Fall 2003 Departmental Distinguished Seminar Series
Paul Davidovits, Ph.D.
Professor, Chemistry Department, Merkert Chemistry Center, Boston College
Mass Accommodation of Gas Phase Species on Octanol as a Function of
Relative Humidity; the Strange Behavior of the Hydrogen Halides Gases
October 24, 2003
Davidovits began the seminar by pointing out that very little is known
about the physical and chemical properties of organic aerosols that have
been shown to be abundant in many regions of the troposphere. In the work
he discussed, 1-octanol was selected as a surrogate for the hydrophobic
oxygenated organic compounds found in such aerosols. Using a droplet train
apparatus, the uptake of several organic gas phase species and gas phase
HCl, HBr, and HI was measured to probe the nature of hydrophobic organic
surfaces as a function of relative humidity. These measurements yielded
the mass accommodation coefficient "a," which is the probability
that a gas phase molecule striking the liquid surface enters the bulk
liquid. The measure uptake of the organic gas phase species is in accord
with expectations. The uptake increases with decreasing temperature and
is independent of relative humidity. On the other hand, the observed uptake
of the gas phase acids is highly surprising. In the absence of water vapor,
"a" for both HBr and HI is unity; however, "a" for
HCl is much smaller on the order 0.015. The values change dramatically
as a function of relative humidity (i.e. the density of water vapor).
As the relative humidity increases, the "a" values for HBr and
HI decrease and "a" for HCl increases. At a relative humidity
of about 50 percent, "a" for all three species reaches values
measured earlier on pure water ("a" between 0.15 and 0.3, depending
on temperature). These results are discussed in terms of the nucleation
model for mass accommodation and a mechanism is proposed to explain the
surprising results related to the uptake of the hydrogen halides. Davidovits'
visit was hosted by Kalliat Valsaraj.
Jacob A. Moulijn, Ph.D.
Professor, Reactor and Catalysis Engineering Group, Department of Chemical
Technology, Delt Technical University, Delft, Netherlands
Structured Catalysts and Reactors, a Contribution to Process Intensification
October 31, 2003
Moulijn began by stating that the creation of a sustainable society is
a goal that gives inspiration to chemists and chemical engineers. He went
on to add that catalysis is the enabling technology for good chemicals
manufacturing methods. It is useful to realize that catalysis, as such,
is a crucial enabling technology, but a catalyst requires a reactor configuration
to perform its desired functions. So, catalytic reactors form the heart
of most chemicals production plants. Given the large variety of processes,
it is no surprise that a large variety of chemical reactors are used.
In the course of the seminar, he stated that the major, basic concerns
(apart from high activity, selectivity, etc.) for catalytic reactors are:
catalyst quality on a microscopic length scale (quality, number of active
sites); catalyst quality on a mesoscopic length scale (diffusion length,
loading, profiles); ease of catalyst separation and handling; heat supply
and removal; hydrodynamics (regimes, controllability, predictability);
transport resistances (rate and selectivity); safety and environmental
aspects (run-aways, hazardous materials, selectivity); and, costs. Why
would one use structured reactors? Moulijin's answer: Precision in catalytic
processes is the basis for clean processes. During the remainder of the
seminar, he proceeded to give his arguments as to why the catalyst and
the reactor should be close to perfect. He discussed the pros and cons
of random packed beds, as well as structured reactors. Moulijin's visit
was hosted by Douglas Harrison.
Jose A. Romagnoli, Ph.D.
Professor and Chair of Process Systems Engineering, Department of Chemical
Engineering, The University of Sydney
Process Systems Engineering at The University of Sydney
November 24, 2003
Romagnoli's presentation covered the different activities under investigation
at the Laboratory for Process Systems Engineering at the University of
Sydney. Focus was on recent developments in areas such as: intelligent/smart
process monitoring, modeling/optimization/control of complex systems (polymerization
reactors, crystallization processes), and integrated systems design/operation.
Also in his presentation, current developments in the laboratory's experimental
facilities were discussed, as well as their integration into his main
areas of research. Finally, some of the laboratory's industrial activities
were described. Romagnoli's visit was hosted by Danny Reible.
Richard Alkire, Ph.D.
Professor, Department of Chemical and Biomolecular Engineering, National
Computational Science Alliance, University of Illinois at Urbana-Champaign
Use of High Performance Computing Tools to Integrate Experimental Data
with Multi-scale Simulations of Copper Electrodeposition with Additives
February 20, 2004
Allkire's seminar focused on the idea that the sustained growth of information
technologies will enable a period of unprecedented advances in "molecular
engineering," as well as creation of next-generation research tools.
He went on to state that electrodeposition represents an ideal testbed
for developing such tools because, in part, it includes many systems that
are complex and have multiple phenomena that span as many as ten orders
of magnitude in time and length scales. In addition, electrodeposition
applications support a large, economically significant industry that is
characterized by rapidly changing technological opportunities. The field
is but one of many where it is critically important to speed innovation
from discovery to application. He focused here on the role that trace
quantities of solution additives play by influencing complex multi-scale,
multi-phenomena events during electrodeposition, especially product quality.
In this selective overview of science and engineering aspects, he highlighted
areas where numerical simulation tools were needed for predicting molecular
behavior that leads to self-assembly, as well as for design and control
of fabrication processes. The range of topics discussed included Atomic
Scale Simulations of early stages of nucleation and growth, coarse-grained
Meso-Scale Simulations that bridge the region between atomistic
and macroscopic (continuum) scales, Multi-scale Simulations that
link methods appropriate for different scales that are applied simultaneously
to achieve a comprehensive description of a system, and Parameter Assessment
for determining which parameters are the most sensitive, estimating their
values, and selecting from among multiple reasonable reaction mechanisms
when there is uncertainty in experimental data. Alkire's visit was hosted
by Elizabeth Podlaha-Murphy.
Jan Andersson, Ph.D.
Professor, Institut für Anorganische und Analytische Chemie, Westfälische
Wilhelms-Universität Münster, Münster, Germany
"Why does the sun make petroleum toxic?"
March 17, 2004
Andersson began his seminar by stating that large amounts of crude oil
are spilled into the oceans every year and that photochemical oxidation
is a major but not very well investigated route of removal of the components
of crude oil. It is known that toxic products are produced through the
photooxidation of crudes. He and his research team have studied the polycyclic
aromatic sulfur heterocycles to learn about the products that are formed
through this oxidation. They have found that a range of compounds belonging
to different chemical classes are produced from the benzothiophenes and
dibenzothiophenes studied. Andersson's visit was hosted by Judy Wornat.
James G. Goodwin, Jr., Ph.D.
Professor and Chair, Department of Chemical Engineering, Clemson University,
Clemson, South Carolina
Investigation of Reaction at the Site Level: Selective Oxidation of
CO on Pt-Based Catalysts
March 19, 2004
Most reaction techniques used in the study of heterogeneous catalysis
do not provide any information about reaction at the catalytic site level.
Isotopic transient kinetic analysis (also known as SSITKA) is a powerful
technique using isotopic switching during reaction that allows the in-situ
study of adsorption and catalysis on catalytic surfaces at reaction conditions.
Parameters able to be determined include the surface concentration of
reactive intermediates, a measure of the site turnover frequency, and
the distribution of site activities.
Goodwin focused on the use of SSITKA to investigate the selective oxidation
of CO. Poisoning of the electrode in a proton exchange membrane fuel cell
by CO impurities in the hydrogen fuel is a severe problem in using hydrogen
derived from hydrocarbons by reforming or partial oxidation. In order
to reduce the amount of CO in the hydrogen stream to a tolerable level
(<10 ppm) without losing too much hydrogen, a suitable catalyst must
be used downstream of hydrogen generation to selectively oxidize the CO
to CO2. Pt/Al2O3 and, especially, Fe-promoted
Pt/Al2O3 based catalysts have been found to be suitable
for this purpose. The seminar also focused on the use of SSITKA to explain
the cause of the rapid initial deactivation of these catalysts to steady-state
activity and to explain why Fe-promotion results in a more active catalyst.
Goodwin's visit was hosted by Jerry Spivey.
R. Bertrum Diemer, Jr., Ph.D.
Prinicipal Consultant, DuPont Engineering Technology
Advances in Population Balance Modeling
April 30, 2004
Many of the current advances in simulation tools are occurring as new
and efficient ways are found for combining capabilities from previously
stand-alone applications. For example, process simulators are being interfaced
with improved or specialized databases to expand simulation capabilities
beyond their classic applications in petrochemicals to include rigorous
simulation of aqueous and pyrometallurgical systems. Likewise, there is
a drive to expand the capability of CFD to simulate reacting flows and
multiphase flows, including situations in which particle formation and
growth processes are important.
The first part of Diemer's seminar was devoted to approaches aimed at
solving two significant problems encountered in moment methods: model
closure and distribution reconstruction. Several closure approaches were
discussed (MOMIC (method of moments with interpolative closure) and QMOM/DQMOM
(quadrature method of moments/direct quadrature method of moments) and
their performance compared to a classic discretized population balance
(DPB) model on a test problem. A method for developing efficient basis
sets for distribution reconstruction was presented and applied to the
problem of simultaneous aggregation and breakage. As part of this development,
a generalization of daughter distribution was discussed.
The second part of the seminar was turned toward bivariate problems.
This was introduced by a discussion of similarity solutions to the univariate
coagulation problem following which a general method of attack was presented
for bivariate problems aimed at both finding similarity solutions and
developing moment models. This was applied to the problems of multicomponent
coagulation, collision and coalescence of particles, and polymerization
with crosslinking. Diemer's visit was hosted by Ben McCoy.
^ Top
|
