Organization of Chemical Engineering Graduate Students University of Houston

22nd Annual Symposium – 2007


Oral Presentations


Steven M. Kemper1, Tim Sherlock2, Kaajal D. Shah3, Eliedonna Cacao1, Paul Ruchhoeft2, Robert L. Atmar4,
Richard C. Willson1,3
1 University of Houston (UH) Department of Chemical and Biomolecular Engineering, 2 UH Department of Electrical and Computer Engineering, 3 UH Department of Biology and Biochemistry, 4 Baylor College of Medicine, Houston, TX

Many current bioanalytical methods rely on labeling of the species of interest and the use of specialized detection equipment to monitor the label. Fluorescence and luminescence are common techniques, but suffer from photobleaching and signal loss from dispersion of the emitted signal light in all directions. Retro-reflectors may offer a solution to this problem, as they return a light signal back to the source in a narrow directional beam. Retro-reflectors are among the most detectable man-made objects, with corner-cube retro-reflectors having been placed on the lunar surface for use in measuring the distance between the earth and moon.

While the high signal fidelity from fully pre-assembled retro-reflectors makes them attractive as taggants or labels, our current work focuses on analyte-responsive modulation of signal from micron-scale retro-reflectors. In this approach, a gold corner-cube retro-reflector is fabricated on a silicon wafer prior to functionalization with antibodies specific to the antigen of interest, eg Norwalk virus. The surface is then exposed to the sample and any virus will attach to the antibodies. Gold nanoparticles similarly coupled with another antibody are then added to the surface, and light scattering reduces reflectance if analyte is present. A similar device can be envisioned for in vivo glucose monitoring, wherein glucose competes with a dextran bound to gold nanoparticles for a covalently bound lectin. As glucose level increases, the gold-bound sugar is displaced from the surface and overall reflectance increases. In addition to high detectability, these devices would benefit from being easily producible and robust. Initial studies using microfabricated retro-reflectors show promise for these applications.

Katerina Kourentzi, Mohan Srinivasan, Sandra J. Smith-Gill and Richard C. Willson
Department of Chemical and Biomolecular Engineering, University of Houston, Houston, TX 77204-4004 USA

Recent therapeutic successes with humanized forms of monoclonal antibodies have vitalized interest in a mechanistic understanding of the details of antibody-antigen recognition. We have examined the energetics and kinetics of association of three structurally-characterized anti-lysozyme (HEL) antibodies (H8, H10 and H26) recognizing a common epitope on hen egg white lysozyme (HEL), and the avian variant Japanese quail egg white lysozyme (JQL). Competitive dissociation induced by addition of excess unlabeled HEL after varied periods of antibody-antigen association was followed in real time using fluorescence anisotropy. Dissociation was in many cases non-single-exponential and observed off-rates became slower as complex age increased, implying multi-step association kinetics (consistent with an encounter-docking view of protein-protein interactions). The fully-docked fraction of the complexes just prior to inducing dissociation was high for the HEL complexes but was dramatically reduced for JQL complexes, i.e. final docking was antigen-mutation sensitive. Calorimetric characterization of association energetics suggests that the cross-reactivity of H8 may be mediated by a combination of conformational flexibility and less-specific intermolecular interactions. On the other hand, minimal structural perturbations occur in the intramolecular salt-bridge rigidified H26 and the large loss of enthalpic energy with mutant antigen JQL reflects its specificity. The observed changes in enthalpic energy correlate well with the observed differences in functional behavior of the antibodies, and indicate that the thermodynamic characteristics of cross-reactivity and of specific recognition are fundamentally different.

Veselina Uzunova, Dr. Peter G. Vekilov, PhD

Hemoglobin S (HbS) polymerization is one of the best-studied phase transitions. This process is a primary pathogenic event in a severe debilitating disease known as sickle cell disease (SCD). So far the therapy of SCD is palliative and not at a satisfactory level. The basic idea behind research in the area is that delaying the polymerization with ameliorate the symptoms of the disease and blocking polymerization will lead to ultimate curing. Since the clinical variability of the disease is great even in genetically identical individuals, it is obvious that not only genetics underlines this condition. The properties of the mutant protein and the thermodynamic and kinetics of the process have been studied extensively in order to uncover the mechanism of this phase transition. The possible role of small molecules as modulators of the process deserves attention. We investigated the effects of heme on the first stage of polymerization – nucleation.

HbS from a patient with SCD was purified and nucleation rates and delay times of polymerization were measured at different temperatures and concentrations of HbS. The rest of the stock solution was dialyzed. Immediately after dialysis an attempt was made to measure the nucleation parameters but it was found that the hemoglobin does not nucleate.

After addition of heme nucleation was resumed. The effects of addition of 3 different heme concentrations were measured at different temperatures and compared to the measured nucleation rates before dialysis of HbS. In order to address the question of possible release of heme from hemoglobin we followed the mass spectra of HbS solution aged at room temperature in three consecutive days and found that indeed heme is released and the release of heme in vitro increases with time under those conditions.

Tirtha Chatterjee and Ramanan Krishnamoorti
Department of Chemical and Biomolecular Engineering, University of Houston, Houston, TX, 77004

Polymer nanocomposites are widely considered as the next generation materials where nanoscale fillers dispersed in polymer matrices extend their utility while maintaining the processing flexibility inherent to the parent polymer. Single walled carbons nanotubes (SWNTs) with their remarkable set of intrinsic properties are strong candidates as nanoparticles to develop polymer nanocomposites. Despite this promise, the successful dispersion of the tubes and its proper characterization are the most significant challenges. Due to the strong intertube cohesive forces, nanotubes tend to form clusters. This along with the high anisotropy of the individual tubes, leads to a percolated tube network structure at extremely low loading. Hence, a detailed understanding of the network structure and related properties is fundamentally important from both the industrial and academic points of view.

In this research, SWNTs are successfully dispersed in poly(ethylene oxide) matrix using lithium dodecyl sulfate , an anionic surfactant, as a compatibilizer. A high state of dispersion is characterized by a low geometric percolation threshold (pc) (at ~ 0.09 vol % SWNTs loading) as obtained from melt state rheological measurements. Systematic investigation of the network structure for semidilute tube dispersion (3 ≤ p/pc ≤15) reveals a hierarchical fractal network structure with mass fractal dimension 2.3 ± 0.2. The network consists of micron length tube cluster which shows concentration invariance as opposed to the mesh size (ζ) which shows diffusion limited dependence with concentration (ζ ~ p-0.4±0.08).

As a direct consequence of the self-similar fractal network, the linear viscoelastic parameters display ‘time-temperature-composition’ superposition. But remarkably, this superposability can be extended for non-linear deformations when the non-linear properties are scaled by the local strain experienced by the elements of the network.

Saurabh Y. Joshi, Michael P. Harold, Vemuri Balakotaiah

We derive simplified low-dimensional models for real time simulation, control and optimization of catalytic converters used in automobiles. Liapunov-Schmidt (LS) technique of bifurcation theory is used to average the convection-diffusion-reaction equations in the transverse direction. The resulting low-dimensional models are described by a system of differential algebraic equations involving multiple concentration and temperature modes. These “multi-mode models” capture all the important features involving exchange of mass and thermal energy between different phases. For the case of CO oxidation reaction carried out in a catalytic monolith, we compare the solutions obtained by the averaged model and detailed model. It is observed that there is a very good agreement between both the predictions. Since the solution of low dimensional models requires considerably less computation time while retaining all the parameters and essential physics of the detailed model, they can be used for accurate design and control of catalytic monoliths used in automobiles.

A. M. Sani, and K. K. Mohanty

During the process of crude oil/gasoline loading and storage, significant amounts of volatile organic compounds (VOC) are emitted to the environment. VOC emission pollutes the environment, is a fire hazard and a threat to workplace and people’s safety. Our goal is to develop stable aqueous foam formulations to control the VOC emission by providing a mass transfer barrier during the process of loading crude oil/gasoline

In this work novel aqueous foams have been formulated by incorporating nano-clays, polymers and surfactants in an aqueous solution. The use of nano-clays in the foam formulation creates continuous interlinked and overlapping layers that significantly reduced the rate of vapor diffusion through the lamella and through the whole foam network for the first 40 hours. It also decreases the rate at which the liquid drains out of the foam and thus decreases the rate of quality increase. Increase in polymer and nano-clay concentration significantly increases foam stability. The stability and mass transfer of these novel foams were also investigated at high temperatures (105 and 125 oF) and they were found to be highly stable both in the presence of gasoline and crude oil loading processes. These foam systems should be generally safe and technically and logistically easy to handle.

Nitika Kalia, Dr. Vemuri Balakotaiah

Carbonate reservoirs containing oil and gas deposits are routinely treated with acid to increase production. The acid reaction with carbonate rocks results in the formation of complex patterns, such as wormholes. The reactive dissolution phenomenon is governed by the operating conditions such as injection rate, type of acid and the mineralogy of the formation (calcite, dolomite). The kind of patterns formed, wormhole density and wormhole penetration depth in a medium also depend on the nature of flow (radial or linear) and medium heterogeneities. In this work, both radial and linear flow of acid in a porous medium is theoretically investigated and the phenomenon is simulated using a two-scale continuum model. The three main types of patterns observed experimentally, namely, compact, wormhole and uniform patterns are numerically simulated. The fractal nature of wormholes observed in experiments is confirmed through simulations and the fractal dimension is quantitatively matched. The dependence of wormhole fractal dimension, optimum injection rate and minimum pore volumes required to breakthrough the medium on medium heterogeneity is investigated in detail for both linear and radial flows. The heterogeneity study involves studying the core-scale heterogeneities and introducing a new way of quantifying the magnitude and length scale. The proposed method is validated by numerical simulations and the key results are noted.


Poster Presentations

Mohan Boggara, Dr. Krishnamoorti

Non-steroidal anti-inflammatory drugs (NSAIDs) are the most widely prescribed drugs worldwide [1] for their pain, fever and inflammation reducing action. Chronic usage of NSAIDs, however, leads to gastrointestinal (GI) toxicity. The mechanism by which NSAIDs cause GI toxicity remains unclear and recent clinical evidences strongly point that the cause could be due to direct interactions between NSAIDs and phospholipid membranes. However, the molecular details of such interactions are not fully elucidated. Also, NSAIDs pre-associated with phospholipid vesicles are purported to be safer and therapeutically more effective alternatives to the unmodified ones.

Our initial studies on the partitioning of two most common NSAIDs (Aspirin and Ibuprofen), by experiments and computer simulations, clearly indicate the role played by the structure of the drug in their interaction with the lipid membrane. Motivated by those results, we systematically performed molecular dynamics (MD) simulations of lipid membranes with various NSAIDs that are of different size, structure and pKa values. Our MD results suggest high partition coefficients for these NSAIDs in bilayer membrane as compared to water and strong thinning effect on the bilayer in the presence of NSAIDs. Also, our recent neutron scattering studies (small angle neutron scattering and neutron reflectivity) on DMPC-Ibuprofen systems indicate that the drug affects both the ~5 nm thick bilayer as well as the overall ~ 100 nm diameter vesicle, indicating that NSAIDs affect lipid vesicles on various length scales. In this presentation, we will discuss the structural perturbations to lipid membranes due to NSAIDs at clinically relevant molar ratios to lipid molecules and other aspects like drug diffusion mechanism, size, charge state and hydrophobicity and their implications on the use of lipid vesicles as drug delivery vehicles for NSAIDs.

1. Donnelly, M. T. and C. J. Hawkey (1997). “Review article: COX-II inhibitors-a new generation of safer NSAIDs?” Aliment Pharmacol. Ther 11: 227-236.

Michael Clark, Dr. Krishnamoorti

Carbon nanotubes are unique materials, which are at the forefront of the nanotechnology revolution due to their extraordinary mechanical, thermal, and electrical properties. They are amongst the strongest materials known to man making them ideal candidates as reinforcing elements in ceramic based materials. Their use in this capacity is contingent on dispersing the nanotubes throughout the host material and fostering nanotube – matrix interactions while ensuring the integrity of the nanotubes. However, strong inter-tubular van der Waals forces facilitate nanotube bundling limiting their solubility in common solvents and ultimately, the host matrix. Furthermore, the impurities often present in synthesized carbon nanotubes may interfere with nanotube – matrix interactions and are detrimental to both nanotube thermal stability and overall nanocomposite performance.

We are primarily interested in understanding the fundamental thermodynamics that govern the solution behavior of carbon nanotubes and using that knowledge to successfully generate ceramic nanocomposites with enhanced mechanical and thermal properties. In that pursuit, we have successfully developed a purification procedure to eliminate impurities that undermine our ability to analyze generated thermodynamic data. Furthermore, this purification process has yielded a dramatic increase in nanotube crystallinity and thermal stability, making the carbon nanotubes more compatible with high temperature ceramic processing techniques.

Ratndeep Srivastav, Dr. Strasser

Fuel cells are long been pursued as the vibrant technology for generating electricity as upshot of growing consensus to adopt clean energy strategy and the expansion of much power consuming utility horizons. We have been working on a novel class of de-alloyed Pt-alloy materials which have shown tremendous activities in RDE as well as single PEMFC’s. Active cathode electrocatalyst layer was prepared by in-situ de-alloying using cyclic voltammogram of a base metal rich cathode catalyst layer. The electrocatalyst was prepared by unique liquid precursor based freeze drying method. Various Pt-alloy compositions were prepared; the powders were further alloyed using the hydrogen reduction method. They were annealed under Ar/H2 (96% Ar) environment at 600oC to control the particle size of the catalyst. XRD, TEM analysis on the catalyst shows small sized particles of the order 3-4 nm. To investigate the ORR activity of these catalysts, they were examined under a rotating disk electrode and as cathode catalyst in a H2/O2 fuel cell. Electrode layer were prepared by CCM technique, further the catalyst layer was de-alloyed in-situ a fuel cell and then ion-exchanged with sulfuric acid to remove any base metal left inside the fuel cell.

Shirlaine Koh1, Chengfei Yu1, Jennifer E. Leisch2, Michael F. Toney2 and Peter Strasser1
1 Department of Chemical and Biomolecular Engineering, University of Houston, Houston, Texas 77204 (USA)
2 Stanford Synchrotron Radiation Laboratory (SSRL), Stanford Linear Accelerator Center (SLAC), Menlo Park, California 94025 (USA)

Polymer Electrolyte Membrane fuel cells (PEMFC) is a type of hydrogen fuel cells that runs at relatively low temperatures. It converts chemical energy of a fuel directly into electrical energy with high efficiency and low emission. Currently, Pt is the best catalyst for this energy conversion. However, recent work had shown that Pt alloys exhibit activity improvements of up to 3 times over pure Pt. For fuel cells to be commercially feasible, the catalyst activity needs to be improved by at least a factor of 4 over pure Pt and need to maintain these activities over sustained periods of time. To design and discover catalyst with the required activity characteristics, we started by understanding the fundamental structure-activity relationships of binary and ternary Pt alloys as well as their activity-stability relationships. Electrocatalytic mass and Pt-surface-area based activities were measured using rotating-ring-disk-electrode while changes in catalytic particle composition, uniformity and size distribution before and after electrochemical testing were obtained using Energy Dispersive spectroscopy (EDS), and X-ray Diffraction (XRD).

In this work, structure-activity-stability relationships of Pt-Cu binary electrocatalysts will be presented. For these Pt-binary alloy electrocatalysts, multiple phases and smaller particle sizes were observed when catalysts were annealed at 600 °C while single phase and larger particle sizes were seen in those annealed at 950 °C. The de-alloying process created highly active Pt nanoparticle electrocatalysts. It was observed that de-alloying of base metal rich phases generally resulted in more active de-alloyed particles compared to base metal deficient ones. Comparing the compositional data from XRD with that from EDS suggested the presence of lattice strain in the catalysts upon base-metal dissolution. The presence of strain was further confirmed when they were subsequently relaxed by further annealing at higher temperatures. Our experimental findings together with preliminary Density Function Theory (DFT) computations from our collaborator
indicated that compressive lattice strain improves the electrocatalytic surface reactivity of Pt.

References

1. Koh, S.; Strasser, P., JACS Comm. (2007) (Accepted)
2. Gasteiger, H. A., Kocha, S. S., Sompalli, B. and Wagner, F. T. Appl. Catalysis B: Environmental 56, 9-35 (2005)
3. Koh, S.; Yu, C.; Strasser, P. ECS fall meeting transactions, Washington, DC, 2007

Ajish S. R. Potty (1), Katerina Kourentzi (1), Han Fang (2), George W. Jackson (1#), Peter Schuck (3), Dar-Chone Chow (2), Glen Legge (4), and Richard C. Willson (1,4*)
(1) Department of Chemical and Biomolecular Engineering, University of Houston 4800 Calhoun Rd, Houston, TX 77204-4004, USA
(2) Department of Chemistry, University of Houston, 4800 Calhoun Rd, Houston, TX 77204-5003, USA
(3) Division of Bioengineering and Physical Science, National Institute of Health 13 South Dr, Bethesda, MD 20893-5766, USA
(4) Department of Biology and Biochemistry, University of Houston 4800 Calhoun Rd, Houston, TX 77204-5001, USA
#Present address: BioTex, Inc., 8058 El Rio St, Houston, TX 77054, USA
*Author for correspondence: fax: 713 743-4323; willson@uh.edu

Aptamers are RNA or DNA molecules which selectively recognize and bind pre-selected target molecules. Aptamers bind with high affinity and specificity to a variety of targets; one is now an FDA-approved pharmaceutical for treatment of wet macular degeneration by sequestration of vascular endothelial growth factor (VEGF). Our research primarily involves understanding the underlying features of aptamer-target recognition.

To this end, we have characterized the binding of DNA and RNA aptamers with hen egg white lysozyme (HEL) or vascular endothelial growth factor (VEGF) using fluorescence anisotropy, isothermal titration calorimetry (ITC), surface plasmon resonance (SPR) and analytical sedimentation. The anti-HEL DNA aptamer (despite being polyanionic) is selective for HEL over cationic cytochrome C. However, binding of HEL to the aptamer is strongly sensitive to ionic strength, and is greatly suppressed at 100 mM NaCl. ITC and fluorescence anisotropy experiments demonstrated that association is entropically-driven at 25 C, and accompanied with the release of 2.3 monovalent ions. The association (> 10^8 M^-1 s^-1) and dissociation (> 10 s^-1) rates were too fast to quantify.

On the other hand, binding of anti-VEGF aptamer to VEGF is more robust to salt, and is enthalpically-driven at 20 C. Moreover, the kinetics is characterized by fast association (Kon = 2.7 * 10^4 M^-1s^-1) and slow
dissociation (Koff = 4.2 * 10^-4 s^-1), typical of most antibody-antigen interactions. Furthermore, we examined the key recognition hot-spots by a combination of mutations, truncations and extensions. Many single-nucleotide mutations completely suppress binding while some have a minimal effect. Moreover, we observed the binding stoichiometry to be 1(aptamer):1(VEGF dimer), suggesting the aptamer wraps around the native VEGF dimer to prevent access to its extracellular biological ligand.

Binh Vu, Dr. Willson

Gold nanoparticles were recently reported to reduce the formation of non-specific products in PCR at remarkably low temperatures. In contrast to these reports, we report that gold nanoparticles do not enhance the specificity of PCR, but rather suppress the amplification of longer products while favoring amplification of shorter products, independent of specificity. Gold nanoparticles bearing a self-assembled monolayer of hexadecanethiol did not affect PCR, suggesting that surface interactions play an essential role. This role was further confirmed by experiments using gold nanoparticles ranging in size from 20 nm to 200 nm, in which a similar effect on PCR was observed for the same total surface area of particles over a 100-fold range of per- particle surface areas. The effect was seen with Taq and Tfl polymerases but not with Vent polymerase, suggesting the involvement of polymerase in nanoparticle effects. The effects of nanoparticles can be reversed by increasing the polymerase concentration or by adding BSA. Transient high-temperature nanoparticle exposure of PCR mix containing polymerase but not template or primers, followed by nanoparticle removal, modified subsequent nanoparticle-free PCR. Taken together, these results suggest that the surface interactions which are most important are those of the polymerase, rather than with the template or primers.

Katerina Kourentzi, Mohan Srinivasan, Sandra J. Smith-Gill and Richard C. Willson
Department of Chemical and Biomolecular Engineering, University of Houston, Houston, TX 77204-4004 USA

Recent therapeutic successes with humanized forms of monoclonal antibodies have vitalized interest in a mechanistic understanding of the details of antibody-antigen recognition. We have examined the energetics and kinetics of association of three structurally-characterized anti-lysozyme (HEL) antibodies (H8, H10 and H26) recognizing a common epitope on hen egg white lysozyme (HEL), and the avian variant Japanese quail egg white lysozyme (JQL). Competitive dissociation induced by addition of excess unlabeled HEL after varied periods of antibody-antigen association was followed in real time using fluorescence anisotropy. Dissociation was in many cases non-single-exponential and observed off-rates became slower as complex age increased, implying multi-step association kinetics (consistent with an encounter-docking view of protein-protein interactions). The fully-docked fraction of the complexes just prior to inducing dissociation was high for the HEL complexes but was dramatically reduced for JQL complexes, i.e. final docking was antigen-mutation sensitive. Calorimetric characterization of association energetics suggests that the cross-reactivity of H8 may be mediated by a combination of conformational flexibility and less-specific intermolecular interactions. On the other hand, minimal structural perturbations occur in the intramolecular salt-bridge rigidified H26 and the large loss of enthalpic energy with mutant antigen JQL reflects its specificity. The observed changes in enthalpic energy correlate well with the observed differences in functional behavior of the antibodies, and indicate that the thermodynamic characteristics of cross-reactivity and of specific recognition are fundamentally different.

Xing Zhang 1, Ajish S. R. Potty 2, George W. Jackson 3, George E. Fox 1, and Richard C. Willson 1, 2
Department of Biology and Biochemistry 1, Department of Chemical Engineering 2, University of Houston, and BioTex, Inc. 3, Houston, TX

Aptamers are library-selected DNA or RNA ligands that bind to pre-selected target molecules. In this work, we inserted known aptamer-coding sequences into a plasmid containing an artificial 5S rRNA deletion cassette that is strongly expressed in E. coli under T7 control. The expressed RNA is structurally similar to 5S rRNA and accumulates to high levels in E. coli but is not incorporated into 70S ribosomes. A surface plasmon resonance competitive binding assay demonstrated that a VEGF aptamer/5S rRNA chimera produced in vitro by transcriptional runoff could compete with VEGF DNA aptamer for VEGF, suggesting binding between the VEGF “ribosomal-RNA aptamer” and VEGF. Separately, a 5S rRNA chimera displaying an aptamer known to increase the fluorescence of malachite green (MG) also enhanced MG fluorescence. Wild-type 5S rRNA neither binds VEGF nor enhances MG fluorescence. These results indicate that the aptamer’s useful properties are not lost when it is expressed in the context of the stable 5S rRNA carrier. Inclusion of the aptamer in the carrier may facilitate production of large quantities of RNA aptamers with potentially extended half-lives in vivo, and may open the possibility of screening aptamer libraries in vivo.