Wallace Anderson, Sc.D.
Professor of Electrical Engineering
Cullen College of Engineering, UH
The general principle being explored in this research is data extraction directly from two-dimensionally recorded information using optical analog methods. The term "data extraction" in this context can equally be understood as "data compression," "data reduction," "pattern recognition," etc.; hence, this study relates directly to a great variety of image-handling tasks, ranging from video telecommunication and HDTV to feature extraction from still pictures.
The application being used as a vehicle for this investigation (and one of great importance in its own right) is more specialized: it is the problem of estimating 3-D vector velocity fields in turbulent fluids. The starting point for data extraction in this case is in the manner of processing reconstructed diffraction fields from double-exposure "3-D snapshots," i.e., short-exposure in-line holograms, of small (5-100m) neutral buoyancy particles suspended in moving fluids. The conventional approach is direct focusing of reconstructed images onto a CCD camera and digitization followed by intensive computer processing. This procedure is extremely slow and is severely hampered by poor signal/noise ratio. Its resolving power for particle location varies from useless for depth resolution to adequate for lateral resolution.
Our approach has several advantages over conventional practices. We operate in the Fourier plane, which gives us the flexibility to utilize as much as we need of the wide-angle scattering essential for good resolution of particle locations, and which gives vastly improved signal/noise ratio. This operation is incorporated in an optical mask or "filter" that provides simultaneous, i.e., parallel, readout of all particle locations in the plane being viewed. (Mathematically the procedure is exactly that of "wavelet transformation," a subject of keen interest usually in connection with one-dimensional data analysis during the past decade or so and is, therefore, fundamentally a scaled cross-correlation.) The only burden on the workstation assigned is to record, for each frame, locations of correlation peaks higher than a threshold value set within the camera, so that each frame can be processed very rapidly.
We have found three different optical configurations that can achieve the scaling required for depth search together with the necessary correlation operation. One of these is described in a paper: "2-D wavelet transform and application to holographic particle velocimetry," by W. L. Anderson and Hongyan Diao, that has been accepted for publication in Applied Optics. Investigation of different possibilities for design of the optical mask is underway. It appears that the greatest promise lies in some type of POF (phase-only filter).
A number of intricacies are involved in the design of such a filter. Because its fabrication is necessarily by electron beam lithography, it is intrinsically quite expensive; so we are seeking at least a tentative compromise between performance and cost before making up a budget for consideration by an outside agency. For instance, we may have to strike a balance between high selectivity (hence, high resolution) and the need to accommodate slight variations in particle diameter. Also, the refractive indices of commonly used particles will surely have a small imaginary component; this would especially affect the small-angle portions of the scattering patterns. Since this factor depends on surface quality as well as bulk properties, it may not be available from the manufacturer and may be difficult to determine experimentally.
Back to Table of Contents
Farouk G. Attia, Ph.D.
Associate Professor of Mechanical Technology
Department of Civil, Mechanical, and Technical Mathematics Technology
College of Technology, UH
Existing Proportional-Integral-Derivative (PID) robotic controllers usually rely on an inverse kinematic model to convert user-defined cartesian trajectory coordinates to joint variables. These joints experience friction, stiction and gear backlash effects. Due to lack of proper linearization of these effects, modern control theory based on state space methods cannot provide adequate control for robotic systems. Fuzzy logic is a fast emerging alternative to conventional control systems in situations where it may not be feasible to formulate a proper model of the control system. This technique tracks user-defined trajectory without having the host computer to explicitly solve the nonlinear inverse kinematic equations. A telerobot system has a great potential for servicing large and unmanned space applications because of its flexibility and maneuverability in space. This project describes a control strategy for robotic manipulators that incorporate proportional plus integral (PI) controller with a simple fuzzy logic controller. The effectiveness of this controller is demonstrated through simulations involving the control of a two-arm, two-link manipulator.
Robots and telerobots for advanced space applications will require manipulators with redundant degrees of freedom, that are capable of employing a variety of geometric configurations to avoid obstacles. Conventional methods for control of manipulators, based on solving the inverse kinematics cannot be easily extended to these situations. Fuzzy logic-based controllers (FLC) offer simpler and faster solutions to satisfy these needs such as, increased robustness in spite of high ambient noise levels or sensor failures, ability to handle nonlinearities without control system degradation, and easy formulation of fuzzy rules.
A new control strategy for robotic manipulators that incorporates a proportional plus integral controller with a simple fuzzy logic controller has been proposed. The effectiveness of the proposed strategy has been demonstrated using a two-joint, two-link manipulator. It can be shown that only a maximum of 15 rules are required to evaluate individual joint axis motion and that a linear relationship exists between the number of rules and the degree of freedom of the robot. The fuzzy logic controller approach is 33 percent faster than the traditional kinematic equation. More studies of the proposed control strategy such as, trajectory tracking and higher degree-of-freedom, are desirable in order to demonstrate its potential for real-time control of manipulators.
Back to Table of Contents
Vemur Balakotaiah, Ph.D.
Professor of Chemical Engineering
Cullen College of Engineering, UH
Abraham Dukler, Ph.D. (deceased)
Professor of Chemical Engineering
Cullen College of Engineering, UH
L.-Q. Yu
Doctoral Candidate in Chemical Engineering
Department of Chemical Engineering, UH
Figure 1. A typical experimentally measured time trace of film thickness at Reynolds number 760, developed by Dr. Balakotaiah and his research team.
Two-phase gas-liquid flows are important in the design of many components of chemical and process equipment. Some examples include the flow of oil and gas in petroleum production facilities, flow of vapor and liquid in reboilers and condensers, and the simultaneous flow of gas and liquid in gas-liquid reactors, trickle bed contactors and wetted wall absorbers. In addition to these flows at normal gravity conditions, it has been suggested that gas-liquid (or vapor-liquid) two phase flows are important in a variety of future space operations such as the operation of the thermal bus for the space station, power cycles, and storage and transfer of cryogenic fluids in space.
For conditions of technological interest, there are three major types of flow regimes observed for gas-liquid flows in vertical pipes. At low gas flow rates, a "bubbly flow" pattern, in which, small gas bubbles are uniformly distributed in the liquid, is obtained. Increasing the gas flow rate leads to "slug flow." This flow pattern is characterized by large bullet-shaped gas bubbles separated by liquid slugs. Increasing the gas flow rate further leads to the annular flow regime in which the liquid moves along the pipe wall in a thin, wavy film and the gas flows in the core region. A large fraction of the gas-liquid flows of engineering interest falls in the annular flow regime.
Experiments have shown that in the annular flow regime the waves appearing on the liquid film have a profound influence on the transfer of heat, mass, and momentum in these systems. Neglecting the wavy nature of the film can seriously underestimate transfer rates. For example, the heat and mass transfer rates into the liquid may be enhanced by 100 to 300 percent in the presence of the wave, as compared to the flat film case.
In the simplest form of the annular flow regime, the liquid film falls under gravity through either a stagnant or an upward moving gas stream (causing interfacial shear). Figure 1 shows a typical experimentally measured film thickness profile for a freely falling film at a high Reynolds number (corresponding to a typical liquid flow rate of practical interest).
The liquid film surface consists of a series of large waves accompanied by an array of wavelets of smaller amplitude and length which exist between these large waves. The large waves are steeper in the front and slope more gradually in the back with slopes not exceeding ten degrees. The amplitude of these large waves can be two to three times the mean film thickness, and they carry a large fraction of the total mass flow. The wave amplitude and spacing display a chaotic behavior similar to that observed in many lower dimensional deterministic models. Researchers speculate that the large waves contain stagnation points and recirculating regions which enhance the transport rates.
We have successfully predicted the statistical properties of the waves (celerity, r.m.s values of film thickness, peak-to-substrate ratio, film thickness power spectrum, etc.) in a free falling film by solving the two-dimensional Navier-Stokes equations and using the methods of nonlinear dynamics. Current research is aimed at exploring the effects of interfacial stress on the wave profiles and wave statistics and using the hydrodynamic model to predict the heat and mass transfer enhancement in annular flows.
Back to Table of Contents
Edwin Carrasquillo-Molina, Ph.D.
Associate Professor of Chemistry
College of Natural Sciences and Mathematics, UH
Research during the award period focused on the continued development and implementation of a new approach in our laboratory, collision-assisted spectroscopy (CAS), to probe and characterize the rovibrational structure of the HCN molecule at high excitation. These studies, first of all, probed the structure of the first excited electronic state of HCN and showed that this state is not very well characterized. They revealed previously undetected perturbations in some upper state vibrational levels and earlier misassignment of the rotational structure in at least one state for this well studied molecule. From these investigations, a marked improvement in the rovibrational modeling of the upper electronic state has been derived (to be published). Characterization of the upper state also allowed the laser induced fluorescence (LIF) spectra from high lying vibrational levels in the ground electronic state to be fully unravelled. The experiments showed that CAS is more sensitive to CN stretch excitation and to the vibrational fine structure in the ground electronic state than previous methods. Detailed characterization of the ground electronic state vibrational structure from 4,000 12,000 cm-l has now been accomplished (to be published). This research has significantly improved the sensitivity and state-selectivity of the LIF detection capabilities for highly vibrationally excited HCN and now permits investigation of the CN + H2 exothermic reaction.
Back to Table of Contents
G. Ron Chen, Ph.D.
Associate Professor of Electrical and Computer Engineering
Cullen College of Engineering, UH
This project seeks to develop some new robust fuzzy control theory and techniques for adaptive control of space robots working in an uncertain environment where sudden changes of conditions and/or unpredictable faults may occur during the control processes. Our approach consists of a few key components in the design of the tentative control system:
Under support from this ISSO grant, we have completed the three tasks listed above. This grant of a total of $7,453 for 1994 has been used to support 1/2 months of summer salary for the PI and 3 months of summer salary for two graduate students.
The results obtained from Parts (1)(2) constitute the M.S. thesis of the PI's student Mr. Weiming Tang, who graduated in August of 1994 on schedule.
Back to Table of Contents
David R. Criswell, Ph.D.
Director of the Institute of Space Systems Operations
Office of the Senior Vice President, Academic Affairs / Provost, UH
Research was conducted and published on the implications of industrial development of the moon and future international exploration of the moon. Recent research focuses on describing the biosphere of Earth in terms generally applied to a manned spacecraft on a long-duration mission. Research summarizes key scales of Earth's biosphere and people and their needs. It reviews the growth of industry on Earth, the opportunities to achieve transition to solar energy from space, and the economic benefits of solar electric power from space. Arguments find that lunar industries will enable space solar electric power during the first part of the 21st century. Lunar industries conceivably lay the foundation for the migration of humans into space habitats. The needs, time duration, resources, knowledge, corpus of educated people, and growth path now exist to move mankind from the precarious exterior of spaceship Earth to more interesting engineered venues off Earth. The next fifty years can envision the continued stifling poverty of billions of people or the creation of a wealthy, healthy future for mankind.
Back to Table of Contents
David R. Criswell, Ph.D.
Director of the Institute for Space Systems Operations (ISSO)
Office of the Senior Vice-President, Academic Affairs / Provost, UH
ISSO supports selected aspects of UH's participation in the Texas Space Grant Consortium. Dr. Criswell is a member of the TSGC Board of Directors, an advisor on proposals submitted to TSGC, and a participant in proposal development and other community activities.
Award
David Criswell earned an award for chairing the National Space Grant Directors' Workshop
at the National Academy of Sciences Woods Hole Conference Center and directing the writing
and production of the report Mission to America, presented at the Third National Space
Grant Conference, NASA Johnson Space Center, January 11, 1993.
Back to Table of Contents
William E. Fitzgibbon, Ph.D.
Professor of Mathematics
College of Natural Sciences and Mathematics, UH
In 1994, W. Fitzgibbon received an ISSO Grant of $2,500 for travel to plan and organize a conference and workshop on high speed flow fields. The results of the workshop will be archived in a permanent research database located at the University of Houston.
The conference is planned to attract engineers and scientists involved in high speed aircraft and spacecraft design, researchers in computational and experimental fluid mechanics, applied mathematicians, physicists and those involved in code validation for scientific and industrial design. The conference will consist of invited plenary lectures delivered by internationally recognized experts. It is intended to be a forum bringing together international experts in modelization, experimentation, and simulation. Computational and experimental lectures on topics previously described have been suggested. A portion of the lectures be devoted to presentations concerning the following topics:
In addition to invited lectures, the conference will have contributed poster sessions, a panel discussion, and several informal social activities designed to encourage and stimulate the scientific interaction of conference participants.
The workshop is probably the most complicated component of the proposed project. The workshop will consist of presentations of solutions to test case problems designed to compare available experimental data with computation simulation. The goals are to assess the validity of physical models and to determine the accuracy, efficiency and robustness of the numerical simulation.
Workshop test cases will be selected to explore a variety of critical fluid dynamic and flow chemistry issues. They include:
Test case problems will be restricted to simple two- and three- dimensional geometric configurations.
Two panels of experts, one at research installations in the United States and the other at Western European research installations, are currently working to define two lists of approximately ten to twelve test case problems. It is expected that there will be considerable intersection of these two lists, enough, in fact, to extract on the order of ten final workshop test problems. Two additional test problems will come from Russian scientists and one from scientists in Japan.
Each of the finalized test problems will have two designated test case co-chairs, one from the United States and one from Europe. The test case co-chairs will preside over the workshop sessions. They will have the overall responsibility of completely defining and specifying their test problem and determining what is an acceptable problem solution. All test case co-chairs, as well as certain other individuals, will comprise a global Working Group. The Working Group co-chairs will coordinate the test case chairs each on his own continent. They will channel the flow of information between the conference/workshop organizers and place individuals interested in workshop participation in contact with the appropriate test case co-chair.
Workshop presentations will take place in the special sessions that involve real time visualization of solutions and comparison of experimental and computational solutions. Graphics workstations will be used to project the visualizations onto large screens. Visualization software will be provided by the European Hypersonic Database located at INRIA-Sophia Antipolis, in France. The organizers anticipate obtaining the necessary visualization hardware on loan from commercial vendors. Participants in the workshop will develop solutions to the test case problems which will be transmitted electronically to the University of Houston and stored in a database. During these sessions, these solutions will be drawn from the database for visual presentation. The test co-chairs will have the responsibility of compiling the proceedings of their sessions.
The online database will be maintained for later computational and experimental validation as well as for code calibration. It will be well documented and will be easily accessible to the research and industrial communities. The documentation will include written reports, book, and an on-line browser. The database will be stored on a devoted disk attached to an Internet accessible node. Backup tapes will be maintained by the University of Houston. The INRIA database software will be delivered to the University by the end of summer or early fall 1994.
Information stored on the European Hypersonic Database will also be stored on the database located in Houston. It will be transmitted electronically to Houston in the fall of 1994. The results of the proposed workshop will be included into the Houston database. In addition, the Houston database will continue to accept new solutions to test case problems well after the workshop has concluded. The suitability of new solutions will be determined by a committee extracted from the test case Working Group. Criteria for acceptance of new solutions will include accuracy, robustness or, perhaps, prior unseen phenomena. Thus in a sense the database will function as a continuing reference bank and electronic journal. The contents and new additions to the Houston database will be permanently maintained at the University of Houston and access to the database will only be available through the University of Houston.
ISSO funds have enabled researchers to attend conferences and committee meetings to promote university research. In January of 1994, Dr. Fitzgibbon traveled to Reno, Nevada, to present the workshop to the annual AIAA (American Institute of Aeronautics and Astronautics) meeting. Dr. W. E. Fitzgibbon and Dr. J. Periaux, Assistant Director of Scientific Strategy and Head of the Numerical Analysis Group, Dassault Aviation, France appeared before the Fluid Dynamics Technical Committee to secure the official AIAA endorsement of the proposed activities. As a result of their appearance, the official endorsement and an advertisement appeared in the August, 1994 issue of the monthly magazine of the AIAA, Aerospace America. Late summer of 1994, Dr. Fitzgibbon and Dr. Periaux traveled to Washington, D. C. to discuss potential funding from agency officials. Dr. Fitzgibbon also attended the annual IMACS (International Mathematical and Computing Sciences) meeting where he organized and conducted a special session on reaction diffusion systems and discussed the proposed event with many individuals including I. Stakgold of the University of Delaware, past president of Society for Industrial and Applied Mathematics (SIAM) and current lobbyist for the mathematical sciences in Washington, D.C., and W.F. Ames, chair of the IMACS conference.
Back to Table of Contents
George E. Fox, Ph.D.
Professor of Biochemical and Biophysical Sciences
College of Natural Sciences and Mathematics, UH
In support of long duration space missions, NASA is extremely interested in developing portable microbiological instrumentation. This reflects established needs to monitor the general level of microbial activity in the space craft environment and to achieve the specific identification of possible pathogens. In addition, regenerative life support systems may depend on specific species of bacteria. It has been previously shown that ribosomal RNAs (rRNAs) and their corresponding genes (rDNAs) make excellent monitoring targets for individual organisms. Work conducted during this period had the purpose of determining the feasibility of beginning the development of automated instrumentation for use in the space craft environment. The essence of the idea is to simultaneously immobilize a number of rRNA specific hybridization probes to a silicon surface which can then be exposed to a single biological sample obtained from space craft air or water. A detection system would then inform a computer system which probes tested positive and thus allow determination of the organisms present in the test sample. We anticipate that the strength of the hybridization signal will also be informative about the relative size of various populations. We showed during the past year that probes of length 30 could be designed that readily distinguish multiple species with redundancy. RNA samples were also prepared for preliminary tests of the relative utility of alternative probe target locations in the 16S rRNA.
Back to Table of Contents
Theodore F. Leibfried, Jr., Ph.D.
Associate Professor, Division of Computer Science, Engineering, and Mathematics
School of Natural and Applied Sciences, UHCL
Alfredo J. Perez-Davila, Ph.D.
Associate Professor
School of Natural and Applied Sciences, UHCL
Jung-Chang Huang, Ph.D.
Professor and Chair of Computer Sciences
College of Natural Sciences and Mathematics, UH
Sadegh Davari, Ph.D.
Associate Professor of Computer Science
School of Natural and Applied Sciences, UHCL
This is the final report on "Constructing a Runtime Environment for Real-Time Processes," a project partially funded by ISSO, which supported two students half time for three summer months. We proposed to investigate the following important practical problems to enhance rate-monotonic scheduling in real-time systems:
Since the proposal was funded at a level less than 40 percent of the proposed budgets, we stated in our letter of acceptance that we would concentrate upon establishing an RMS environment and then attempt the others only to the extent possible. In fact, the college support of $3800 proferred by the School of Natural and Applied Sciences (NAS) had to be rescinded because of a budget cut Nevertheless, we can report the achievement of some progress on the other items.
Bare Machine Environment
The bare-machine environment was a major part of the effort promised in accepting the
award at less than half of the funding requested. The initial work was related to a former
NIO project entitled "RMS Operational Testing Environment." Under this project,
an RMS schedule testing environment was created using the LYNX Operating System. The main
driver for the preceding simulation program, RMS_SIM,ADA has been divided into two
separate programs. These two programs are NEWTEST.ADA and NEWSIM.ADA. The first determines
the schedulability of a set of tasks. The second, which utilizes the task set analyzed by
NEWTEST.ADA, is a simulation program that runs separately from the analysis program and is
amenable to running on a bare-machine environment.
This work was accomplished to make the test and simulation environment directly usable by NASA since the operational environment for the Space Station Freedom has been changed from one which will operate in the LynxOS environment on an Intel386DX in the so-called Standard Data processor (SDP) box, to one which is a "bare machine," Ada runtime environment on an Intel386SX in the so called "super" MDM (Multiplexer deMultiplexer).
In addition to the work described, several technical discussions of work to be accomplished were held. In one of the early meetings, a basic project plan was discussed which included the project, "Enhancements to Rate Monotonic Scheduling." The original intent to address Ada-9X issues had to be modified since the Ada-9X compilers available did not support tasks essential to an investigation where concurrency is paramount. Discussion held at that meeting entailed a review of the five principal techniques for handling non-periodic or asynchronous events, viz. (1) polling, (2) priority exchange, (3) deferrable server, (4) sporadic server, and (5) the slackstealing algorithm. The two most promising avenues of research are with the sporadic server and the slack stealing algorithm.
Additional meetings discussed the concept of utilizing period transformation as a mechanism of increasing the priority, and, hence, the importance of low rate tasks in a system. What this does is to split the required budget for a given task into two or more equal subintervals, where each subinterval is structured to be run in a period correspondingly shorter than the original period. Since the RMS paradigm assigns priority to tasks according to their periods, the priority assigned to such a task would be correspondingly higher. This enables a hard-deadline but slow periodic task to attain a priority higher than a companion soft-deadline task which happens to operate at a faster periodic rate. In later meetings, the discussion of handling blocking time and deadlock, where low and high priority tasks share serially reusable resources, led to the decision to try to employ the more robust priority ceiling protocol (PCP), rather than priority inheritance protocol (PIP). PIP is simpler but more subject to what is called unbounded priority inversions and deadlock. The limited funding enabled us only to establish the basic bare machine enviroment and prepare some test programs which, as yet, have not been run because of delays in the installation of hardware. The minutes of these successive meetings are available upon request.
Frame Overruns
The next phase of our research involved identifying methods of preventing periodic and
periodic processes from executing for more than their maximum allowed time (frame
overruns). Two UNIX functions that help enforce the timing requirements were identified.
These findings were discarded in favor of the VERDIX VADScross crosscom-piler environment
for the bare machine (80386/80486).
The VADS cross compiler environment has implemented certain features that facilitate realtime programming. It contains a delay_until statement which has great advantages over the basic delay statement. It has Interrupt handling and semaphores In addition, it allows priorities of tasks to be changed during execution (dynamic priorities) and allows the user to control the scheduling policy. For example, it allows the user to set/disable time slicing and enable/disable preemption.
The most promising feature is the install_callout procedure. This procedure allows the user to install a function to be called at program or task events. Some of the events are EXIT_EVENT, PROGRAM_SWITCH _EVENT, TASK CREATE_EVENT, TASK_SWITCH_EVENT, and TASK _COMPLETE_EVENT. These services are of most interest because they can be used for time stamping as well as for the enforcement of timing requirements.
The VADS cross compiler has been installed on the DIAMOND server for the SUN Laboratory. The system has been configured, and a basic RMS scheduler has been coded . This code has been successfully compiled, but it cannot be executed because of the want of a serial line from the DIAMOND server to the allocated 80386 machine.
Future Work
As soon as the required hardware becomes available, the RMS scheduler will be executed and
debugged. After this, an aperiodic server will be added to the system. The next stage will
include eliminating frame overruns, A serious attempt will be made to complete the above
by the end of the fall semester 1994. This limited work for the fall is now being
supported by a NASA-NIO joint grant to UHCL and TAMU but not ISSO.
Significant effort was made to stay abreast of all the events taking place in Washington, D. C. on an initiative known as ''Industrial Computing." This initiative is headed by the Advanced Technology Program (ATP) of the National Institute of Standards and Technology (NIST) branch of the U. S. Department of Commerce. In this regard, with funding from RICIS, we attended a two-day brainstorming workshop in Palo Alto, California, on August 4-5. The workshop provided the program director with ideas from United States industries for the creation of a new focused group under the temporary title of "Dependable and Renewable Industrial Systems." Five focused areas currently exist. The collected ideas will be presented by the program director on October, 1994. Workshop participants approved announcements for request-for-proposals, which will appear sometime in November. This initiative is to help United States industries prove competitive in the world market by encouraging them to perform research, which, though risky, has potential for profitability. The initiative provides half of the cost.
Since any proposal submitted to NIST must be initiated by one or more industrial organizations, we have been talking to the realtime-systems people of CAELINK for submission of future joint proposals. We have had three brainstorming sessions so far. Realtime activities of the two organizations are complementary to each other, and the possibility of developing a joint proposal looks promising, Efforts are also underway to develop proposals for submission to the NSF, NASA, or DARPA.
Communications
Proposals and papers have been prepared and presentations delivered related to
"RealTime Process Run-Time Environments" and "Enhancements to Rate
Monotonic Schedul-ing." These companion proposals reflect continuing effort by the
UHCL realtime systems research teams. The UHCL team has also investigated realtime issues
not covered by the two proposals, for example, the slack-stealing algorithm comparison to
the rate-monotonic scheduling algorithm.
Back to Table of Contents
John H. Miller, Jr., Ph.D.
Professor of Physics and the Texas Center for Superconductivity at UH (TCSUH)
College of Natural Sciences and Mathematics, UH
The goal of this research project is to develop a noninvasive, noncontact magnetocardiography system for the diagnosis of cardiac arrhythmias. The initial phase of this program (Tralshawala, et al.) has achieved a three-channel system, with three superconducting magnetic sensors, a cryostat, and the design of a gantry for holding the cryostat. The three high-Tc SQUID magnetometers have been successfully fabricated and tested. Each SQUID sensor consists of an integrated YBCO SQUID and pickup loop on a SrTiO3 bicrystal substrate, as illustrated in Fig. 1.
Figure 1. Configuration of
a YBCO integrated singleturn magnetometer, based on a design by Miklich and his
colleagues. (Miklich et al.) The pickup loop connects to the SQUID body on either side of
the slit, just below the two junctions. The four vertical lines extending from the SQUID
are contact pads.
The YBCO films were deposited by laser ablation and were photolithographically patterned. The sensitivity of these sensors is 150 fT or, better, down to 5 Hz, which is comparable to the world's best high-Tc SQUID sensors. In addition, a three-channel SQUID controller system, including flux locked loop electronics and an IEEE-488 interface, has been purchased from Conductus, Inc. The liquid nitrogen dewar and gantry are scheduled for completion in October (Fan and Miller). Novel image surface gradiometers, designed to exclude external magnetic field noise, have been designed and mathematically modeled. The results of these modeling studies will be reported at the Applied Superconductivity Conference, October 17-21, 1994, in Boston.1 We will seek ISSO funding in 1994-95 to fabricate high-Tc superconducting cylinders and other shapes to be used, together with the SQUID sensors, for the modified image surface gradiometers that we have designed and modeled. (Miklich et al.). A finite element source localization code has been written for two dimensions and tested on two-dimensional cross-sections of cylindrical shapes and magnetic resonance images of the brain. Some of this work will be reported at the 1994 Applied Superconductivity Conference.2 Design of a novel nonmagnetic bed with precise x-y translational capability has been initiated. One of the novel features of the bed is the of pressurized air to levitate one stage above another while it is being translated in the x- or the y- direction. This bed is expected to be completed in late October. Animal and clinical studies are scheduled to be carried out in 1995.
References 1N. Q. Fan. and J. H. Miller, Jr. "A Scalar Field Solution to the Magnetic Inverse Problem for Biomagnetic Sowse Localization." Applied Superconductivity Conf., Boston, MA, Oct. 17-21 1994. 2N. Tralshawala, et al. "High-Tc Supercon-ducting Image Surface Magnetometers," Applied Superconductivity Conf., Boston, MA, Oct. 17-21 1994. A. H. Miklich, et al. IEEE Trans. on Applied Superconductivity 3 (1993): 2434.
Back to Table of Contents
James T. Richardson, Ph.D.
Professor and Chair of Chemical Engineering
Cullen College of Engineering, UH
Dan Luss, Ph.D.
Professor of Chemical Engineering
Cullen College of Engineering, UH
The objectives of this research are to determine the kinetics of La1-xSrxCrO3 formation and to identify and solve engineering problems associated with largescale synthesis of these powders. This work will aid in the design of improved fabrication processes for Solid Oxide Fuel Cells (SOFC) and lead to improvements in component stability during SOFC operations. The kinetics of Lal-xSrxCrO formation are followed by timeresolved XRD, collecting data over a 5°ree;2- theta range every 30-60 seconds to quantify variations in mass fractions of various phases with time. Formation of solid solutions from chromites is followed by measuring peak shifts at about 32°ree;2- theta. We are investigating the impact of initial La/Sr ratio temperatures gas composition, and precursor particle size, with special attention given to SrCrO3 segregation, the presence of various phases during sintering and operation, and the thermal expansion coefficient, a major concern for stability at interfaces.
During this first year, mixtures of xSrCrO and (1-x)LaCr03, with x from 0 to 0.20 and calcined at 1260°ree; to 1310°ree; C to ensure complete conversion, were examined at various temperatures in the range 1000°ree; to 1200°ree; C. The position of the 32°ree;2- theta peak was found to be linear with x, thereby verifying Vegard's law and forming a calibration series. Complicating instrumental problems and techniques for the Rietveld analysis were solved. Kinetic measurements on the samples at 1050°ree; C confirm the feasibility of the analysis and indicate Avarami-type nucleation kinetics similar to those found in the 1-2-3 system. Reproducibility and other checks have been completed.
Back to Table of Contents
Irving N. Rothman, Ph.D.
Professor of English
College of Humanities, Fine Arts, and Communication, UH
ISSO provided the editorial support staff for the Woods Hole Conference of the National Space Grant Consortium and edited Mission to America. The Space Grant Consortium operates on the premise that "by the year 2000, NASA will have a prominent and permanent presence throughout the nation's academic institutions." The Space Grant Program will prove NASA's agent for change, coordination, and cooperation in technical education, for the enhancement of research infrastructure, and for the promotion of lifelong learning of math, science engineering, and technology. Editorial staff assisted in writing proposals for submission to NASA on 1000-day missions to the moon and Mars. Senior project editors included Bryan Bear, master's student in English Studies, and Chris Dow, English undergraduate senior who was to be named the outstanding English major the following year at the University of Houston. Engaged in tasks requiring compilation, documentation, writing and editing, student editors gained experience in publications and scientific resources.
The 1992-93 Annual Report of the Institute of Space Systems Operations listed 38 separate projects supported by the Institute in (1) aerospace power, propulsion, and electronics, (2) education, (3) environment, (4) fluids and thermal, (5) life sciences, (6) robotics, (7) software and analysis techniques, (8) space industrialization and economics, and (9) space science and astrophysics. The ISSO Annual Report (1992-93) comprised 48 pages of documentation and introduced research faculty to readers by means of the photographic portraiture of William Ashley. ISSO maintains the highest quality desktop proofreading capability, including the management of QuarkXPress for the publication of scientific documentation.
Back to Table of Contents
W. R. Sheldon, Ph.D.
Professor of Physics
College of Natural Sciences and Mathematics, UH
With ISSO support, W. R. Sheldon was an invited participant in the Balloon Measurement Workshop held at San Juan Capistrano, California, July 11-12, 1994. The workshop, organized by NASA headquarters, included presentations both by research workers and personnel involved in balloon flight operations. A report on the workshop may be obtained directly from Prof. Sheldon.
Prof. Sheldon detailed the University of Houston's capability to conduct in situ ozone measurements from a high-altitude balloon or during parachute descent, following a rocket-boosted ascent to the upper stratosphere. Measurements following a flight aboard a Nike-Orion rocket were presented. Measurements from the EASOE campaign during Arctic winter were also reviewed.
Back to Table of Contents
Cynthia Stokes, Ph.D.
Assistant Professor of Chemical Engineering
Cullen College of Engineering, UH
The efficacy of targeted therapeutics such as immunotoxins is directly related both to the extent of distribution achievable and the degree of drug internalization by individual cells in the tissue of interest. Factors that influence the tissue distribution of such drugs include drug transport, receptor/drug binding and cellular pharmacology, and the processing and routing of the drug within cells. To examine the importance of cellular pharmacology, we developed a mathematical model for drug transport in tissues that include drug and receptor internalization, recycling, and degradation, as well as drug diffusion in the extracellular space and binding to cellsurface receptors. We applied this "cellular pharmacology model" to a model drug/cell system, specifically, transferrin and the well-defined transferrin cycle in CHO cells.
We compared simulation results to models with extracellular diffusion only or diffusion with binding to cellsurface receptors. This comparison showed that inclusion of intracellular trafficking significantly increases the total transferrin concentration throughout the tissue while decreasing penetration depth (see Figs. 1 and 2).
 |
|
| Figure 1. Spatial profile of extracellular transferrin concentration compared for three cases: (a) diffusion in the tissue only with no drug/cell interactions; (b) diffusion with drug/cell binding but no intracellular trafficking; (c) diffusion, drug/cell binding and intracellular trafficking. Note that penetration of the tissue is impeded by taking into account the cellular pharmacology. | Figure 2. Spatial profile of total cell associated transferrin (cell surface plus intracellular) for two cases: (a) diffusion in the tissue with drug/cell binding but no intracellular trafficking; (b) diffusion, drug/cell binding and intracellular trafficking. Note that inclusion of in-tracellular processes greatly enhances the amount of cellassociated drug (b) compared to (a). |
Increasing receptor affinity or tissue receptor density reduces permeation of extracellular drug while increasing intracellular drug concentration, resulting in "internal trapping" of transferrin near the source; this could account for heterogeneity of drug distributions observed in experimental systems. Other results indicate that, when intracellular drug is required for a therapeutic effect, the optimal treatment may not result from conditions that produce the maximal total drug distribution. These results should help guide rational drug design and provide useful information for whole body pharmacokinetic studies.
Back to Table of Contents
Russell G. Thompson, Ph.D.
Professor of Decision and Information Sciences (DISC)
College of Business Administration, UH
This research focused on measuring the relative efficiency of lunar solar power (LSP) technology versus four major terrestrial technology options, specifically, fossil, nuclear, solar thermal, and solar photovoltaic. Data reported by U. S. Government agencies were used to estimate the inputs used to produce one gigawatt-year of electricity; data in the published literature were used to estimate the bounds on the modeled prices evaluated, e.g., externality costs. Data Envelopment Analysis (DEA) methods were used to measure relative efficiencies. The analysis showed LSP to be much more efficient than any of the other technology options; in fact, the resource cost of LSP could have been increased nearly 11-fold. Making that increase in costs and applying the multiplier bounds showed LSP still to be the most efficient technology option in the face of the modeled price bounds.
Another facet of the research involved an analysis of the impact of implementing LSP on the technical progress of the United States economy relative to other leading industrialized nations. It assumed that the United States would underwrite all of the startup costs and export electrical power to other nations (following startup). This implementation required all of the initial government funding in the startup phase to be paid back to the United States government, plus interest, within the first 20-year period. With LSP, modeling results showed that the United States would maintain its position on the efficiency frontier of leading nations and accelerate its total factor productivity (TFP) growth. However, in the absence of the LSP initiative, the U.S. economy failed to maintain its position on the efficiency frontier, and its TFP growth decelerated. Thus, LSP provides a positive economic perspective for the U.S. in the 21st century. Other leading nations may join the United States in this prosperity and with a cleaner environment.
Back to Table of Contents
Larry Witte, Ph.D.
Professor of Mechanical Engineering
Cullen College of Engineering, UH
Insulating materials are used abundantly for thermal protection of space hardware. Many insulating materials either used now or envisaged for use in future space applications are fibrous in nature, that is, they have fibers existing in a gas matrix. Nomex™ is such an insulator and has strong potential for space-related applications.
Futschik (ISSO Fellow, 1991-93) and Witte (1993) recently studied the behavior of fibrous insulation under the influence of different fillgases and fillgas pressures, both analytically and experimentally. The objective of this study was to examine theoretically the gas conduction portion of the effective thermal conductivity to see if the reason for the correction factor for CO2 can be determined. Comparisons were made between two models developed to predict the effective thermal conductivity of Nomex™ samples under different gas pressures. Experiments were conducted to demonstrate the applicability of the models. In this study, a good correlation was found between the model predictions and measurements of thermal conductivity for Nomex™ with air and nitrogen fillgases, but the model predictions for CO2 varied somewhat from the measurements. Applying a constant correction factor to the gas conduction component of heat transport for CO2 brought predictions into line with measurements.
Since the effective pore size of an open pore system like Nomex™ is not physically well defined because of fiber bundles and protruding free ends of fibers, researchers might conclude, as did Williams and Curry (1993), that the thermal conductivity of a gas within a fibrous material would require a correction factor independent of the gas type and based purely on temperature, pressure and an additional empirical factor which was density dependent. For purposes of this study, the form of the correction proposed by Williams and Curry was adopted and its dependence upon fillgas pressure determined for various fillgases in Nomex™.
Excellent agreement was achieved between experimental values of effective thermal conductivity for CO2 and the modified model. Results show that conventional models for the effective thermal conductivity of fibrous materials can be applied if an appropriate gas conduction correction factor is evaluated for a particular operating condition.
Back to Table of Contents
Lowell T. Wood, Ph.D.
Associate Professor of Physics
College of Natural Sciences and Mathematics, UH
Mohammad A. Rob, Ph.D
Assistant Professor of Natural and Applied Sciences
School of Natural and Applied Sciences, UHCL
The primary objective of the project was to develop a tunable coherent source in the 4.5-5.5 mm wavelength range by frequency doubling a TEA CO2 laser that operates in the 9-11 mm wavelength range. The nonlinear crystal selected for the frequency doubling was silver gallium selenide (AgGaSe2), a crystal transparent in the CO2 wavelength range with a large nonlinear coefficient. Once we have developed the tunable source, we will use it to detect atmospheric pollutants in the 3-5 mm atmospheric window.
Figure 1 shows that the experimental arrangement seeks to study second harmonic generation. Infrared radiation (9-11 mm) from the pulsed CO2 laser reflects from a flat mirror (M1), transmits through a beamsplitter (BS) that can be used to attenuate the signal, reflects from another flat mirror (M2), and then focuses from a 5-meter radius mirror onto the sample. Both the primary and second harmonic signal emerge from the sample, but a sapphire filter transmits only the second harmonic signal that strikes the detector.
Figure 1. Experimental arrangement for CO minimum detection limit.
The intensity 1(2w) of the second harmonic signal as a function of the relevant variables is given by
where mo and eo are universal constants, n(w) is the refractive index for the fundamental
frequency, n(2w) is the refractive index for the second harmonic signal, d is the
nonlinear coefficient for the crystal, L is the length of the crystal, I(w) is the
intensity of the fundamental signal, w is the frequency of the fundamental, and;
Δk is the difference between the propagation number of the second harmonic
signal and the fundamental signal.
The parameters that we could easily vary in our measurements are Δk and I(w). The last bracketed part of the equation pertains to the phase matching condition that allows one to produce a relatively large SHG signal. If Dk=0, then the bracketed term is one, and the fundamental and SHG signal are phasematched. The SHG signal reaches its maximum value. Dk can be varied by changing the angle of incidence of the fundamental signal on the face of the crystal and adjusting the angle until a maximum SHG signal is obtained. I(w) can be varied by controlling the output power of the CO2 laser itself.
Before one can maximize the SHG signal, it is important to characterize the beam from the CO2 laser to make sure the beam is uniform and does not overfill the face of the sample. This ensures that no energy is lost so that the sample interacts with the fundamental signal to produce a large SHG signal. Figures 2-4 show the beam profile at the three different points indicated on Fig. 1.
 Figure 2. Beam Profile at point A. |
 Figure 3. Beam profile at point B. |
In each case, the solid line is a Gaussian fit and the dots are the data points. Points A and B are essentially the same as if no focusing of the beam has occurred. The beam is approximately Gaussian and about 4 mm in diameter. Figure 4 shows the focused beam as it enters the sample. The beam is about 2 mm in diameter so it nicely fits onto the face of the sample we used.
Figure 4. Beam profile at point C.
Figure 5 shows the SHG power as a function of the angle of incidence on the crystal. The solid line shows the fitted curve; the dots are the experimental data. The sin x/x behavior is clearly evident and the phase matching condition has at least partially been met, at least in the plane of rotation used for these data.
Figure 6 shows the SHG intensity I(w) as a function of the fundamental intensity I(w). Theory predicts a quadratic dependence; the solid line is the fitted curve and the dots are the data. Once again, the data and theory are in good agreement.
In summary, our initial measurements have demonstrated that we can obtain SHG signals that have the behavior predicted by theory. The single major problem we have is that the SHG signal efficiency is small, only about 2 percent. We should be able to improve this because the theory predicts that about 15 percent should be feasible. We believe that we need better control of both angles which define the angle of incidence and that the beam may need to be focused a little tighter. It is also possible that the output end of the crystal has some damage and may need to be recoated. We have ordered a smaller radius mirror to achieve a tighter focus and are considering an improved mounting system so that we have better control of the incident angle on the crystal.
 Figure 5. Power vs. external angle. |
 Figure 6. Output power vs. input power. |
Note on graphs: Some of the units on the graphs do not match the quantities because conversions have not yet been made. In all cases, however, the values shown are directly proportional to the quantities represented by the units.
Contents
ISSO -- Institute for Space Systems Operations
1994-1995 Annual Report
|
|
