ONGOING PROJECTS
For many years
molecular level computer simulations have been used to provide an unambiguous
means to test theoretical assumptions and to observe and obtain a better
understanding of microscopic structure and transport mechanisms. However, with the rapid (and sustained) growth in computing
power along with improvements in empirical molecular force field models and
simulation techniques, we are very rapidly approaching a situation where
simulations can provide “experimental” data and resolve questions concerning
conflicting experimental data. It is fair to say that in the near future, molecular level
simulation results will be used to make conceptual design level calculations and
to some degree, eliminate the need for costly laboratory experimentation
. One of the stumbling blocks to achieving this goal has been
in the uncertainty in model force-field parameters that must be used in these
simulations and the lack of transferability of optimized force fields from one
type of simulation to another. This is
particularly prevalent when looking at the simultaneous simulation of
equilibrium and transport properties. Ideally, computational
quantum chemistry would provide the necessary force-fields, but in spite of the
incredible advances in computational power, this possibility is still restricted
to relatively simple molecular systems consisting of fewer than 10s of atoms.
Thus, in the near future, we will continue to use parameterized site-site
model force fields. In this project we are performing
a systematic investigation of force fields for simulation of equilibrium and
transport properties. Our focus is on the non-bonded interactions and we
investigating the effects of simultaneously changing both the repulsion and
dispersion parts of the potential (e.g., exp-6-8 or m-6-8).
Computer Simulation
Equation of State
Development
During
recent years we have made substantial advances in equation of state
determination, most of which have been directed towards pure fluids.
We use accurate equations of state as input to our mixture studies, and
have, therefore worked to continually develop ultra-high accuracy pure fluid
equations of state. Our current work in this area has three components:
1)
Use of selection algorithm methodology to develop an equation of state
that explicitly incorporates crossover in the critical region
In
the area of equation of state structure, a topic of great interest for us is to
develop an equation of state that incorporates crossover in its design.
Fluids of interest are those of high commercial impact, for example, N2,
C2H4, CO2, O2, CH4, C3H8, R134a and Ar amongst others. Our belief is that by
super-imposing rescaling or crossover theory on the selection algorithm, many,
if not most, of the terms resulting from the selection algorithms will
disappear. The reason for this is
that the main function of many of these terms is to cancel each other out to
make the classical equation of state appear nonclassical near the critical
point. This research incorporates
what we know about the critical point directly and uses the selection algorithm
to find the background terms. The
net result of this study should be more compact, higher accuracy equations of
state for fluids. We have also
applied this methodology to improve the performance of the SAFT and Patel-Teja
equation of state models
2)
Use of selection algorithm methodology to simultaneously optimize the
structure of a single equation of state that can be used to represent the first
ten alkanes
We
have been investigating a revision of the extended corresponding states
principle which at least partially eliminates the artificial shape factor
structure that is created near the critical point and during the past three
years we have developed several new high accuracy MBWR32 equations of state in
conjunction with this effort. These equations have been used along with previously
developed equations to create a Teja-like variable critical compressibility
reference fluid.The resulting correspondence at subcritical temperatures is
excellent with the shape factors being very close to one. As we move into the supercritical region, however, an
apparent density dependence arises which is much more complicated than would be
expected from the work of Lee and Kesler and Teja and coworkers. In our opinion,
this apparent density dependence in the supercritical region is due to
differences in data distribution and accuracy in the data sets that were used to
determine the MBWR-32 coefficients. In fact, the functional form of the MBWR-32
equation of state was developed for nitrogen and there is no reason to believe
that it is optimal (especially in the supercritical region) for other fluids.
Thus
far we have developed MBWR32 equations of state for all of the normal alkanes up
to C10 and for isobutane and isohexane. We
intend to develop a similar equation for neopentane and isooctane and then
extend our simulated annealing/stepwise regression algorithm to be able to
simultaneously optimize functional form of an equation of state using data sets
for all of the different substances. The algorithm will then result in a
functional form which will yield on average the best results for all of the
compounds included in the data bank and will hopefully eliminate some if not all
of the apparent density dependence observed for shape factors in the
supercritical region. The program
being developed is a relatively simple MPI based parallel algorithm to perform
the optimization on our SP2. An
extension of this study will be to explicitly include material constants such as
the acentric factor, dipole moment, etc., in the term bank, thereby giving an
optimized, generalized equation of state.
3)
Development of an equation of state which contains explicit
composition dependence.
A long standing goal of our equation of state work is to incorporate our theoretical results into the development of mixture equations of state. This work differs from previous conformal solution studies in that here the theory is used to define specific equation of state terms rather than to define a scaled temperature or density that can be used in a pure reference fluid equation for mixture calculations.
Pedagogy
CCHE: Beginning in July of 1997, the Colorado Commission for Higher Education (CCHE) started funding of a Program of Excellence Enhancement Award to the Chemical Engineering Department at the Colorado School of Mines. This enhancement project supports hardware and software acquisition and curriculum development to implement molecular modeling at the undergraduate chemical engineering level. Curriculum development work generally involves the development of new learning materials for inclusion in existing courses and the development of new courses that focus on the molecular aspects of chemical engineering. Another aspect of this project includes consultation with other experts in the field such as the CACHE Molecular Modeling Task Force, participation in meetings, short course and, for example, consortia for software development and collaborations with industrial groups that are already actively using these tools. The ultimate objective of this enhancement project is to develop an educational program for the chemical engineer of the 21st century.
Top
GECO The Golden Energy Computing Organization (GECO)
is a multidisciplinary/multi-institutional partnership consisting of Mines
faculty and students and researchers from the National Renewable Energy
Laboratory (NREL) and the National Center for Atmospheric Research (NCAR). This
consortium will promote activities crossing disciplinary lines and fostering
links between education, scientific inquiry and industrial pursuits. The
facility will bring scientists and engineers together to cover a broad spectrum
of energy-related research. CSM Departments and/or Divisions currently involved
with the program include the following:
Consistent with Mines' mission to foster curricula
and research geared toward responsible stewardship of the earth and its
resources, the GECO facility will enable faculty and students initially
to investigate eight primary challenge topics:
Students in the program are expected to become proficient in
computing technology, including numerical computation and the practical use
of advanced computer architectures, as well as in one or more applied
disciplines, such as solid mechanics, materials science, chemical
engineering, etc. Thesis research by students is expected to
be computationally oriented and actively advised by faculty members from
member departments.
CSM Campus Projects
Bioengineering and the Life Sciences (BELS) are becoming increasingly significant in fulfilling the role and mission of the Colorado School of Mines. Many intellectual frontiers within the fields of environment, energy, materials, and their associated fields of science and engineering, are being driven by advances in the biosciences and the application of engineering to living processes. The program in Bioengineering and Life Sciences (BELS) at the Colorado School of Mines is managed by the Chemical Engineering Department but is interdisciplinary in its course offerings. Participating departments/divisions include the Divisions of Engineering, Environmental Science and Engineering, and Liberal Arts and International Studies, and the Departments of Chemical Engineering, Chemistry and Geochemistry, Geology and Geological Engineering, Mathematical and Computer Sciences, Metallurgical and Materials Engineering, and Physics. The mission of the BELS program is to offer Minors and Areas of Special Interest (ASI) at the undergraduate level, and support areas of specialization at the graduate level, as well as to enable research opportunities for CSM students in bioengineering and the life sciences.
Top ...................................................................................................................................................................