Abstracts

External stress can accelerate molecular mobility of amorphous solids by several orders of magnitude. The changes in mobility are commonly interpreted through the Eyring model, which invokes an empirical activation volume whose origin remains poorly understood. Here, we analyze constant-stress molecular dynamics simulations and propose an extension of the Eyring model with a machine-learned field, softness. Our model connects the activation volume, which is an empirical parameter to a structure property (softness) for the first time. Our results show that stress-induced mobility depends on local structure and explains the narrower distribution of relaxation time observed under stress.

The unique structure of bottlebrush polymers, featuring a linear polymer backbone grafted densely with polymer side-chains, allows for the creation of interesting and unique morphologies. By combining structurally dissimilar backbone and grafts, core-shell bottlebrushes serve as candidates for materials with unique combinations of properties, such as mechanically rigid backbones combined with conductive grafts, a combination which could be useful as membranes in battery applications. In this talk, we present a field-theoretic model for the structure of bottlebrush copolymers and explore how architecture alters the traditional block copolymer phase diagram. We accomplish this via our recently developed theoretically informed Langevin dynamics (TILD) package for LAMMPS, in which the non-bonded forces are calculated using density fields in lieu of particle-based interactions. The TILD simulations capture the fluctuating version of the model, which is compared to the analytic RPA predictions for the ODT boundaries. Finally, we provide insight into the changes in the chain conformations and relate them to the shifts in the phase boundaries we observe. 

The effect of catalyst addition on H2 evolution from composite electrodes of La0.7Sr0.3TiO3 (LST) and BaZr0.1Ce0.7Y0.1Yb0.1O3−δ (BZCYYb) was studied. Starting with symmetric cells (LST||BZCYYb||LST), Pt was added to one or both electrodes, after which i-V polarization measurements were performed in humidified H2 at 723 and 773 K. The base cells showed very high impedances but these decreased dramatically upon addition of Pt to both electrodes. When Pt was added to only one electrode, the cells performed as diodes, showing that Pt was necessary for H2 dissociation but not for H recombination. The effects of adding Ru, W, Re and Fe were also studied. DFT calculations helped confirm that H recombination on BaZrO3 is expected to be barrierless. The implications of these results for potential application to electrochemical synthesis of ammonia are discussed. 

Syk and Src family kinases (SFK) inhibitors interfere with signaling from GPVI, α2β1, αIIbβ3, and GPIb-IX-V to reduce thrombotic risk or induce bleeding episodes. Collagen-mediated clustering of platelet GPVI results in phosphorylation of SFKs such as Lyn and Fyn, and active Lyn is constitutively bound to GPVI to allow rapid signaling. During clotting under flow, the generation of fibrin can have diverse influences on platelet signaling by sequestering thrombin and potentially activating GPVI signaling within the clot interior. These inhibitors tackle the thrombus formation at earlier stages since the platelets reach the activation surface. Direct inhibition of GPVI was used to compare the difference between inhibition of subsequent pathways. Using microfluidics, the effects of these inhibitors can be explored under defined hemodynamic flows and procoagulant surface triggers. Additionally, the drug may be present in the blood at desired time of clotting by perfusion switching to drug-treated blood. This experimental design allows exploration of platelet response at different stages of clotting through the measurement of drug potency to modulate clotting on different procoagulant surface conditions.

The effect of adding transition-metal catalysts to the surfaces of Solid-Oxide-Fuel-Cell (SOFC) cathodes was examined. Monolayer amounts of Pd, Ni, and Co were deposited by atomic layer deposition (ALD) to cathodes prepared by impregnation of La0.8Sr0.2FeO3 (LSF), La0.8Sr0.2CoO3 (LSCo), and La0.6Sr0.4 Co0.2Fe0.8O3 (LSCF) into YSZ scaffolds. Incorporating the catalysts by ALD allowed modification of the surface composition without changing the cathode surface area or structure. X-ray Photoelectron Spectroscopy and High Sensitivity-Low Energy Ion Scattering were used to confirm ALD growth rates that were measured gravimetrically. Electrochemical impedance spectroscopy indicated that the addition of Pd had a small negative effect on all of the cathodes. Ni and Co exhibited only a small effect on LSCo and LSCF cathodes but both decreased the cathode impedance for LSF. It is suggested that an essential feature in cathode promotion by Co and Ni is that these cations are incorporated into the perovskite surface and thereby increase the concentration of oxygen vacancies, while Pd only blocked adsorption sites. 

The potential to add catalytic activity to inert perovskite anodes was also investigated using ALD. Solid Oxide Fuel Cells (SOFC) with La0.3Sr0.7TiO3 (LST)–yttria-stabilized ZrO2 (YSZ) anodes were prepared by impregnation of LST into porous YSZ scaffolds and then modified by Atomic Layer Deposition (ALD) of Ni, Pt, Pd, Fe, Co. and CeO2. High Sensitivity-Low Energy Ion Scattering were used to confirm ALD growth rates that were measured gravimetrically.  Weight loadings as low as 0.01% of Pt, Ni, and Pd were sufficient to decrease anode impedances by orders of magnitude for operation in humidified H2 at 973 K. The effects of CeO2, Co, and Fe were less but still significant. 

¹Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA

²Department of Bioengineering, University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA

³Department of Electrical and Systems Engineering, University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA

Hydrogel particles (microgels) generated using microfluidic methods have superb properties such as high size uniformity and precise control over degradation and release profiles, making them useful for applications in wound healing and injectable drug delivery. However, the throughput of microfluidics is constrained by the physics governing the flow of immiscible fluids confined within microchannels. This throughput tends to be several orders of magnitude lower than what would be necessary for commercial and clinical applications. Here, we demonstrate the scaling up of on-chip synthesis of microgels by parallelizing the microfluidic channels. Taking advantage of the established fabrication technologies developed by the semiconductor industry and a high flow control system, a 4-inch silicon microfluidic chip integrating more than 4,000 microfluidic devices is developed. By incorporating a high energy flood UV source, this chip allows the synthesis of poly (ethylene glycol) diacrylate microgel particles with diameter down to 30 µm at a throughput above 1kg/hr. By using photomasks that enable milli-second scale control of the UV exposure, the stiffness of microgels can be varied between 103 to 104 Pa. Large-scale production of microgels will enable construction of large-scale tissue scaffold with well-defined physiochemical properties, and will provide scalability for translation to clinical settings. 

Polymer nanocomposites are an important class of novel, synthetic materials made by the inclusion of a
wide variety of nanoparticles in polymer melts. The challenge to making these unique materials in the
laboratory has been to develop scalable processes to achieve desired morphology. Leaching-enabled
Capillary Rise Infiltration (LeCaRI) is a novel, unique technique for the creation of polymer composites by
inducing motion of high-mobility polymers from gels into the pores of a disordered nanoparticle packing.1,2
The resultant composite structure falls under the class of highly-filled PINFs (or Polymer Infiltrated
Nanoparticle Films) where the high loading of NPs in the polymer composite and their dispersed state
gives unique properties to these composites.3,4 The details of this process and effect of atmospheric
humidity, by way of capillary condensation on the infiltration thermodynamics, will be discussed in my
talk. By infiltration of polymers from free-standing gels instead of glass polymer films, we can tap into
another interesting feature: the ability to pattern the gel to make patterned composites. Patterned
composites find a wide variety of applications in generating optical lenses, microelectronic devices, and
selective wetting surfaces. The patterned composites so produced have microdomains of composites in
the nanoparticle packings. These composites can either be self-erasing(due to the slow motion of the
leached polymers out of the patterned domains into the hydrophilic non-patterned domains) or
permanent(due to the polymers being frozen into space by UV-initiated cross-linking). Polymer transport
can be studied by tracking the diffusive front of polymer from a patterned domain into the non-patterned
regions. We observe that the front motion shows dependence on the atmospheric humidity owing to the
presence of capillary-condensed water in the nanopores of the silica packings. The presence of capillary
bridges of water inside the high-curvature necks and pores inside the packings makes it an interesting
problem where two species: polymer and water are drawn by capillarity to explore narrow regions of the
porous domain. I will present results on the dynamics of lateral motion of polymer in the nanoporous
LeCaRI films.

1. Venkatesh, R. B., Han, S. H. & Lee, D. Patterning polymer-filled nanoparticle films via leaching-
enabled capillary rise infiltration (LeCaRI). Nanoscale Horizons 4, 933–939 (2019).

2. Tran, H. H., Venkatesh, R. B., Kim, Y., Lee, D. & Riassetto, D. Multifunctional composite films with
vertically aligned ZnO nanowires by leaching-enabled capillary rise infiltration. Nanoscale 11,
22099–22107 (2019).
3. Venkatesh, R. B. et al. Polymer-Infiltrated Nanoparticle Films Using Capillarity-Based Techniques:
Toward Multifunctional Coatings and Membranes. Annu. Rev. Chem. Biomol. Eng. 12, 411–437
(2021).
4. Donovan, B. F. et al. Elimination of Extreme Boundary Scattering via Polymer Thermal Bridging in
Silica Nanoparticle Packings: Implications for Thermal Management. ACS Appl. Nano Mater. 2,
6662–6669 (2019).

^aDepartment of Materials Science and Engineering, ^bDepartment of Mechanical Engineering and Applied Mechanics and ^cDepartment of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, Pennsylvania 19104, United States

The mechanical properties of disordered nanoparticle (NP) packings can be significantly improved through the infiltration of polymers to enhance the interactions between the NPs. Previous nanoindentation-based fracture measurements have shown that infiltrating polymers into disordered NP packings via capillary rise infiltration (CaRI) is a simple and highly effective route to toughen these packings. However, it is challenging to fully understand how the relative size of polymer and nanoparticles and the extent of confinement affect the toughness. In the present work, a thin-film fracture testing method based on the double cantilever beam (DCB) specimen is developed and used to investigate the fracture properties of polymer-infiltrated nanoparticle films. In the DCB specimen, a crack is propagated in NP films over distances of tens of millimeters, allowing for highly accurate measurements of toughness in a mode I (tensile opening) configuration. The fracture toughness of the polymer-infiltrated NP films is found to be strongly dependent on polymer molecular weight (MW) and NP size. Low MW, unentangled polymers, effectively toughen small NP packings; whereas high MW, entangled polymers, show enhanced toughening in large NP packings. Possible toughening mechanisms, including confinement-induced molecular bridging and polymer entanglement, will be discussed.

¹Institute for Medicine and Engineering, ²Department of Chemical and Biomolecular Engineering

University of Pennsylvania, 1024 Vagelos Research Laboratories, Philadelphia, PA 19104

Platelet contractility plays a crucial role in clot contraction to provide rigidity and stability to thrombi. Clot contraction has been studied extensively in static conditions, but there are fewer studies that evaluate how shear flow can affect platelet contraction. In particular, there have been limited studies evaluating the role of secondary platelet aggregation on platelet contraction under flow. Here, we wanted to evaluate how inhibition ADP and thromboxane A2 (TXA2) would affect clot contraction. We ran PPACK-treated whole blood ± acetylsalicylic acid (ASA), 2-MethylthioAMP (2-MeSAMP), and/or MRS-2179 over collagen (100 s-1) for 7.5 minutes, then stopped flow to observe contraction for 7.5 minutes. We developed two automated methods of quantifying platelet contraction: (1) a one dimensional, “global” measurement of clot length in the direction of flow, and (2) a two dimensional, “local” measurement of the change in area coverage of platelets in a clot. We found that platelet fluorescence intensity (FI) decreased and global platelet contraction was inhibited when ASA, 2-MeSAMP, or MRS-2179 were added to inhibit TXA2 or ADP production. We observed a correlation between platelet FI and global platelet contraction (R2 = 0.72). Unlike global platelet contraction, local platelet contraction was more pronounced across all conditions; however, we observed that in conditions with ASA, there was significantly reduced local platelet contraction relative to conditions without ASA. We also evaluated P-selectin FI to determine how highly activated platelets were affected by ADP and TXA2 inhibition. P-selectin FI was significantly reduced by ADP and TXA2 inhibition. There was limited global and local contraction in P-selectin+ platelets across all conditions. Our results demonstrate that global platelet contraction is inhibited by ASA, 2-MeSAMP, and MRS-2179, while ASA has a more pronounced inhibitory effect on local platelet contraction. These results are significant in understanding how different platelet antagonists affect clot contraction and ultimately clot resolution. 

In an effort to move towards a more sustainable future, improvements are needed for a wide range of materials design problems. Cobalt oxide is a valuable material that can is used in various energy applications, such as, a cathode in batteries or a catalyst. Density Functional Theory is a computational tool that can be used to study material structures and properties, helping to accelerate the material design process. Doping cobalt oxide nanoparticles with other metals has been shown to improve catalytic ability (by Fe-doping) and battery capacity (by Al-doping). By using a scheme utilzing DFT calculations of well-defined experimental systems, we are able to illustrate how nanostructuring materials design can improve material performance. Our detailed study on Fe-doped CoOx nanoislands for the Oxygen Evolution Reaction (OER) demonstrates the importance of site-specific catalyst design. Fe has anisotropic doping patterns in CoOx and the role of Fe is driven by where it is located in the system. Similarly, Al-doping of LiCoO2 increases long term battery capacity with the distribution of Al effecting this improvement. This work elucidates the reason behind these improvements and provides insight in how to further improve cobalt oxide materials for energy applications. 

The advent of self-propelled colloids that move absent external force brings important and as yet largely untapped degrees of freedom to interfacial engineering. Bacteria and reactive Janus beads are examples of active colloids that propel themselves and interact with fluid boundaries. All such active colloids obey similar hydrodynamic descriptions, which we develop and explore.

Because of their directed motion, active colloids accumulate near interfaces. They can become trapped and swim adjacent to interfaces via hydrodynamic interactions, or they can adsorb directly and swim in an adhered state with complex trajectories that differ from those in bulk in both form and spatio-temporal implications.

In this research, we have observed a variety of behaviors for swimming bacteria, which we associate with the trapped orientation of the bacteria at the fluid interface. We image the displacement fields of passive colloidal tracers around the swimming bacteria over short lag times to approximate the hydrodynamic flow field, and we compare that flow field to predictions based on hydrodynamic multipoles for Stokes flow in a 2-D incompressible domain. The flow field of an interfacially trapped bacterium is well described by two hydrodynamic dipolar modes which are associated with interfacial incompressibility. With our understanding of the hydrodynamic pair interactions of passive colloids with active bacteria, we study how tracers are advected through hydrodynamic interactions and understand their enhanced diffusion.  

Additionally, we have identified new opportunities to study the collective motion of colloids as either driven colloids or active bacteria swim through a colloidal crystal of passive colloids stabilized by electrostatic repulsion at fluid interfaces. By understanding these implications, we will establish new design rules for active interfaces to enhance transport in multiphase systems.

The dynamics and structure of systems beyond the glass and jamming transition have been of interest for a long time due to their importance in a wide range of fields. Molecular glasses are essential active layers in organic light-emitting diodes, while the phenomena of jamming is important in dense emulsions and granular flows. The slow dynamics and complexity of theses systems has restricted the ability to simulate them using conventional methods, particularly at low temperatures or high volume fractions. The properties of the low energy states are of particular importance to answering many fundamental questions surrounding the glass transition and in connecting simulation results to experiments. In this study, we use an energy landscape based approach to probe the low-energy inherent structures and provide an algorithm to drive the system to lower energy states. The algorithm is inspired by metadynamics, and proceeds as a high dimensional basin filing algorithm that allows us to explore the energy landscape of a model soft-sphere foam system. We demonstrate the ability of the algorithm to reach low energies while revealing interesting geometric characteristics about the nature of the landscape. We study how connected metabasins are organized relative to a starting energy minimum and their orientations relative to each other. Finally, we compare our method with conventionally used methods like simulated annealing and swap Monte Carlo and discuss the advantages and disadvantages of each.

Complexation between polyelectrolytes (PEs) and nanoparticles (NPs) have been used to assemble functional films and membranes that respond to multiple inputs and stimuli. Interfacial complexation between oppositely charged polyelectrolytes and nanoparticles at the interface of aqueous two-phase systems has emerged as a powerful method to assemble these functional membranes. Remarkably, these membranes can grow continuously to thicknesses approaching 6 mm, naturally raising the question as to what is the underlying mechanism for the growth of these membranes. By taking advantage of a microfluidic setup, we study the growth mechanism behind interfacial complexation between silica (SiO2) nanoparticles and Polydiallyldimethylammonium chloride (PDADMAC). The sequential insertion of PDADMAC and fluorescently labeled PDADMAC has shown that the membranes grow through a permeation mechanism, in which PEs diffuse and permeate continuously to the growing front of the membrane. Taking advantage of this growth mechanism, we demonstrate that PDADMAC/SiO2 nanoparticle membranes can be spatially structured to embed specific functionality, which could potentially be useful for programmed delivery of actives from the membrane.

¹ Institute for Medicine and Engineering, ² Department of Chemical and Biomolecular Engineering University of Pennsylvania, 1024 Vagelos Research Laboratories, Philadelphia, PA 19104

Hemostatic thrombi formation in mouse models is driven by a core/shell hierarchy. This organization is
comprised of a highly activated, tightly packed core of P-selectin positive platelets surrounded by a less activated, loosely packed P-selectin negative shell. While this hierarchy has been observed in human adult blood previously, the stability and strength of the core and shell have not been assessed. Utilizing an 8-channel microfluidic assay, we interrogated thrombi through 2 different methods. For both methods, we perfused CTI treated whole blood (WB) ± agonists over a collagen/tissue factor (TF) surface at an initial wall shear rate of 100 s-1. In one method, we switched out the WB after 3 minutes and replaced it with buffer. We then increased the initial wall shear to 1000 s^-1 and monitored platelet erosion by calculating the change in normalized platelet fluorescent intensity (FI). In our second method, we paused WB perfusion after 7.5 minutes and measured clot % contraction over an additional 7.5 minutes. We calculated this by measuring the difference in clot lengths in the direction of blood flow. Utilizing the buffer exchange method, clot growth on collagen is immediately stopped when switching to buffer and increasing the shear rate resulted in a ~20% reduction in platelet FI. Treating blood with PPACK to block thrombin and therefore fibrin generation led to an increase in platelet erosion (~40% reduction) and a decrease in P-selectin expression compared to control blood. Gly-Pro-Arg-Pro (GPRP) blocks fibrin formation but still allows for thrombin to be produced and resulted in no change in platelet erosion while increasing P-selectin expression compared to the control. In both treated and untreated blood, there was no decrease in P-selectin FI during buffer perfusion and all three conditions were significantly different from each other (P < .0001). These data show that thrombin, but not fibrin, inhibition lowers stability of thrombi and suggest that fibrin is only present in the already stable core region. Shell stability was examined further by using secondary agonist inhibitors that target the ADP receptors P2Y1 and P2Y12 and thromboxane A2 (TxA2) production. These inhibitors resulted in a reduction in P-selectin expression and an increase in platelet erosion. Compared to ~40% reduction in platelet FI observed in the PPACK control, TxA2 inhibition resulted in ~60% reduction, and P2Y1 and P2Y12 antagonism resulted in ~55% and ~60% reduction, respectively. These data suggest that secondary agonists contribute to core formation and stabilize the shell region. Stopping the flow of WB with and without PPACK allowed for the examination of both overall clot and core contraction. Fibrin and thrombin inhibition resulted in an increase in both total (20.3% vs 8.9%) and core contraction (8.3% vs 2.5%). There was significantly less contraction observed in P-selectin positive platelets which suggests that the core region is more tightly packed and more resistant to contractile forces.

Kinases are a class of proteins that play important roles in cell signaling, differentiation, and proliferation. They are found to be frequently mutated in cancer and are the second-largest therapy targets of medicinal drugs in clinical research after GPCRs. The activation status of mutated kinases in cancer can profoundly impact phenotypic outcomes not limited to tumor progression and drug sensitivity. To quantify these phenotypic outcomes through mutated kinase activities, optical approaches to control cell signaling by photoswitchable kinases have been transformational by relying on certain gain-of-function mutations. To better understand this at the molecular level, the role of mutations in intrinsic kinase activity needs to be quantified.

We perform free energy calculations through enhanced sampling techniques to facilitate our understanding of the structural stabilization of MEK kinase and its mutated systems. Additionally, we quantify and rank the degree of alterations caused by the mutated systems to the kinase’s structure-activity relationship based on changed protein dynamics and the resulting proximity between wildtype and mutated kinase systems using our proposed technique of Boltzmann weighted correlations between the kinase conformations sampled on the free energy landscape. We also discuss the kinetics of the kinase transition.  Finally, we trade the computational complexity with improved data scalability to find structural fingerprints in kinase transition that predict the mutational effects with drastically reduced false positives.

Protocells are synthetic cells that mimic the size and organization of biological cells but can be built from novel components, including polymers. Unlike lipid-based protocells, which are fragile and have limited chemical functionality, polymer-based protocells are stable and can be imparted with advanced functionalities that enable greater control over their properties. A goal of our research is to make motile polymer-based protocells whose motion is driven by enzymatic reaction. The motive for using enzyme catalysis to drive particle motion is based on the demonstrated principle that enzymatic turnover of substrate to product generates a displacement force. Our prior work demonstrated that catalase-containing surface-adherent polymersomes made from poly(ethylene oxide)–b–poly(1,2 butadiene) display random autonomous motion in the presence of hydrogen peroxide. However, these experiments used heterogenous populations of protocells made from thin-film rehydration and resulted in nonuniform motion, even within a single batch of particles. To develop a more elegant platform for enzyme-driven motility, we prepare uniform protocells templated from water-in-oil-in-water (W/O/W) double emulsions droplets, allowing for precise control over the size of the particles as well as their composition. Unlike polymersomes prepared by microfluidic assembly, which are slow to template and adopt a partially dewetted configuration during the dewetting transition that limits their mechanical stability, microcapsules based on poly(lactic-co-glycolic acid) (PLGA) maintain a core-shell morphology during solvent extraction, leading to the formation of relatively uniform shells with high stability. We demonstrate the use of these capsules by attaching enzymes, including catalase, to the shell through a combination of carbodiimide-amine and biotin-avidin chemistry. We then examine their motion in solution in response to enzymatic turnover. Our work offers unique approaches for preparing highly uniform protocells that can be used to robustly screen differences between the thermodynamic and kinetics properties of enzymes and the effects such properties have on observed particle motion.

Side-chain liquid crystal (SCLC) diblock copolymers possess the characteristics of both block copolymers and liquid crystalline polymers. This combination makes them useful for membrane separation applications and more responsive to external stimuli. Currently, there is a lack of mesoscale models that effectively captures block copolymer self-assembly and ordering of liquid crystals. We have developed a mesoscale model that we are using to study order-disorder phase transitions of SCLC diblock copolymers. Similar to traditional block copolymer microphase separation, SCLC diblock copolymers can phase separate into lamellar and cylindrical phases with the liquid crystals also forming a nematically ordered phase. The coexistence of the diblock copolymer and liquid crystal phase transitions makes the overall phase behavior very complex. We also have studied the interference between the order to disorder transition of the block copolymer phases and the isotropic to nematic transition of the liquid crystal phases. A comparison of phase behaviors between SCLC diblock copolymers and normal diblock copolymers is made. Additionally, we characterize how the chain conformations change as we move through the phase transitions.

Time-of-flight secondary ion mass spectrometry (ToF-SIMS) has been widely used to study the surface and near surface chemistry properties of polymer materials. However, ion beam bombardment can lead to damage to the chemical structures of the surface and the complicated matrix effect for organic materials. Therefore, when using ToF-SIMS, it is necessary to newly determine the optimal conditions such as ion beam energy and current for different polymer materials. The system of interest is poly(methyl methacrylate) grafted silica nanoparticles, denoted as PMMA-NP, in a poly(styrene-co-acrylonitrile) matrix, in which the grafted nanoparticles should segregate to the free surface after annealing at 190 °C and lead to a surface excess. For this specific system, the optimal ToF-SIMS conditions include a Xe ion beam with an energy of 30 kV and current of 5 nA, rastering over an area of 100 μm x 100 μm. The charging effect is also minimized by selecting appropriate electron beam conditions (5 kV, 100 nA ~ 142 nA) for charge compensation, which increases the depth profile signals by over one order of magnitude, compared with no charge compensation. ToF-SIMS depth profiles of Si show surface peaks after annealing for a variety of times (up to 24 hours), indicating an enriched layer of PMMA-NPs after annealing. The calculated surface excess, Z*, by fitting the depth profiles increases with longer annealing times, confirming the hypothesis that as annealing time increases, more PMMA-NPs diffuse to the surface.

In order to perform crucial immunological functions, T lymphocytes must adhere and migrate along the endothelial surface during the leukocyte adhesion cascade. Strong adhesion and migration are mediated through integrin binding. Two binding pathways are of particular impact to T lymphocyte migration: Lymphocyte function associated antigen-1 (LFA-1) integrins binding Intercellular Adhesion Molecule-1 (ICAM-1) and very late antigen- 4 (VLA-1) binding vascular adhesion molecule-1 (VCAM-1). ICAM-1 and VCAM-1 may be expressed concurrently by endothelial cells. Previously, it has been studied that T cells migrating along ICAM-1 under shear flow will orient and travel upstream against the direction of flow. T cells on VCAM-1 only surface will migration downstream along the direction of flow. With any mixture of ICAM-1 and VCAM-1 presented on a surface, T cells will exhibit upstream migration. Furthermore, it has been shown that T cells migrating against the flow on ICAM-1 will transmigrate through the endothelial surface more quickly than cells migrating downstream on the same surfaces.

In this study we hypoethesize that we should be able to direct T cell migration through microcontact printing ICAM-1 and VCAM-1 surfaces in specific patterns to utilize upstream migration. We printed parallel stripe patterns of mixed ICAM-1 and VCAM-1 parallel to the direction of flow with widths of 50μm, 100μm, and 200μm. Cells experienced an unpredicted inhibition of upstream migratory behavior on the thinner stripes. We analyzed the behavior of individual cells including interactions with the ICAM-1/VCAM-1 edge boundary to determine the source of the variable behavior. Furthermore, we confirmed that the surface concentrations of ICAM-1 and VCAM-1 were comparable between the different striped surfaces and uniform surfaces from previous studies.

We investigated the diffusional limitations of thrombin and other coagulation proteins within fibrin gels, plasma clots, and whole blood clots formed in microfluidic devices. A fluorescence based recovery after photobleaching (FRAP) assay was used to determine several metrics of diffusion and recovery of thrombin, albumin, and fibrin(ogen) within clots. The simplified diffusion coefficient, Dc, will provide a metric for comparison between our experimental conditions to determine the relative differences in diffusion of these fluorescently labeled proteins. The extent of thrombin binding to fibrin fibers was also investigated by centrifugation of fibrin gels of various qualities. Thrombin concentration in the supernatant of gels formed under isotropic conditions was measured by the rate of cleavage of a thrombin sensitive peptide.

It is known that an enhancer regulates transcriptional activity of its gene through activators and repressors by establishing the spatiotemporal expression of a target gene. Since enhancers may contain multiple binding sites for the same transcription factors (TFs), it is not yet understood how each binding site affects the overall transcriptional competency of the enhancer. In this study, we have implemented the MS2-MCP live imaging technique to quantitatively analyze dynamic transcriptional activity of a region of the snail distal enhancer. Through systematically modulating 3 Dorsal (Dl) and 1 Twist (Twi) binding domains, we found that the mutations in these binding sites caused a drastic reduction in the maximum intensity, resulting in a reduction in total mRNA production. While the timing of activation was not substantially affected in the mutated constructs, the bursting dynamics and nuclei activation were altered to varying degrees. We characterized the synergistic capabilities of each binding site by quantifying the contribution of each binding site to the total transcriptional dynamics. Using mathematical modeling, we determined that cooperative action between TFs was necessary to maintain normal expression levels. Through this study, we identified distinct mechanisms by which TFs regulate proper gene expression during development.

Micro zinc-air batteries (micro-ZAB) as a promising power source for miniature devices attract great attention due to their high energy density, biodegradability, and safety. Operating ZAB under a lean electrolyte regime is favorable for a higher energy density because electrolyte volume is minimally packed. However, lean electrolyte is more susceptible to unfavorable side reactions than bulk electrolyte. In particular, hydroxide ion (OH^–) consumption significantly hinders the electrochemical performance of the micro-ZAB. The mechanisms of side reactions are studied through titrations and electrochemical impedance spectroscopy (EIS). The experimental results indicate that both carbonation and zincate accumulation contribute to the consumption of the OH^– in the alkaline electrolyte and cause poor electrochemical performance. The average consumption rate of the OH^– due to the carbonation is measured to be 5.2210^-7 mol min^-1 cm^-2. The electrochemical performance of the micro-ZAB is predicted from consumption rates of the OH^–, and experimental data are in good agreement with the prediction.

Machine learning (ML) has emerged as a powerful tool in many different fields for predicting statistical outcomes both accurately and efficiently – and the catalysis field is no different. Broadly, ML has been used to automatically create captions for images [1], track objects in a live-feed video [2] and detect heart arrythmias from electrocardiograms [3].  Within catalysis, ML has been used to determine new density functional theory (DFT) exchange-correlation functions [4], develop convolutional neural networks based on the local atomic coordination [5], and much more. However, its usefulness as a tool for materials discovery and understanding in catalysis has not been leveraged to its fullest extent. To this end, a convolutional neural network has shown promise in using DFT-determined electronic features for catalytic activity prediction. We hypothesize that the electronic structure of a material, outlined in its density of states (DOS), contains important clues that indicate its catalytic performance. The model, called DOSnet, uses the DOS of unary and binary metals to make predictions on the adsorption energy of important reaction intermediates [6]. By including 11 adsorbates and 2000 unique metal alloys, the dataset was comprised of over 37,000 datapoints. In our work, we further the scope by introducing the DOS of more complex materials, namely ternary metals, carbides, sulfides, and MXenes, among others, with the goal that this model can identify which features of the DOS are important in determining catalytic activity. The model can successfully group together materials that perform similarly, as well as determine a principal component that correlates with the adsorption energy. This result will help identify if there is a feature (or a combination of features) within a material’s electronic structure that is critical in determining its catalytic activity, and that it could potentially be identified through future feature analysis. This work helps advance materials design by elucidating understanding about the electronic structure’s role in determining catalytic activity and further the development of important catalytic descriptors for novel materials design.  

[1] LeCun,Y., Bengio,Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436.

[2] He, K., Gkioxari, G., Dollár, P., & Girshick, R. (2017, October). Mask r-cnn. 2017 IEEE International Conference on Computer Vision (ICCV).

[3] Rajpurkar, P., Hannun,A.Y., Haghpanahi, M., Bourn, C., & Ng,A.Y. (2017). Cardiologist-level arrhythmia detection with convolutional neural networks. arXiv preprint arXiv:1707.01836

[4] Lei, X.; Medford, A. J. Design and analysis of machine learning exchange-correlation functionals via rotationally invariant convolutional descriptors. 063801, 1–18 (2019).

[5] Back, S.; Yoon, J.; Tian, N.; Zhong, W.; Tran, K.; Ulissi, Z. W. Convolutional Neural Network of Atomic Surface Structures to Predict Binding Energies for High-Throughput Screening of Catalysts. Journal of Physical Chemistry Letters, 10, 4401–4408 (2019). 

[6] Fung, V., Hu, G., Ganesh, P. et al. Machine learned features from density of states for accurate adsorption energy prediction. Nat Commun 12, 88 (2021). 

As water scarcity increases in society at large, water purification by filtering undesired species such as perfluoroalkyl substances, salts, and microplastics becomes a larger concern. Prior work by the Osuji group highlighted the utility of crosslinked self-assembled lyotropic mesophases of 2-(methacryloyloxy)ethyl tetradecyl dimethyl ammonium bromide (METDAB) for filtering solutes as small as 1 nm from aqueous solutions. In this work, we investigate related mesophases for controlling diffusive ion transport, in addition to exploring how altering crosslinkers and continuous phase composition can impact filtration properties. Unidirectional fluid flow cells are used to investigate the size rejection of self-assembled membranes, while electrochemical impedance spectroscopy is used to investigate the ion transfer properties of the self-assembled mesophases. New molecular species and their use as platforms for developing other self-assembled structures (i.e. gyroid phases) are discussed. Results show the promising potential for using self-assembled thin-film membranes in both size rejection and ion conductivity.

Global climate goals call for the aggressive removal of carbon dioxide from the atmosphere: 10 gigatons per year by 2050 and 20 gigatons per year thereafter. Carbon removal technologies require a durable, economically viable solution for carbon storage. Atmospheric carbon dioxide is naturally stored, via rock weathering, in the form of carbonate minerals. This process takes place on a geologic time scale; enhanced weathering research focuses on facilitating faster time scales to mitigate climate change. One pathway uses industrial mine wastes for carbon storage through the carbonation reaction. Most industrial mine wastes are unreactive with ambient carbon dioxide and require pretreatment for carbon storage; the pretreatment method is dependent on the specific mineralogy of a material. This project investigated process parameters (temperature, reaction time, and ratio of reagents) for a thermal cation extraction process on mine wastes from a Platinum Group Elements mine in Nye, Montana. Process variables were compared on the basis of extraction efficiencies and related back to the mineralogy of the material. Results from this study reveal the carbon storage potential of the thermal cation extraction process. Our findings offer insights into tuning enhanced weathering processes for industrial wastes as potential feedstocks to meet global climate needs.

During plaque rupture that leads to a heart attack, flowing blood is exposed to a highly procoagulant surface containing collagen and tissue factor (TF), causing platelets to become activated. Activated platelets release soluble agonists such as ADP and thromboxane, causing further activation and platelet recruitment, resulting in thrombus growth that distorts blood flow. Computational modeling of thrombus growth under flow is important to evaluate thrombotic risk under different pharmacological and hemodynamical conditions. In this work, we have developed a fully spatially resolved 3D multiscale framework for patient-specific prediction of thrombus growth under flow on a defined surface containing collagen and TF. The multiscale framework is composed of four modules: a neural network that accounts for platelet signaling, a lattice kinetic Monte Carlo solver for tracking platelet positions, a finite volume method solver for convection-diffusion-reaction equations for agonist release and transport, and a lattice Boltzmann method solver for computing the blood flow field over the growing thrombus. We validated our 3D model by comparing model predictions to in vitro experiments of whole blood perfusion in an experimental microfluidics assay. The model was able to accurately capture the evolution and morphology of the growing thrombus; model predictions agreed well with fluorescent micrographs of thrombus formation over time observed in experiments. The generalizability of the 3D multiscale solver allowed the simulations of important clinical situations such as cylindrical blood vessels and acute flow narrowings (stenosis) linked to heart attacks. Moreover, the model can accurately predict thrombus dynamics under different pharmacological conditions corresponding to drug treatments, such as aspirin for COX-1 inhibition, P2Y1 inhibitor MRS-2179, and iloprost, a prostacyclin analog; and under blood disorders such as von Willebrand disease.

Infiltration of polymer into the interstices of dense nanoparticle packings leads to the formation of
highly loaded nanocomposites with superb mechanical and transport properties¹. Capillary rise
infiltration (CaRI) has emerged as a powerful method to prepare highly loaded composites with a wide
range of polymers and nanoparticles². The CaRI system is unique in that regimes of confinement in
which the polymer radius of gyration is up to an order of magnitude larger than the nanoparticle packing
pore radius can be easily reached. Here, we present a computational study using self-consistent field
theory (SCFT) to understand the thermodynamics of these highly confined composites containing two
different polymers. We investigate the effects of confinement and polymer-nanoparticle interactions on
the miscibility between the two polymers. Two polymers that would undergo macroscopic phase
separation become miscible when they are subjected to extreme nanoconfinement. In addition, when a
strong repulsive interaction exists between one of the polymers and the nanoparticles, the polymers
become miscible over a much larger region. We will discuss the possible thermodynamic origin of such
miscibility as a function of polymer-particle and polymer-polymer interactions, and the confinement
ratio. The ability to create miscible polymer nanocomposites out of normally incompatible blends would
unlock a myriad of novel properties and applications in the future.
References
1. Harito C, Bavykin DV, Yuliarto B, Dipojono HK, Walsh FC. 2019. Polymer nanocomposites
having a high filler content: synthesis, structures, properties, and applications. Nanoscale
11(11):4653–82.
2. Huang, Y. R. et al. Polymer nanocomposite films with extremely high nanoparticle loadings via
capillary rise infiltration (CaRI). Nanoscale 7, 798–805 (2015).

Electrowetting (EW) and electrowetting-on-dielectric (EWOD) are two traditional methods for droplet wetting state modulation via external physical stimuli. However, EW and EWOD require direct droplet contact with an electrode, which may be challenging and undesirable when dealing with electrically sensitive cargo in the droplet or with microscale droplets. Although contactless methods that apply external physical stimuli to alter surface wetting properties have been reported, the range of contact angle modulation remains limited. Here we demonstrate a contactless method to induce reversible droplet contact angle modulation on chemically inert substrates via corona discharge physics, called dielectric charge injection (DCI). The method involves a sharp, conductive probe that can induce dielectric breakdown of the surrounding dielectric medium, such as air, under voltages exceeding the medium’s dielectric strength. Breakdown leads to ionization of the dielectric, which then accelerates away from the sharp tip due to electrostatic repulsion, resulting in charge injection onto a target surface. With DCI, we induce wetting of a pure water droplet on non-wetting, non-contacting surfaces in hexadecane, a non-polar ambient phase. DCI can induce wetting transition of a non-wetting droplet separated from the surface by a nanoscale ambient phase film. Upon inducing wetting, DCI can achieve up to 140° contact angle modulation–competitive or even exceeding the capabilities of traditional EW and EWOD. Furthermore, upon removal of the voltage and/or probe, the droplet undergoes dewetting and returns completely to the initial non-wetting state. We show that DCI can be used to induce deposition of encapsulated materials from droplets to the non-wetting surface. DCI can also be applied for recovery of materials from such a surface. DCI presents a unique strategy for contactlessly and reversibly manipulate droplet wetting that is simple and powerful, with a wide application space that remains to be explored, especially in contexts where EW and EWOD become inapplicable.

This paper presents improvements upon methods that explore rare process trajectories leading to rare safety and reliability events.  It applies forward-flux sampling (FFS) from the family of sampling algorithms developed to discover rare molecular dynamics pathways.  For a relatively simple, dynamic exothermic CSTR model with noisy feed concentration, it shows how to apply the FFS algorithm to simulate and analyze rare trajectories between high- and low-conversion steady states.  First, it compares results with a less efficient brute-force (BF) method, and then with a transition-path sampling (TPS) method applied in simulation studies for an exothermic CSTR. The effects of varying key process parameters, i.e., the residence time, τ, noise variance, 2 , and the controller gain, KC, which impact the rareness of an event, are investigated. Rates of rare-path transition between high- and low-conversion steady states, forward and backward, are shown to exhibit equilibrium ratios independent of 2, with the forward rates decreasing with τ and KC, and the backward rates increasing with KC, whereas both increase with 2. These unanticipated relationships should lead to improved alarm notifications being developed in this research work.

Keywords: Forward-flux Sampling, Molecular Dynamics, Residence time, Noise variance, Controller Gain