O
O
p
p
t
t
i
i
m
m
i
i
z
z
a
a
t
t
i
i
o
o
n
n


P
P
r
r
o
o
g
g
r
r
a
a
m
m


Experimental Data
Depository

Engine Components

Struct. Components

Computational
upscaling

Model
Calibration

Prediction/
Design

Fig.
1
:
Multisc
ale design components


Mathematical

Upscaling

by minimizing the error between experimental data at coupon and fine
-
scale (nanoindentation
tests) levels hosted in the Experimental Depository.


One of the critical a
spects of multiscale approach is quantifying uncertainties introduced by
finer scales.
As a guiding principle for assessing the need for finer scales, it is appropriate to
recall the statement made by Einstein, who stated that “the model used should be the

simplest
one possible, but not simpler.” We will emphasize that the use of multiscale approach has to be
carefully weighted on case
-
by
-
case basis. For example, in case of metal matrix composites
(MMC) with almost periodic arrangement of fibers, introducin
g finer scales might be
advantageous since the bulk material typically does not follow normality rules and developing a
phenomenological coarse scale constitutive model might be challenging at best. The behaviour
of each

phase is well understood and obtain
ing the overall response of the material from its fine
scale constituents can be obtained using homogenization. On the other hand, in brittle
ceramics composites (CMC), the microcracks are often randomly distributed and
characterization of their interface

properties is difficult. In this case, the use of multiscale
approach may not be desirable
, or might need the integration of stochastic tools such as Monte
Carlo with deterministic continuum mechanics.



The quality of model predictions depends on the f
idelity of the model, the data used for
calibration, and the numerical and statistical methods used for simulations. To quantify the
uncertainty in model predictions, students will be trained in statistics.

A suitable MATH pre
-
requisite
, or the development

of applied mathematics courses in the Engineering College are
some options. For example, the Electrical and Computer Engineering department will develop
and offer an Engineering Mathematics course (ECE 200) this Fall 08 geared towards enhancing
and supple
menting the mathematical skills, background and knowledge of undergraduate
s
t
udents.


2.4
Facilities Used in the Proposed Research


a) ODU


The ODU


Center for Advanced Engineering Environments (CAEE) has acquired
several facilities that can be ad
apted to this project. These include:



Comprehensive Information Retrieval, Customization and Summarization



Multimedia and game engine software for generating learning modules


In addition t
he

ODU
extensive computational and experimental
facilities, equipme
nt and
resources are available to the PI’s for
use in the project
.


b) RPI


At Rensselaer there are several facilities that would be utilized for this project including



Multiscale
Science and

Engineering
Center
(MSEC) at Rensselaer, which consists of
more than
60 faculty from nine departments and over 100 graduate
students,

will
engage undergraduate students in ongoing multiscale research activities. MSEC will
leverage its
funding
from
numerous
government

agencies including
DOD, DOE, NSF
and more than
20 companies from industry.



The f
a
cilities at the RPI supercomputing center, which operates a 32,768
-
processor, 90
-
teraFlop Blue Gene supercomputer and a 10
-
teraFlop AMD Opteron cluster, will be
available to the undergraduate students involved in the proje
ct.



The International Journal of Multiscale Computation Engineering published by Begell
House (Prof. Fish is the Editor
-
in
-
Chief



Ref. 12
) agreed to publish the educational
materials and tutorials developed as a part of this project.

The Tutorials

will b
e
posted
on the MESHub and made
available to all the undergraduate students involved in
the project.
























a) Proposed MESHub






b) Future Extension


Figure 2


Proposed MESHub and Future Extension

3.
Proposed Ta
sks


The proposed tasks for the project can be divided into four major categories, namely:

(a)

Development of
a pilot undergraduate course on multiscale engineering and
science. The content of the course will include
engaging interactive learning
modules
(desc
ribed in section 3.
3
) and tutorials
on
m
ultiscale
e
ngineering and
s
cience
.

(b)

Development of the MESHub

(c)

Designing collaborative course projects for ODU / RPI students

(d)

Assessment of the effectiveness of the learning modules, MESHub and course
projects


The tas
ks are described subsequently

(See Figure 2)
.


3.1
The Hub


The hub will integrate the facilities listed in
S
ection 2.1 into a flexible facility and a
workspace in which learners can explore, understand and utilize the information. Specifically,



The lear
ning modules
, which constitute part of the curriculum
content,

will be posted on
the hub, after being presented at live lectures (using distance learning facilities at both
ODU and RPI)
.
Also, the tutorials generated for the special issue of the Journal wi
ll be
posted on the Hub.

The learning modules and tutorials will
provide online learning tools

for the students
.



Knowledge repository, with links to diverse relevant digital and multimedia information on
various aspects of
m
ultiscale
e
ngineering and
s
cienc
e.



Comprehensive information retrieval, customization and summarization tools. An
intelligent question / answering system will be incorporated to provide answers in an
intuitive manner.



A flexible wiki to facilitate communication, collaboration and knowled
ge generation by the
learners. The wiki will also be made available to all individuals and organizations
interested in
m
ultiscale
e
ngineering and
s
cience.


3.2
Pilot Undergraduate Course


The course will provide an overview and introduction to multiscale

Engineering and
Science (MES), at the undergraduate level, including: definitions and goals of MES; limitations
of conventional modeling and simulation strategies in coping with events and phenomena
occurring at disparate spatial and temporal scales; exam
ples of
possible
changes in the
principal physics governing the events with scale; various application areas in engineering and
science requiring multiscale modeling and visual simulations; brief description of some
multiscale linking / modeling approaches
; coupling of different time integration operators;
applications to microwave electronic circuits and nerve networks.


Some of the course material will be provided in the form of interactive, engaging
multimedia learning modules (described subsequently); t
he remaining material will
include both
the
PowerPoint
presentations

and the
tutorials in the special issue of the International Journal of
Multiscale Computation Engineering. All the course material will be posted on the MESHub.

The
course will be deliver
ed simultaneously at ODU and RPI using the distance learning facilities at
both universities.


3.3 Interactive Multimedia Learning Modules


A summary of
some of
the learning modules

and their

multiscale and multidisciplinary
aspects
drawn from different
e
ngineering
and science
fields is given
subsequently
. Some of
the
interactive multimedia tools along with the simulation methodologies and tasks required fo
r

the
learners are also
described
. These descriptions only serve as a sampling. In
the future
, perio
dic
updates, refinements, and revisions to the objectives and design criteria and tasks will be
implemented as necessary and indicated through course assessments, evaluations, and
student feedback
.


A.
Overview of
the
Multiscale Engineering and Science
(MES):


This introductory module will set the stage for subsequent four modules and will include:
definitions and goals of MES; various application areas in engineering and science requiring
multiscale modeling and visual simulations; brief introduction of

various approaches in
multiscale modeling and simulation including
length and time scales in hierarchical modeling;
key components, enabling technologies and tools for multiscale modeling and visual
simulations; verification, validation and model calibrat
ion; synergistic coupling of multiscale
simulations and physical experiments; role and potential of multiscale modeling and
visual
simulations in industrial design


current barriers, challenges and pacing items.

B.
Information
-
Passing and Concurrent Appr
oaches


This l
earning module

will
provide an overview of
two types of multiscale technologies in
space and time, namely: information
-
passing and concurrent

approaches (Ref. 6).
In the
concurrent multiscale methods both, the fine and co
a
rse scales are simu
ltaneously resolved,
whereas in the information
-
passing schemes, the fine scale is modeled and its gross response
is infused
into the
coarse

scale
. The modules will systematically address the major issues of
multiscale science and engineering including:

o

W
hat is the information that needs to be transferred from one scale to another?

o

What are the correct ways to achieve such transfer of information?

o

What physical principles
and conservation laws
must be satisfied during the transfer of
information or simulat
ion results?

The t
able below describes
the
various
topics that will be
briefly covered

in the
learning module

o
n Information Passing and Concurrent Approaches.

Type/Domain

Spatial Multiscale Methods

Temporal Multiscale Methods

Information
Passing

Asymptot
ic Multiscale method

o

Mathematical homogenization for linear
problems in 1D

o

Mathematical Homogenization for linear
problems in 3D

o

Homogenization Error Estimation

o

Material Design

o

Boundary Layers

o

Nonlinear Homogenization

Temporal Homogenization

Generalized
Mathematical
Homogenization

Constrained dynamics

Equation
-
Free Method

Langevin framework

Kinetic Monte
-
Carlo

Concurrent

Domain decomposition based methods

Current practices (submodeling, global
-
local)

Overlapping domains: Alternating Schwarz

Disjoint dom
ains: Lagrange multipliers

Discrete
-
to
-
continuum DD methods

Superposition based methods (coexisting
domains)

Multilevel based methods

Introduction to multigrid methods

Multigrid
-
like multiscale methods

Space
-
time multilevel
methods

Multiple Timestep (MTS)
Methods

Subcycling Methods

Dual purpose

Multiscale Enrichment based on Partition of
Unity

Quasicontinuum



C.
Coupling of Different Time Integration Operators


For certain kind of problems, such as the analysis of fluid
-
structure systems, the total
eleme
nt

assemblage of stiffness and mass characteristics are quite different in different parts of
the assemblage (such as the fluid is very flexible as compared to the stiffness of the structure).
Thus, the explicit time integration of the fluid response using

the conditionally stable central
difference method, in conjunction with an implicit unconditional stable time integration (such as
Newmark) method for the structure response seem to be a natural choice. The critical time step
size for an explicit time int
egration of the fluid response is usually much
smaller

than the one for
the structural response. It may also be efficient to use different time step sizes for the explicit
and implicit integrations, with one (time) step size being a multiple of the other.
Finite element
(FE) domain decomposition (DD) has been considered as amongst the most effective
procedures for solving large
-
scale engineering/science problems
(
Refs. 1
3
,

1
4)
, and FE
-
DD
procedures have also recently been used for solving multiscale problem
s
(
Ref. 1
5)
. Thus, in this
learning module, students will be briefly introduced (at the elementary level) to various technical
topics, such as solving PDE by
FE


and Finite Volume methods
, explicit
-
implicit time integration
algorithms, DD step
-
by
-
step proc
edures, and large
-
scale parallel
-
sparse equation solvers,
which can be utilized to solve multiscale problems.


D
. Electro
-
thermal Analysis of Microwave Electronic Circuits:


Semiconductor based microwave integrated circuits find wide applications that
include
radar technology, pulsed
-
power systems, satellite and cellular communications networks. With
emphasis on device miniaturization, higher frequency operation and “close
-
packing” for ever
-
higher “chip densities”, the issue of electro
-
thermal analysis
(Ref.
16
)
becomes important. This
coupled with short wavelength operation at the high
microwave

frequencies, effectively reduces
the spatial scale within the devices, and requires distributed modeling to account for phase
-
effects that can be coherent. The
strongly disparate time scales for electronic phenomena as
compared to thermal propagation and heat transfer, calls for appropriate decomposition of the
temporal scales as well. This is thus a good example
for

a
pplying the multi
-
scale principles to
enginee
ring design and
analysis
. Simulation analysis (transient and steady
-
state) can be carried
out with different types of devices, applied signals waveforms, and different boundary conditions
(both electrical and thermal). Available software tools (such as SPI
CE

and
MEDICI
) and f
ast
-
solvers
,

such as
SUPER_LU based on decomposition of sparse
matrices
,

will also be applied
as necessary.


E
. Electrically Induced Action Potential Generation in Nerve Networks:


Interest in biomedical engineering is ever
-
increasin
g, especially given that many
processes are electrically controlled (e.g., voltage gating of ion
-
channels, action
-
potential
propagation in neural fibers, arrhythmia in the heart etc.)
.
Furthermore, there is a strong promise
of applying electrical pulses to

biological system
s

for many novel applications that include: (a)
Electrically triggered signaling based on neurotransmitter release and secondary calcium
messenger events

(Ref.
17
)
, (b) Irregular nerve impulse stabilization, (c) Cellular electroporation
a
nd directed killing of tumor cells via apoptosis

(Ref.
18
)
,
and
(d) Gene therapy
th
rough
“electropermeabilization” of cell membranes.


Here we propose a demonstrative module for the study and analysis of electrical
signaling in nerves based on a distribute
d Hodgkin
-
Huxley type transmission line model

(Ref.
19
)
. This will be coupled with tissue
-
level macroscale electrical responses

and
include statistical
uncertainty in ion
-
gating and voltage
-
dependent conductance modulation. This will enable
students to app
ly their knowledge of probability and stochastic processes (as already covered in
ECE 304) to a practical, inter
-
disciplinary area.

Fig.
3:

(A) Biological
membrane, (B) nature
-
inspired membrane



3.
4

Collaborative Project
s


Each team of students will select a project, as a practical application of the knowledge
dev
eloped in the course.
In future, f
aculty and graduate students from different
engineering
departments

with expertise in these applications will be working closely with the undergraduate
students
.
Each
project will cover
an aspect
of
one of
the following re
search areas (but
is
not
limited to them):

a.)
Nature
-
Inspired Membranes

Maintaining access to pure, fresh water is one of the greatest
challenges of the 21
st

Century. Water desalination is a process of great
technological relevance, which is carried out

with remarkable efficiency
in cells. Biological membranes are extremely selective while
maintaining enormous flows through the channels that connect the
inside and the outside of the cell or the nucleus

(Fig. 3)
. Multiscale
simulation is the quintessentia
l tool to unravel the flow and separation
mechanisms. Multiscale methods would enable the access of such
long time
-
scales, since finely resolved simulations, with time steps on
the order of fs (10
-
15

s), are necessary to simulate the channel while
gating p
henomena cannot be observed below 10
-
6

s. However, the MD
techniques reveal details of the internal structure that facilitates and
produces the barriers and their selectivity (e.g., K
-
channel in
membranes).

b.)
Nanocomposites
It is estimated that advanced
-
composite autobodies could be about 65
percent lighter than today's steel versions. Thus, for mass

reduction and the
best possible fuel economy
, advanced
composites offer the greatest potential for ultralight vehicles. The
system behavior is strongly inf
luenced by the interface
properties. This interfacial region is believed to be responsible for
unique properties of nanocomposite materials. These fillers
(spherical or plate
-
like nanoparticles or carbon nanotubes) are so
small that their interaction with
polymer chains can not be
described by continuum mechanics principles. At the nanoscale,
the interaction of the polymer
-
chains and the nanofillers needs
will be explicitly represented by constructing atomistic unit cell
(AUI), which includes few nanopartic
les interacting with polymer
chains (Fig.

4)
.


3.
5

Assessments and Evaluation


Initial e
valuation of our progress will consist of
student surveys

and the extent of use of
the Hub.
Consistent evaluation of student project work will be done using a standa
rdized rubric.

The assessment will be made by an ODU faculty with expertise in cognitive science.
In
the
future, assessment will include
documenting the number of projects created, the number of
faculty/graduate student mentors, the number of undergraduate

students engaged in the
projects, evaluation of students’ design project work
, and enrollment interest in the
integrated
BS
-
MS programs within the Engineering College.

In addition, we will collect student
demographic information to
assess the

impact
of

di
versity

(gender, major, ethnicity, etc.).
Finally, we will track retention rates of students who participate in our education activities. We
Atomistic

Coating

Bundle

Weave

Composite
tube

Fig.
4
: Nanocomposites coating

2
0
0
n
m

will conduct focus group interviews of students to identify how we may refine our educational
activities to better
meet diverse students’ needs.


Also, since most engineering departments maintain records
and information on their
alumni,

we also envision
tracking the progress of our graduates and getting their feedback from
an industry viewpoint in the future.


3.
6

De
liverables and Timeline


The proposed activities will be coordinated with other related activities (to avoid
duplication, and whenever possible and appropriate leverage other resources). Among the
related NSF projects are the newly
-
funded activities at Was
hington State University, and the
NanoHub project.



The work
on the project
will start one month after the project is funded. The timeline for
the expected completion of the tasks is shown subsequently.

Deliverables and Timeline

First

Second

Third

Fourth

Quarter

Quarter

Quarter

Quarter

a)

Development of Learning
Modules







Pilot
Course Offering







b)

Development of MESHub

First Release









Updates







c)

Designing Collaborative
Projects













Performance of Projects








Workshop











d)

Assessment and Testing










4. Intellectual Merit


The intellectual merit of the proposed project is to provide an attractive, relevant and
connected learning and research experience in Multiscale Engineering and Science for
u
ndergraduate students, and a multidisciplinary systems perspective for real world problems.
The excitement and intellectual content of the learning modules, along with the use of emerging
digital and web technologies of the hub will help in attracting high
ly talented students to the field
of multiscale engineering and science.

5.
Broader Impact and Educational Initiatives


5.1
Expected Significance


Our long
-
term goals include
developing an effective technology
-
based, cyber
enabled
,

collaborative learning

and research environment for multiscale engineering and
science that would accelerate the development of the needed technologies and their
industrial applications.

The current project is a step towards the development of an
undergraduate curriculum that w
ill be part of the environment.

It will also reap the benefits of
multi
-
university collaborations and provide opportunity for students to
share the resources
at

multiple
sites.



We plan to broaden awareness of the results of this project through:



Providin
g links in the MESHub to all the relevant activities at various academic and
research institutions

(including NSF funded NanoHub at P
u
rdue University, and the
new program at Washington State University)
, and encouraging these institutions to
cross link our

MES
Hub
.

Minority universities will be contacted and provided
access to the MESHub.



Organizing a workshop at the end of the project
, posting the projects on the Hub,

and
inviting several institutions to participate and help in assessing our activity. Minor
ity
universities and community colleges will be invited to the workshop.


5.2
Impact


The
pilot
undergraduate course,
the
MESHub and
the multiscale
course project
s

will
prepare the students for
successful
careers
as practitioner dealing with complex engi
neering
system
,
and
as research
ers

and educat
ors

in th
e

field. Th
e

project
will
provide the

undergraduate
s

with
a perspective on

the complexity and multidisciplinary nature of practical
engineering problems. It
may

help

the

students
to be

better prepared
t
o deal with complex
systems
and
provide them with

a competitive advantage in the global workforce.
With t
he
infusion of fresh (multiscale) ideas into undergraduate curriculum, the project may reinvigorate
engineering education that would hopefully
attract
students with
diverse

backgrounds and
career interests, particularly women, and underrepresented minorities.


5.3
Integration of Research and Education


a) Old Dominion University


Old Dominion University has a history of activities targeting advanced le
arning
technologies. In particular
,

the Center for Advanced Engineering Environments led a consortium
of eight universities in 2002
-
2003, with NASA funding, for studying
the feasibility of developing a
hierarchical research and learning network
. The partic
ipating universities included MIT Media
Lab, George Mason University, University of Illinois

at

Urban
-
Champaign, University of Florida


Gainesville, Cornell University and Syracuse University.


b)
Rensselaer Polytechnic Institute


Rensselaer has a history

of developing web
-
accessible educational materials that
demonstrate the methods and practices including interactive educational materials and
technologies for learning/studio
-
based curricula and outreach programs and (ii) tutorial and
simulation units tha
t provide the background necessary to understand and visualize the
technical issues. Educational materials will be based on the successful, multiple award
-
winning
interactive multimedia formats and technologies used by RPI in former MATC and CCLI projects
(Premier Courseware Award for 2000 from the National Engineering Education Delivery System
and 2001 Curriculum Innovation Award from the ASME.)

6.
References

1
. Grigorios A. Pavliotis, Andrew M. Stuart;
Multiscale Methods: Averaging and Homogenization
(Tex
ts in Applied Mathematics),
Springer; 2008.

2.
Sabine Attinger, Petros Koumoutsakos (Editors);
Multiscale Modelling and Simulation
(Lecture Notes in Computational Science and Engineering),

Springer; 2004.

3.
Ahmed K. Noor,
Multiscale Modeling, Simulation a
nd Visualization and Their Potential for
Future Aerospace Systems
,
NASA/CP
-
2002
-
211741 2002
-
07, NASA Langley Research Center,
2002.

4.
Timothy J. Barth, Tony Chan, and Robert Haimes (Editors);
Multiscale and Multiresolution
Methods: Theory and Applications

(Lecture Notes in Computational Science and Engineering),
Springer; 2001.

5
. Curtin, W.A. and R.E. Miller, Atomistic/continuum coupling in computational materials
science, Modeling and Simulation in Materials Science and Engineering. 11 (3) (2003) R33
-
R68
.

6
. Fish, J., Bridging the scales in nano engineering and science, Journal of Nanoparticle
Research. 8 (2006) 577
-
594.

7
. Fish, J., ed. Bridging the Scales in Science and Engineering. Oxford University Press, 2007.

8
. Ghoniem, N.M. and K. Cho, The emergin
g role of multiscale modeling in nano
-

and micro
-
mechanics of materials, Modeling in Engineering and Sciences. 3 (2) (2002) 147
-
173.

9
. Liu, W.K., E.G. Karpov, and et. al., An introduction to computational nanomechanics and
materials, Computer Methods in A
pplied Mechanics and Engineering. 193 (2004) 1529
-
1578.

10
. Khare R, Mielke SL, Paci JT, Zhang SL, Ballarini R, Schatz GC, Belytschko T, Coupled
quantum mechanical/molecular mechanical modeling of the fracture of defective carbon
nanotubes and graphene she
ets, Physical Review B. 75 (7) (2007) Art. No. 075412.

11
. National Nanotechnology Initiative. Supplement to the President’s FY 2004 Budget. National
Science and Technology Council Committee on Technology.

12.
Jacob Fish (Editor),
International Journal for

Multiscale Computational Engineering,
Volume
4, Issue 3; Begell House Inc., 2006.

1
3
. C. Farhat, F.X. Roux, Implicit Parallel Processing in Structural Mechanics, Computational
Mechanics Advances, 2 1
-
124 (1994)

1
4
. D.T. Nguyen, Finite Element Methods: Par
allel
-
Sparse Statics and Eigen
-
Solutions,
Springer Publisher (2006)

1
5
. T. Arbogast, G. Pencheva, M.F. Wheeler, and I. Yotov, A Multiscale Mortar Mixed Finite
Element Method, Multiscale Modeling & Simulation, A SIAM Interdisciplinary Journal, Vol.6,
No.1,
pp. 319
-
346 (January through May 2007).

16
.

G. Zhao, R.P. Joshi, V.K. Lakdawala, and H. Hjalmarson, “Electro
-
Thermal Simulation
Studies for Pulse Induced Energy Absorption in Nan0
-
Crystalline ZnO Varistors”, IEEE Trans.
Dielectr.
a
nd

Electr. Insulation 14,

1007
-
1015, 2007.

1
7
. R.P. Joshi, A. Nguyen, V. Sridhara, Q.
Hu, R. Nuccitelli, and K.H. Schoenbach, “Simulations
of Intra
-
Cellular Calcium Release Dynamics in Response to a High
-
Intensity, Ultra
-
Short Electric
Pulse”, Phys. Rev. E 75, 041920/1
-
10, 2007.

1
8
. A.G. Pakhomov, R. Shevin, J.A. White, J.F. Kolb, O.N. Pakhomova, R.P. Joshi and K.H.
Schoenbach, “Membrane Permeabilization and Cell Damage by Ultrashort Electric Field
Shocks”, Archives of Biochemistry and Biophysics, 465, 109
-
118, 2007.

1
9
. R.P. Joshi
, A. Mishra, J. Song, A. Pakhomova, and K.
H. Schoenbach, “Simulation Studies
of Ultra
-
Short, High
-
Intensity Electric Pulse Induced Action Potential Block in Whole
-
Animal
Nerves”, IEEE Trans. Biomedical Engineering 55, 1391
-
1398, 2008.