1. Precision A3G Base Editors for Disease Modeling and Correction

Tyler Daniel, Xue Gao

Cytosine base editors (CBEs) are advanced tools that enable efficient cytosine-to-thymine (C-to-T) substitutions at specific genomic sites without making double-stranded breaks. A particular variant, A3G-BE, is useful for preferentially editing the second cytosine in a 5’-CC-3’ motif, thereby reducing unintended “bystander” edits nearby. However, this tool faces two major challenges: it often edits multiple CC motifs within its operational window when only one is desired, and its use is restricted by the stringent protospacer adjacent motif (PAM) requirements of the standard SpCas9. To overcome these limitations, we engineered improved A3G-BE variants by optimizing their linkers, applying rational mutagenesis, and integrating them with Cas9 variants (SpG and SpRY) that have more relaxed PAM constraints. These enhancements achieve greater precision by targeting a single cytosine within CC motifs and significantly broaden the targeting scope to previously inaccessible genomic locations. We validated these new editors by precisely installing and correcting mutations that cause cystic fibrosis (CF) in HEK293T cells. By applying them to 16HBE14o- human bronchial epithelial cells, we successfully modulated CFTR mRNA levels, protein expression, and channel function. These precision A3G-BE variants therefore represent a significant advancement for modeling and potentially treating CF and other human diseases.

2. Measuring the Traction Forces of Upstream-migrating Hematopoietic-like KG1a Cells under Shear Flow

Dong-hun Lee and Daniel A. Hammer

Upstream migration is a distinct mode of leukocyte motility in which cells move against the direction of fluid flow on surfaces with Intercellular Adhesion Molecule-1 (ICAM-1), mediated by the integrin Lymphocyte Function-Associated Antigen-1 (LFA-1). This behavior has been observed across multiple immune cell types including T cells, hematopoietic stem and progenitor cells, and neutrophils. Despite its prevalence in immune cell trafficking, the mechanical forces that drive upstream migration have remained uncharacterized. Here, we apply Traction Force Microscopy (TFM) to quantify the spatiotemporal dynamics of force generation of KG1a cells migrating upstream on ICAM-1 functionalized polyacrylamide hydrogels under physiologically relevant shear flow.

We find that upstream migration is supported by a marked increase in traction force generation relative to migration in static conditions, with peak forces nearly doubling under flow. This enhancement in force output was consistently observed across the upstream-migrating cell population. Despite this increase in magnitude, it is not accompanied by a detectable anterior redistribution of traction: the spatial pattern of force application remains similar to that observed in static conditions. These findings indicate that cells adapt to shear flow primarily by increasing how much force they generate, rather than by changing how that force is spatially deployed.

By providing the first direct measurement of traction forces during upstream migration, this work establishes a quantitative reference point for the mechanics of this behavior. The finding that upstream migration relies on force amplification without large-scale reorganization of traction distribution further clarifies the nature of this response, and sets the stage for dissecting the signaling pathways that enable cells to sense and respond to mechanical stimuli during immune cell trafficking.

3. Quantitative Analysis of the Impact of Local Chromatin Organization on Trascriptional Dynamics

Noel Buitrago, Bomyi Lim

Transcription, the conversion of DNA to RNA, is a precisely controlled process that is essential for proper development and maintenance of an organism. Gene promoters and their regulatory enhancers must interact in spatial proximity to initiate transcription. However, in the eukaryotic genome, enhancers and their target promoters are often separated by large linear genomic distances containing numerous non-target genes, calling into question how these elements reliably find each other in the crowded nuclear environment. Insulator elements serve a prominent role in this long-range enhancer-promoter communication by organizing local chromatin into loop-like domains, facilitating interaction within the loop and silencing interaction with elements outside of it. Despite their importance, systematic analysis on the properties of insulators and their effect on transcription is lacking. This study employs single-cell live imaging techniques to elucidate the dynamics of insulator-mediated chromatin looping and its role in transcriptional regulation. Our system consists of a single enhancer that regulates two equidistant promoters in cis. We show that a single enhancer can co-activate two target promoters in cis, such that transcriptional kinetics from the two reporter genes are coordinated. Insulators are then introduced in various positions and orientations to analyze their effect on transcription. Insulator-mediated looping of the system further synchronizes gene activity and increases transcriptional output. Interestingly, bidirectional insulator conformations produce a more pronounced effect than unidirectional ones, indicating an orientation-dependent stability in insulator pairing. Insertion of an insulator between enhancer and promoter results in a sharp decrease in transcriptional output of the blocked reporter gene. Surprisingly, the blocked promoter can escape silencing and often exhibits coordinated transcriptional activity with the unblocked promoter. These results challenge the traditional view that insulator-mediated chromatin loops are largely static and suggest that the dynamic nature of insulator pairing contributes to transcriptional regulation to varying degrees, based upon their relative positions and orientations. Taken together, this work enhances our understanding of the relationship between local 3D chromatin architecture and transcription, providing new insights into the complex mechanisms governing gene expression.

4. Domain Alignment and Solvent Swelling Impact Ion Transport in a Multiblock Copolymer

Benjamin T. Ferko, Viola A. Burlein, Stefan Mecking, Karen I. Winey

The incorporation of polar solvents into ion-containing polymers is known to enhance the ionic conductivity towards application-relevant values. We previously investigated the incorporation of ethylene carbonate into bulk, isotropic samples of a multiblock copolymer consisting of alkyl blocks of precisely 12 carbons alternating with polar blocks containing lithium sulfonate groups (PESLi12). The ionic conductivity was enhanced by several orders of magnitude upon swelling. Another approach for improving the ionic conductivity of nanostructured polymers is alignment which eliminates grain boundaries, mixed orientations, and defects. This study investigates the effect of solvent swelling on aligned thin films of PESLi12 using  in situ grazing incidence X-ray scattering and broadband dielectric spectroscopy with interdigitated electrodes. We demonstrate that the layered ionic assembles remain parallel to the substrate upon swelling with propylene carbonate, γ-butyrolactone, dimethyl carbonate, or diglyme. The in-plane ionic conductivity is significantly increased upon solvent incorporation, but a swollen, isotropic sample swollen with propylene carbonate displays 1.5 orders of magnitude greater ionic conductivity than the equivalent aligned thin film. These findings indicate that grain boundaries and defects in these alternating multiblock copolymers are preferentially swollen with solvent, providing pathways for rapid ion transport and suggesting routes for further improvement of ionic conductivity in solvent-swollen nanostructured polymer electrolytes.

5. Enhanced Allam Cycle Through High-Temperature Conversion of Captured CO₂ to CO

Aryaman Shah, Warren Seider, John O’Connell

Traditional fossil fuel-based industries have long been troubled by their inherently carbon-intensive nature. Decarbonizing such sectors has been a persistent global challenge, driving the development of closed-loop energy systems that integrate power generation with carbon utilization. Among these, the Allam Cycle stands out as a high-efficiency, oxy-fuel, supercritical CO₂ power cycle capable of achieving near-zero-emission electricity generation. Yet, its full potential remains underexploited, as the captured CO₂ is conventionally directed toward long-term storage rather than productive reuse. This study introduces a novel high-temperature thermochemical pathway that integrates seamlessly with the Allam Cycle to convert captured CO₂ by-products into carbon monoxide (CO), effectively transforming the system from carbon-sequestering to a carbon-utilizing chemical-generation process. This innovation not only addresses the decarbonization imperative, but also creates an additional revenue stream through the generation of value-added chemicals and fuels such as methanol, higher alcohols, and syngas derivatives. Moreover, this novel process seeks to decarbonize emissions-heavy industries thus, allowing the U.S. to meet its power demands in a clean, sustainable manner.

6. pH-dependent Interaction between Anionic Silica Nanoparticle and Rare Earth Element (REE) for Selective REE Separation

Rachel Thatcher, Aleksandra Vojvodic

Surface structure plays a critical role in determining the performance of metal oxides in real world applications. However, typical computational studies frequently rely on idealized, bulk-cleaved “perfect” surfaces, neglecting the structural complexity of real-world conditions. By introducing simple surface imperfection such as adsorbates, defects, and vacancies, we demonstrate that these non-ideal terminations are often more stable than their perfect counterparts under thermal and electrochemical conditions. Our findings challenge the conventional assumption that perfect surfaces dominate in reactive environments and highlight the critical influence of realistic terminations on local chemical environments. This study provides a framework for improving surface models in simulations and offers experimentalists insights for more accurate interpretation of surface-sensitive measurements.

7. The Impact of Direct Hexagonal Mesophase Geometry and Hydration on Ion Motion

Ying-Shuo Peng and Talid Sinno

Crystal nucleation is fundamental to the study of self-assembling colloidal systems. Although molecular simulations have been extremely useful for unraveling some of the mechanistic details of colloidal crystal nucleation, a longstanding issue is that simulations consistently underpredict nucleation rates in weakly supersaturated hard-sphere systems, sometimes by orders-of-magnitude. This is true for both direct simulations of crystal nucleation and enhanced sampling approaches such as forward flux sampling. Recently, hydrodynamic correlations between particles due to the solvent have drawn attention as a potential explanation for this mystery. Yet, computational studies that include hydrodynamic interactions (HIs) have mostly shown relatively small effects that arise due to the impact of HIs on single particle attachment kinetics on an already-formed critical nucleus. 

Here, we consider the impact of HIs on density fluctuations in a metastable colloidal fluid prior to the formation of a critical cluster as a potential explanation for the discrepancy between simulation and experiment. Specifically, we employ the total intermediate scattering function, F(k,t), which is related to the autocorrelation function of density fluctuations in reciprocal (𝑘) space, to study the impact of HIs on crystal nucleation. Brownian Dynamics (BD) simulations are used to compute intermediate scattering function for systems without HIs, while Multiparticle Collision Dynamics (MPCD) simulations are used to capture the effect of short and long-range HIs. Our results reveal that in MPCD, density fluctuations decay much more slowly than in BD, indicating that HIs significantly prolong the relaxation time of density fields in the metastable colloidal fluid. This observation leads to a key insight: the enhanced lifetime of pre-critical density fluctuations due to HIs may be crucial for forming large nuclei, especially under weak supersaturation conditions. Moreover, we find that the density-field-stabilizing effect from HIs becomes even more pronounced at larger length scales (lower wavevectors), higher volume fractions, and weaker interparticle interactions—all of which are characteristic of the weakly supersaturated hard-sphere regime where simulation-experiment discrepancies are most severe.

To quantify the impact of extended relaxation times on nucleation dynamics, we develop a toy model to estimate the enhancement in nucleation rate due to HIs. The inputs to the model are (1) the length scale of the density fluctuation relevant for the nucleation event, which we obtain from seeding simulations, and (2) the lifetime of the fluctuation required for the nucleation event to occur, which is estimated from observations of clusters formed in direct simulations. The model suggests a simple mechanism by which HIs can lead to large impacts on nucleation rates observed in simulation. 

8. How do Antifreeze Proteins Inhibit Ice Recrystallization?

Jeongmoon Choi, Yusheng Cai, Amish J. Patel

Recrystallization is a phenomenon in which larger ice crystals grow at the expense of smaller ones. Both theory and experiments have shown that during recrystallization, the cube of the average crystal size grows linearly with time, r̅³ = kt. However, in the presence of antifreeze proteins (AFPs), experiments have shown that ice recrystallization can be retarded and even arrested altogether above a critical AFP concentration, c^∗. In this study, we propose a theoretical model explaining how AFPs suppress ice recrystallization and how their molecular properties influence c^∗. We also interrogate how experimental parameters, such as sucrose/saltconcentration, temperature, and the initial crystal size distribution, influence c^∗. Our findings offer insights into the relationship between AFP properties and c^∗, which could guide the designof novel materials for ice recrystallization inhibition.

9. Machine Learning Accelerated Radiative Transfer Modeling in CFD Fire Simulations

Mitchell Daneker, Lu Lu, Xiaoyi Lu

Radiative heat transfer is the dominant heat transfer mode in many fire scenarios, yet its solution procedure is often a major performance bottleneck of Computational Fluid Dynamics (CFD) simulations. Conventional solvers for the governing radiative transfer equation (RTE) are computationally expensive and can suffer from numerical artifacts. This paper presents a machine learning-accelerated RTE solver based on Deep Operator Networks (DeepONet) to address these limitations. The DeepONet is trained as a neural surrogate to learn the underlying solution operator of the RTE, directly mapping input functions, such as absorption coefficient and temperature, to the output radiative intensity solution. This pre-trained DeepONet model is then integrated into the open source CFD code FireFOAM, replacing the baseline finite-volume discrete ordinate method (fvDOM) solver. The performance of this hybrid ML-CFD framework is evaluated in transient simulations of canonical fire scenarios, including two-dimensional (2D) and three- dimensional (3D) pool fires and a 3D wall fire. Results demonstrate that the DeepONet solver reproduces the radiation solutions with significant computational speed-up. Moreover, when using the DeepONet solver at high resolution, the ray effect is mitigated.

10. Autonomous Motion of Enzymatically Driven Janus Micromotors

Amanda Hopkins, Daeyeon Lee, Daniel A. Hammer

Cellular motion is a dynamic process that plays a critical role in immunology, tissue assembly, and physiological homeostasis. Cell-like motility of artificial cells, or protocells, has the potential to enhance biologist’s understanding of complex cellular processes and to develop new technologies related to drug delivery, biochemical sensing, and cell-to-cell signaling. Understanding the fundamental requirements for protocell motion would provide insights into cellular motion and lead to the development of new technologies for building responsive protocells that could sense and respond to environmental stimuli. In this work, we use the notion that enzymes, when converting substrates to products, can exert forces. Our group recently used this idea to illustrate the enzyme-driven motion of cell-sized, urease-functionalized poly(lactic-co-glycolic acid) (PLGA) microcapsules. These capsules were intentionally developed with a non-uniform enzyme distribution on their shell as enzyme asymmetry has been hypothesized to be the driving force for enhanced motion. In our current work, we have made asymmetric Janus microparticles using microfluidic assembly to test whether asymmetric particles can display enhanced motion. We fabricated Janus microparticles using a mixed polymer system, blending the polymers PLGA and polycaprolactone (PCL). These two polymers partially phase separate during dewetting. Protein conjugation via 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) targets carboxylic acid functional groups which are only present on one lobe of the Janus particle, creating an inherent asymmetry in both shape and enzyme functionalization. This asymmetry can propel Janus particles functionalized with urease. Active motion of these particles provides direct measurements of the relationship between direction of motion and the axis of asymmetry of the Janus particles. We are exploring the relationship between substrate concentration and degree of geometric asymmetry on the avidity of motion.

11. Surface modification of microfluidics for scalable manufacture of anisotropic microparticles

Siddharth Sharma, Daeyeon Lee, Dave Issadore, Gijung Kim, Gabriel Rodriguez-Rivera

Microfluidic technologies offer exceptional control over the fabrication of microparticles, enabling precise tuning of size, shape, and composition. This capability has unlocked new possibilities in fields ranging from drug delivery to materials science. However, the scalable production of complex, anisotropic hydrogel microparticles remains challenging due to clogging caused by increased flow resistance during UV polymerization. This is especially a challenge when using Si/glass parallelized high-throughput material generators. 

To address this, we introduce a surface modification strategy using covalently bonded, liquid-like perfluoropolyether (PFPE) polymer coatings on microchannel walls. These coatings act as lubricating layers that reduce interfacial friction and enable the smooth transport of high-aspect-ratio hydrogels through microfluidic channels. We quantitatively and qualitatively compare fluidic resistance in coated versus uncoated devices, demonstrating that uncoated channels experience a drastic rise in resistance—often leading to clogging—as particle aspect ratios increase. In contrast, PFPE-coated devices maintain comparatively consistent flow and particle transport, even under demanding conditions. To quantify this behavior, we assess the impact of surface coating on generated particle size distribution and transport dynamics by measuring particle velocities and their changes over time to confirm the effect of PFPE on the particle generator walls. 

This work presents a scalable and effective solution for high-throughput microfluidic production of anisotropic hydrogels, removing a key bottleneck in the field and enabling broader application of complex microparticles in research and industry.

12. Diffusion of Bottlebrush Polymer Grafted Nanoparticles: The Role of Matrix Molecular Weight

Ben Indeglia, Karen I. Winey, Jensen N. Sevening, Robert J. Hickey

Nanoparticle diffusion is a valuable heuristic for structure-property relations in polymer nanocomposites. Nanoparticle diffusion in a polymer matrix is known to be sensitive to surface chemistry and size of the nanoparticles, and previous studies have focused on bare nanoparticles in athermal and attractive melts, or nanoparticles with linear grafted chain architecture. This investigation explores the effect of non-linear polymer grafted nanoparticles (PGNs) on nanoparticle diffusion. We report the diffusion coefficients of nanoparticles grafted with bottlebrush polymers in melts of linear polymers. These bottlebrush polymer grafted nanoparticles (PGNs) are comprised of large silica NPs (radius 80 nm) grafted with various backbone degrees of polymerization and fixed polystyrene side chains (4.2 kDa) to give bottlebrush molecular weights of 80 to 388 kDa. The polymer matrix is polystyrene (100 or 425 kDa), and nanoparticle diffusion was studied at 170 °C. After annealing a trilayer sample, time-of-flight secondary ion mass spectroscopy (ToF-SIMS) measured the cross-sectional PGN concentration profiles and extracted diffusion coefficients. The PGNs with high bottlebrush molecular weights relative to the matrix polymer display core-shell diffusion behavior, wherein the shell thickness is comparable to the grafted bottlebrush thickness. Surprisingly, when the bottlebrush molecular weight is smaller than the matrix polymer, the PGNs diffuse 10 – 100 times faster than predicted, which we attribute to a local decrease in viscosity associated with disentanglement of the matrix polymer near the large nanoparticles.

13. Room-Temperature Preservation of mRNA using Deep Eutectic Solvent

Hoang Dinh, Daeyeon Lee, Kathleen Stebe, Sang Yup Lee, Michael Kegel, Drew Weissman, Jilian R. Melamed

Messenger RNA (mRNA) has rapidly emerged as a transformative platform for vaccines and therapeutics, but its intrinsic instability in aqueous media and reliance on cold-chain storage remain critical barriers to global deployment. To overcome these challenges, we explore non-aqueous solvents as protective media for mRNA. Our work introduces a hydrophobic deep eutectic solvent (hDES) system that enables eƯicient partitioning of mRNA away from degradative aqueous environments while maintaining integrity over extended periods at room temperature. This simple, metal-free medium shields mRNA from enzymatic and hydrolytic degradation and is compatible with clinically relevant transcript lengths and modifications. Beyond stabilization, the hDES platform offers opportunities as a shelf-stable precursor for next-generation RNA formulations, including emulsions and lipid nanoparticles. These findings suggest new directions for creating shelf-stable RNA therapeutics that could broaden access to vaccines and gene therapies.

1. A New STEM Teaching Practice Program: Training the Engineering Faculty of Tomorrow

Zeenat Bashir, Sean Holleran, Lorena Grundy

Training of today’s Science, Technology, Engineering, and Mathematics (STEM) graduate students demands more than technical specialization; it requires additional skills essential for communication, collaboration and teaching. While STEM graduate curricula are best at developing disciplinary expertise, very few programs offer structured opportunities to support the pedagogical and interpersonal development required for long-term success in academia or industry. As a result, many graduates enter professional roles underprepared for responsibilities that demand both subject mastery and human-centered teaching and mentoring. This work-in-progress paper describes the development and implementation of a STEM Teaching Pedagogy Apprenticeship Program in the Department of Chemical and Biomolecular Engineering (CBE) at the University of Pennsylvania.

The CBE Teaching Pedagogy Apprenticeship was developed as a one-year structured experience that connects teaching theory with practice. The program pairs graduate students interested in academic or teaching-focused careers with faculty mentors and integrates two components: a Pedagogical Seminar and an Apprenticeship Experience. The semester-long discussion-based pedagogical seminar explores key domains of effective teaching, such as curriculum design and pedagogy, assessment and feedback, student engagement and development, equity and access in learning, technology-enhanced instruction, and educational research and evaluation. The apprenticeship experience is a guided face-to-face teaching practicum where participants assist in undergraduate/graduate courses, gradually taking responsibility for lesson design, classroom facilitation, and student feedback under faculty supervision. Together, these experiences help participants connect theoretical principles of learning with practical application. The program encourages iterative reflection through teaching journals and mentor feedback sessions, helping students internalize effective teaching practices and recognize how pedagogy intersects with scientific communication. In this paper, co-authored by faculty leaders and students in the initial pilot cohort, we conductive qualitative analysis focused on identifying common themes in participants’ evolving teaching philosophies and practices. Data include written reflections, seminar discussions, and end-of-semester surveys assessing participants’ perceptions of teaching development, confidence, and pedagogical growth. Overall, the program effectively advances participants’ pedagogical skills and professional identity as educators, demonstrating the impact of mentorship and reflective practice in graduate teaching development. We share our findings to serve as a model for others working to advance these skills for PhD students in their engineering departments.

2. Development and Validation of HIV Infectious Molecular Clones with Diverse Neutralization Profiles

Jiajie Li, Katharine Bar, Ryan Krause, Himanshu Garg, Jake Robinson, Noah Beratan, Hanah Schrader, Rebecca Lynch

Human immunodeficiency virus (HIV) remains a major global health challenge, and  broadly neutralizing antibodies (bnAbs) show great promise in combating infection.  bnAbs bind to the HIV envelope (Env) glycoprotein on the surface of virions and impede  viral entry into target cells. In our previous study, Panels of relevant Envs and bnAbs,  which were identified from our and published human clinical trials, were screened and  resulted in the selection of seven Envs representing a spectrum of sensitivity and  resistance to a bnAb pair (8ANC195 and N49). 

To characterize respective neutralization sensitivities and replication fitness of seven  Envs, this project focuses on engineering Env-chimeric infectious molecular clones  (IMCs) of replication-competent viruses to serve as research accessible resources. The  well-characterized HIV-1 NL4-3 plasmid was selected as the HIV backbone. Through In-Fusion Assembly, the env region was seamlessly replaced between the EcoRI site  (within vpr, creating overhang across pol-vif-vpr) and the XhoI site (within nef, creating  overhang across nef-3’LTR). This design prevents frame shift disruptions in overlapping  genes and regulatory elements, preserving full replication competence. IMCs’ functionality was validated by infectivity titration through TZM-bl cells first, in which all  IMCs were successfully infectious. Second, we validated replication fitness by in vitro infection of human primary CD4⁺ T cells, and through AlphaLISA detection of p24 antigen (capsid protein), we confirmed robust viral replication in host HIV target cells.  Last, neutralization assays comparing HEK293T-derived IMCs (original production) and  CD4⁺ T cell-passaged IMCs (natural infection) confirmed that Env-specific neutralization  phenotypes from selection criteria were maintained in IMCs. Replication experiments in  primary human CD4+ T cells and humanized mice to test relative replication fitness are  ongoing. This study demonstrates the utility of a scalable, reliable molecular platform for  generating replication-competent HIV clones. Engineered HIV IMCs with unique Envs  that maintain sensitivity/resistance profiles to clinically used bnAbs provide an essential  tool for downstream in vitro and in vivo studies to test combinatorial approaches to bnAb strategies and to highlight synergistic methods for managing HIV.

3. pH-Responsive Adsorption from Mixed-Diameter Gold Nanoparticle Solutions onto Weak Polyelectrolyte Brushes with Tunable Grafting Density

Katie Sun, Russell J. Composto, Karen I. Winey, Zixuan Lin

Weak polyelectrolyte brushes offer tunable control over adsorption through modifications in brush structure and environmental conditions. Solution pH impacts interfacial charge, swelling, and local brush environment, while grafting density directly influences brush conformation, osmotic pressure, and the accessibility of chemical moieties at the interface.1 My research investigates how pH and brush grafting density govern the adsorption behavior of citrate-coated gold nanoparticles (AuNPs). Amine-terminated poly(2-vinylpyridine) (P2VP-NH₂, 40 kg/mol) brushes were prepared at three different grafting densities (σ = 0.10, 0.15, and 0.17 chains/nm2) and challenged with mixed-diameter solutions containing equal picomolar concentrations of 13- and 23-nm diameter AuNPs. After seven days of immersion in AuNP solutions at either pH 4.0 or 6.2, the samples were rinsed, dried, and analyzed using helium ion microscopy (HIM). The HIM images reveal a pH-dependence in adsorption behavior. At pH 4.0, where P2VP-NH₂ is highly protonated and swollen, both nanoparticle sizes adsorb in significant quantities, with adsorption increasing with grafting density. Notably, the system exhibits size-selectivity, with 23 nm AuNPs adsorbing at higher relative densities. In contrast, at pH 6.2, the brushes are mostly deprotonated and collapsed, experiencing much lower overall adsorption and minimal size selectivity. This work demonstrates how weak polyelectrolyte brushes can be engineered to modulate nanoparticle uptake through pH and grafting density changes, contributing to the development of responsive nanoscale separation technologies.

4. Understanding the Effects and Mechanisms of Cytokines and Death Signals on CARM5 T Inflammatory Response, Cell Fitness, and Tumor Control via Multiplex Base Editing

Shimin Liu, John Scholler, Carl June

Chimeric antigen receptor (CAR) T cell therapy has shown remarkable success in hematologic malignancies but remains limited in efficacy against solid tumors. CAR M5 T cells, a new-generation MSLN-CAR T cells, targeting mesothelin expressed in various solid tumors, are being evaluated in clinical trials. To study the effects of various cytokines on CAR M5 T cells, the previous experiments showed that while individual drug neutralization of tumor necrosis factor alpha (TNFα), interferon gamma (IFNγ), or IL1β provided partial protection against lung inflammation, combined blockade of all three cytokines significantly improved survival in human mesothelin knock-in (huMSLN-KI) NSG mouse model. The same combination of blockade also slowed CAR M5 T cell expansion and, in a wt NSG mouse model, dramatically reduced tumor control by CAR M5 T cells. In this study, we aimed to explore how specific cytokine-receptor or combination of TNFα and IFNγ signaling pathways, influence toxic response, function, persistence, and tumor control efficacy of CAR M5 T cell. Using ABE8e and TadCBEd base editors, we screened multiple gRNAs for best knockouts of the genes involved in the cytokine signaling (TNFα, TNFαR1, TNFαR2, IFNγ, and IFNγR). In in vitro killing assay, we found IFNg-KO on CAR M5 T cells significantly impairs their tumor control ability against AsPC1 cells. In the AsPC1-wt NSG mouse model, TNFaR1 knockout (KO) showed a trend toward better tumor control compared to TNFaR2 KO. We also targeted death signaling genes (Fas, FasL, DR3, DR4, DR5, and TRAIL) to study CAR M5 T cell cytotoxic mechanisms mediated by the death receptor pathway and to enhance their persistence. We firstly observed the inverse expression of TRAIL and DR5 on CAR M5 T cells upon activation. Through Incucyte assay, we found that TRAIL-KO on both CAR M5 T+IL18 cells and CAR119+IL18 cells largely decrease their killing ability against Jeko1 WT cells. Consistent with in vitro killing assay, we found in Nalm6-NSG WT mice that TRAIL-KO on CAR119+IL18 cells impairs their killing ability. This study established a base-editing platform to investigate how cytokine and cell-death pathways intrinsically regulate CAR T-cell activity and found that IFNγ and TRAIL are essential for e<ective tumor killing. Further studies will uncover the effects of knockouts of death receptors to improve CAR M5 T cells in persistence and tumor control ability.

5. Decoding Galectin–Galactose Interactions through Molecular Simulations

Rahul Ramaraju, Amish Patel

Galectin–glycan interactions play a central role in cellular recognition, immune regulation, and cancer progression. While their biological importance is well established, quantifying the thermodynamics of binding at atomistic resolution remains challenging. Isothermal titration calorimetry (ITC) provides ΔH, K and estimates of ΔG experimentally, but molecular simulations must capture the same behavior through accurate free-energy landscapes.

In this work, we use enhanced sampling (umbrella sampling) to characterize the binding free energy surface of galectin-galactose. Simple distance-based collective variables (CVs) reveal significant hysteresis and inconsistencies between binding and unbinding profiles, indicating missing internal degrees of freedom. By decomposing the potential energy into protein-sugar electrostatic interactions and solvation effects, we find that most residual hysteresis arises from intramolecular rearrangements within the protein and sugar, rather than solvation differences. 

Together, these results highlight that achieving a hysteresis-free free-energy profile is only the first step; accurately connecting computation to experiment requires integrating intramolecular flexibility, electrostatics, solvation, and restraint corrections into a unified thermodynamic framework.

6. Heterogeneous Ice Nucleation on Textured Surfaces 

Mian Qin, Amish Patel

Heterogeneous ice nucleation (HIN) plays a vital role in atmospheric and cryobiological processes, yet the influence of surface topology remains poorly understood due to its coupling with surface chemistry. In this study, we employ the Phase-Field Model and Interfacial Dynamics to investigate how surface texture—represented by planar surfaces decorated with one or more cylindrical pillars—affects HIN. Contrary to conventional expectations, ice nuclei rarely form axisymmetrically around the pillars; instead, asymmetric structures are typically more stable. To interpret these observations, we derive analytical expressions that reduce the original three-dimensional problem to an equivalent two-dimensional formulation, enabling clearer physical insight. Furthermore, for single-pillar systems, we find that several system properties, including the critical temperature for heterogeneous nucleation, exhibit a sigmoidal dependence symmetric with respect to log r (the pillar radius). These findings establish a foundation for understanding how complex surface topologies influence ice nucleation and suggest the existence of underlying mathematical principles governing such symmetry.

7. Measuring the Traction Forces of Upstream-Migrating Hematopoietic-like KG1a Cells Under Shear Flow

Dong-hun Lee, Daniel A. Hammer

Upstream migration is a distinct mode of leukocyte motility in which cells move against the direction of fluid flow on surfaces with Intercellular Adhesion Molecule-1 (ICAM-1), mediated by the integrin Lymphocyte Function-Associated Antigen-1 (LFA-1). This behavior has been observed across multiple immune cell types including T cells, hematopoietic stem and progenitor cells, and neutrophils. Despite its prevalence in immune cell trafficking, the mechanical forces that drive upstream migration have remained uncharacterized. Here, we apply Traction Force Microscopy (TFM) to quantify the spatiotemporal dynamics of force generation of KG1a cells migrating upstream on ICAM-1 functionalized polyacrylamide hydrogels under physiologically relevant shear flow.

We find that upstream migration is supported by a marked increase in traction force generation relative to migration in static conditions, with peak forces nearly doubling under flow. This enhancement in force output was consistently observed across the upstream-migrating cell population. Despite this increase in magnitude, it is not accompanied by a detectable anterior redistribution of traction: the spatial pattern of force application remains similar to that observed in static conditions. These findings indicate that cells adapt to shear flow primarily by increasing how much force they generate, rather than by changing how that force is spatially deployed.

By providing the first direct measurement of traction forces during upstream migration, this work establishes a quantitative reference point for the mechanics of this behavior. The finding that upstream migration relies on force amplification without large-scale reorganization of traction distribution further clarifies the nature of this response, and sets the stage for dissecting the signaling pathways that enable cells to sense and respond to mechanical stimuli during immune cell trafficking.

8. Surface Modification of microfluidics for scalable manufacture of anisotropic microparticles

Siddharth Sharma, Daeyeon Lee, Dave Issadore, Gijung Kim, Gabriel Rodriguez-Rivera

Microfluidic technologies offer exceptional control over the fabrication of microparticles, enabling precise tuning of size, shape, and composition. This capability has unlocked new possibilities in fields ranging from drug delivery to materials science. However, the scalable production of complex, anisotropic hydrogel microparticles remains challenging due to clogging caused by increased flow resistance during UV polymerization. This is especially a challenge when using Si/glass parallelized high-throughput material generators. 

To address this, we introduce a surface modification strategy using covalently bonded, liquid-like perfluoropolyether (PFPE) polymer coatings on microchannel walls. These coatings act as lubricating layers that reduce interfacial friction and enable the smooth transport of high-aspect-ratio hydrogels through microfluidic channels. We quantitatively and qualitatively compare fluidic resistance in coated versus uncoated devices, demonstrating that uncoated channels experience a drastic rise in resistance—often leading to clogging—as particle aspect ratios increase. In contrast, PFPE-coated devices maintain comparatively consistent flow and particle transport, even under demanding conditions. To quantify this behavior, we assess the impact of surface coating on generated particle size distribution and transport dynamics by measuring particle velocities and their changes over time to confirm the effect of PFPE on the particle generator walls. 

This work presents a scalable and effective solution for high-throughput microfluidic production of anisotropic hydrogels, removing a key bottleneck in the field and enabling broader application of complex microparticles in research and industry.

9. Converting Industrial Waste to High-Value Precipitated Calcium Carbonates

Aline Uwase, Jennifer Wilcox, Max Planck Institute

Carbon dioxide (CO₂) mineralization offers a transformative approach for greenhouse gas mitigation and industrial waste valorization. This study investigates the indirect carbonation of coal fly ash samples sourced from across the United States, focusing on both Class C (calcium-rich, >10% CaO) and Class F (siliceous, <10% CaO) varieties to produce high-purity precipitated calcium carbonate (PCC) for a market projected to reach US$3.16 billion by 2033.

With approximately 32 million tons of coal fly ash produced annually in the United States and only 55% currently beneficially used, our investigation addresses both environmental remediation and economic opportunity. While Class C fly ash offers superior CO₂ sequestration capacity due to its higher calcium content, we demonstrate that Class F fly ash—available in larger quantities but typically limited to use as a concrete additive or landfilled—can also be effectively utilized through optimized processes.

Multiple extraction reagents (glycine, nitric acid, hydrochloric acid, acetic acid, and ammonium salts) were evaluated across diverse fly ash samples to identify optimal leaching protocols. For each extraction system, we established correlations between temperature, residence time, liquid/solid ratio, and reagent concentration to maximize extraction efficiency. Our carbonation studies examined the interplay between temperature, pressure, and reaction duration on both carbonation efficiency and product purity. This work establishes a circular economy framework by transforming industrial CO₂ emissions and coal combustion residuals into marketable calcium carbonate exceeding 97% purity for applications in paper, plastics, paints, pharmaceuticals, and cosmetics, demonstrating the technical and economic viability of carbon mineralization as a sustainable pathway toward industrial decarbonization.

10. Tri-continuous Carbon-based Electrocatalysts for Renewable Energy

Gabriela Gomez-Dopazo, Kathleen Stebe, Daeyeon Lee, Soomin Kim, May El Jamal, Tom Mallouk

Oxygen reduction at air cathodes remains a bottleneck for fuel cells and metal–air batteries due to limited three-phase boundaries (TPBs) and inefficient reactant transport. We introduce a tricontinuous, nanoporous carbon architecture (TRICE) that maximizes TPB density while providing continuous pathways for O_₂ and electrolyte transport. TRICE films are fabricated from a bicontinuous interfacially jammed emulsion gel (BIJEL): a nanoparticle-stabilized oil–water network that is UV-cross-linked and pyrolyzed to form two interpenetrating channels separated by a silica-lined carbon wall. Subsequent silica removal yields a nanoporous carbon wall that exposes abundant TPBs and independent channels for gas and ions. Supported on carbon paper (CP), TRICE significantly improves Zn–air performance; polarization curves show a ~70× increase in peak power density (≈80 mW cm⁻² at 110 mA cm⁻² for CP supported TRICE vs ≈1.2 mW cm⁻² at 4 mA cm⁻² for CP alone), demonstrating the combined benefits of enhanced TPB density and mass transport. To mitigate flooding, we covalently modify the carbon surface via an in-situ diazotization reaction, achieving hydrophobic wetting confirmed by contact-angle measurements. Ongoing work targets asymmetric wetting across the dual-channel network and optimization of TRICE layer thickness to further advance practical air-cathode performance.

11. Site-Specific Enhancements in Electrochemical CO₂ Conversion by Interfacial Additives

Rani Baidoun, Dohyung Kim, Talid Sinno

Electrochemical conversion of CO₂ into fuels and chemicals is a promising approach to address carbon emissions while enabling sustainable energy pathways. While much of catalyst design has focused on tuning its bulk composition and structure, modifying the local reaction environment has recently emerged as a complementary and powerful strategy. In particular, hydrophobic additives have been shown to alter interfacial properties, creating catalytic environments that favor CO₂ reduction over competing reactions. In this work, we investigate how such additives interact with catalytic surfaces at the molecular level and demonstrate that their enhancement effects are highly dependent on the nature of the active sites. Using polycrystalline silver as a model catalyst, we show that hydrophobic cations such as cetyltrimethylammonium dramatically increase the activity of undercoordinated surface sites, boosting CO₂-to-CO turnover frequencies by nearly two orders of magnitude relative to the unmodified interface. In contrast, more coordinated sites display only modest improvements, highlighting a strong site-dependent effect. Importantly, because undercoordinated sites are present in small numbers and are highly sensitive to impurities, even trace amounts of contaminant metals in solution can completely suppress the observed enhancements. These findings reveal both the opportunities and challenges of leveraging local environmental control to accelerate CO₂ electroreduction. More broadly, they underscore the importance of considering site-specific interactions between catalysts and their surrounding environment to design robust and scalable systems. By advancing our understanding of how interfacial additives influence catalysis at distinct surface sites, this work provides new insights into strategies for tailoring catalytic performance and informs the development of next-generation electrochemical systems for CO₂ utilization and beyond.

12. Effect of pendent group placement on the ion transport properties of single ion conducting polymers

Aubry Hymel, Karen I. Winey

Polymer single-ion conductors (SICs) are promising candidates for mechanically robust electrolytes for electrochemical devices due to their reduced flammability hazards and improved charge transference compared to typical organic solvents. However, the ionic conductivities of SICs are orders of magnitude below those necessary for commercial application because of their slow backbone dynamics and large energy barriers for ion transport. Previously, Paren et al. studied a SIC with a polyethylene-like backbone and a Li-neutralized phenyl sulfonate pendant on every 5th carbon (p5PhSA-Li) that sequesters ions into percolated aggregates and demonstrates improved ionic conductivities. We compare the morphology and electrochemical properties of a precise stereo- and sequence-controlled polymer synthesized by the Gutekunst group with vicinal Li-neutralized sulfonated phenyl pendants every 7th and 8th carbons (p8[PhSA-Li]₂) to p5PhSA-Li and commercial lithium polystyrene sulfonate (>95% sulfonated) to investigate the effect of pendent group spacing on ion transport since this parameter changes the ion content and the chain packing arrangements available. Dielectric relaxation spectroscopy and small angle X-ray scattering results show that ion content directly affects the ionic conductivity and inter-aggregate spacing, but electrode polarization analysis shows a non-monotonic trend in free ion concentration and free ion mobility. These results suggest that polymer pendent group placement affects the local aggregate structure and the availability and mobility of ions participating in conduction. The study of these analogous materials aids in understanding how ionic group packing influenced by pendent group placement can enhance SICs’ local ion transport performance. 

13. Fourier Transform Infrared Spectroscopy of Water Dynamics in Hydrocarbon Proton Exchange Membranes

Lindsay Jones, Karen I. Winey, Daniel Vigil, Amaile L. Frischknecht

The effect of sulfonation (Y) on the morphology and proton conductivity of a linear polyethylene with pendant sulfonated phenyl groups on every fifth carbon (p5PhSH-Y) was recently investigated. Because the local water environment of the sulfonate groups is key to understanding conductivity and ion dynamics in the system, we explore the water-polymer interactions using Fourier Transform Infrared Spectroscopy and perform both experiments and simulations as a function of sulfonation levels and relative humidities. The OD stretch peak was deconvoluted with a fitting function containing three Gaussian peaks, representing bulk-like, tightly bound, and other water. The fractional peak areas of the OD stretch peak correspond to the relative amounts of these water types and reflect the nature and strength of the hydrogen bonding between water and the sulfonic acid anion. This additional insight into the water environment will help to design optimized proton exchange membranes. 

14. Phase Behavior of Thin-Film Metal Oxides

Shiqiang Wang, Aleksandra Vojvodic, Yingjie Shi

ABO_x-type (A = Ce, La; B = Fe, Mn, V, Cr; x = 2–4) oxides exhibit tunable redox and catalytic properties, making them promising candidates for energy conversion and environmental applications. However, the factors controlling their thin-film phase behavior remain poorly understood, as these systems involve a combinatorial interplay of composition, stoichiometry, and dimensional confinement. In this study, we employ density functional theory (DFT) to investigate how cation substitution, defect, and strain effects govern the structural stability of representative ABO_x oxides. By understanding the thermodynamics of transformation between bulk and thin-film phases, we reveal the leading factors that stabilize perovskite phases relative to their fluorite counterparts. This work provides atomistic insight into the structure–stability relationships of ABO_x oxides. There, we establish design principles for developing robust oxide catalysts and adaptive thin-film materials.

15. Enhanced Allam Cycle Through High-Temperature Conversion of Captured CO₂ to CO

Aryaman Shah, Warren Seider, John O’Connell

Traditional fossil fuel-based industries have long been troubled by their inherently carbon-intensive nature. Decarbonizing such sectors has been a persistent global challenge, driving the development of closed-loop energy systems that integrate power generation with carbon utilization. Among these, the Allam Cycle stands out as a high-efficiency, oxy-fuel, supercritical CO₂ power cycle capable of achieving near-zero-emission electricity generation. Yet, its full potential remains underexploited, as the captured CO₂ is conventionally directed toward long-term storage rather than productive reuse. This study introduces a novel high-temperature thermochemical pathway that integrates seamlessly with the Allam Cycle to convert captured CO₂ by-products into carbon monoxide (CO), effectively transforming the system from carbon-sequestering to a carbon-utilizing chemical-generation process. This innovation not only addresses the decarbonization imperative, but also creates an additional revenue stream through the generation of value-added chemicals and fuels such as methanol, higher alcohols, and syngas derivatives. Moreover, this novel process seeks to decarbonize emissions-heavy industries thus, allowing the U.S. to meet its power demands in a clean, sustainable manner.

16. Morphological Impacts and Diffusive Properties of Organic Solvents in Self-Assembled Nanoporous Membranes

Jack Granite, Chinedum Osuji

Membrane technologies hold promise to achieve highly selective separations, reducing energy costs in some cases by a factor of 2–3 compared to thermal methods. Conventional polymer membranes suffer from tradeoffs in their permeance (flux normalized by operating pressure) and selectivity due to broad distributions in pores sizes, which limits their industrial implementation. Self-assembled membranes address this issue− they have monodisperse pores of one transport limiting dimension due to the governance of pore formation by the thermodynamics of self-assembly. To date, these novel membranes have been heavily explored for their capability in aqueous nanofiltration and ion separations, and have shown promising performance compared to conventional polymer membranes. However, organic solvent separations remain mostly unexplored in these systems. In this work, we reveal the critical role of polymerization sites in amphiphilic monomer design and highlight key materials design principles for stability of these membranes in industrially relevant organic solvents. Additionally, we investigate the sorption of organic species in these membranes and show that uptake depends on solvent polarity, since these membranes are charged. Finally, through simple h-cell experiments coupled with gas chromatography, we elucidate the diffusive properties of organic species in these membranes and highlight selective separations based on both size-exclusion and chemical affinity.

17. Pathways to Carbon-Negative Aviation: A Techno-Economic and Life-Cycle Perspective

Shrey Patel, Jennifer Wilcox, Hélène Pilorgé, Greg Cooney, Makenna Damhorst

Aviation currently accounts for approximately 3.3% of U.S. carbon dioxide emissions, and its share could double by 2050 without major intervention. While sustainable aviation fuels (SAFs) represent a key route toward decarbonizing the aviation industry, their production costs remain two to ten times higher than those of conventional jet fuel. This study presents a combined techno-economic analysis (TEA) and life-cycle assessment (LCA) of using Fischer-Tropsch (FT) synthesis integrated with biomass gasification and direct air capture (DAC) to produce not just carbon-neutral, but carbon-negative SAF. 

Sustainably sourced biomass is gasified in an oxy-fired gasifierto produce hydrogen, and the produced carbon dioxide is captured and sequestered underground yielding net-negative emissions. Direct Air Capture captures atmospheric carbon dioxide which is converted to carbon monoxide via reverse water-gas shift (RWGS) and reacted with renewable hydrogen from the gasification step for Fischer-Tropsch synthesis, followed by hydrocracking and isomerization steps yielding carbon-negative SAF as the final product.

Carbon intensities of -17 to -26 kg CO2 per gallon of SAF can be achieved with this process configuration. Modeling for California’s policy framework shows that the estimated production cost of $13-16 per gallon can be reduced substantially through policy incentives such as 45Q, 45V, and LCFS, bringing the ecective cost of a carbon-neutral blend with conventional jet fuel to $3.5-4.5 per gallon, approaching parity with conventional fossil-derived fuel.

The produced carbon-negative SAF can be blended with conventional Jet A to obtain a carbon-neutral Jet fuel mix, further reducing the fuel cost as a near-term strategy. Other pathways using alternative hydrogen and carbon dioxide sources are also analyzed for comparison. This analysis demonstrates that integrating biomass conversion, carbon capture, and fuel synthesis can deliver a practical, near-term pathway to low-carbon aviation fuels, providing a bridge between today’s infrastructure and a fully decarbonized aviation sector.

18. Vibrational frequencies from ab initio and deep potential molecular dynamics at SnO₂-water interfaces

Daniel Intriago, Aleksandra Vojvodic, Karen I. Winey

The entropic effects of intermediates interacting with the surface at electrocatalyst-electrolyte interfaces potentially play a vital role in the water dissociation reaction barrier. While some of these effects can be measured with experimental techniques (surface-sensitive water orientation measurements and IR), the use of computational methods can determine effects of surface proximity and specific ligand chemistry on these properties at the atomistic level. Density functional theory molecular dynamics (DFT-MD) was used to sample configurations at the catalyst-electrolyte interface of rutile SnO₂ systems in water. The autocorrelation of dipoles from H₂O, OH, and O species were calculated to generate theoretical IR spectra as a function of surface proximity. This technique resulted in calculating relevant frequency shifts that arise from dissociative water adsorption – specifically, strong redshifts of up to 400 cm^-1 of species near the catalyst surface were found in systems where dissociative water adsorption spontaneously occurs. Very weak redshifts can still be observed in non-dissociative water adsorption, which are attributed to adsorbate-surface interaction effects. Finally, the results from DFT-MD were then used to train a small-package machine learned interatomic potential (MLIP) to increase sampling of the vibrational frequencies at these SnO₂-water interfaces. Not only did this provide sufficient sampling for low-frequency modes – and thus more representative relative peak heights – but spontaneous water dissociation was still observed in these deep-potential molecular dynamics systems.

19. Chemistries of IrOx Morphologies

Joseph Nicolas, Aleksandra Vojvodic

Iridium oxide (IrO₂) is the state-of-the-art electrocatalyst for water oxidation in electrolyzers, yet it suffers from instability under operating conditions. Here, we combine first-principles modeling with in-situ solid–liquid characterization to resolve the atomic-scale morphology and dissolution dynamics of IrO₂ nanocrystals. Our computational Wulff constructions uniquely incorporate high- index facets, providing new insights into thermodynamic facet-dependent stability under operating conditions. Atomically resolved studies reveal multiple dissolution pathways, including high-index facet formation via step-edge restructuring on {110} surfaces and monolayer delamination. Ab initio molecular dynamics simulations further show that initial dissolution kinetics are facet-dependent. These findings highlight how combining in-situ imaging with modeling reveals atomic-scale dynamics that influence material performance.

20. Distribution of local active sites in the electrocatalytic behavior of mutlt-metallic 2D carbides

Bo Spinnler, Aleks Vojvodic, Anupma Thakur, Brian Cecil Wyatt, Babak Anasori

Recent studies have successfully synthesized multi-metallic MXenes, a type of 2D layered carbide material which can contain between 2 and 9 different transition metals. MXenes have garnered great interest as electrochemical catalysts due to the tunability of MXenes chemistry combined with their metal-like electrical conductivity and hydrophilicity. However, the effect of introducing multiple metallic elements into the MXene structure on electrocatalysis has not been well-studied experimentally nor computationally. This work uses the electrochemical hydrogen evolution reaction (HER) as a model reaction to measure the catalytic activity of several multi-metallic MXene compositions. We report that (MoTiVW)C₃T_x is the best-performing multi-metallic MXene out of the experimentally screened compositions for HER. Additionally, density functional theory (DFT) calculations were performed to better understand the catalytic behavior of the 4TM MXenes. This work applies a statistical approach to best model the catalytic activity of multi-metal MXenes and finds that the catalytic activity is strongly related to the probability distribution of local active sites present on the MXene surface, as well as the theoretical overpotential of each local active site. With a statistical approach, there is agreement HER activity trends between experiment and computation for the various compositions of MXenes screened. This work re-emphasizes the tunability of MXenes and introduces the surface site distribution and local active sites as a path forward to improve the catalytic activity of multi-metal MXenes which may also be tuned for other electrochemical reactions.

21. Towards Biophysics-Driven Digital Twins for Personalized Immunotherapy

Stephanie Monson, Ravi Radhakrishnan, Bomyi Lim, Jina Ko

Digital twins are data-driven virtual models of real-world systems used to analyze and predict system performance in industries ranging from healthcare to manufacturing. In biomedicine, a digital twin of a patient’s tumor programmed with diagnostic data can be used to predict response to immunotherapy options. This prediction is otherwise challenging due to the heterogeneous and patient-specific nature of cancer. According to a 2023 National Academies report, an integral research requirement for digital twins is the development of models which are multiscale, mechanistic, and fit-for-purpose. Recent studies on cancer mechanisms have uncovered a complex, multiscale relationship between tumor-derived exosomes, immune modulation, tumor progression, and therapeutic resistance. Thus, an effective tumor digital twin must incorporate mechanistic models of these phenomena.

In my poster and lightning talk, I demonstrate how we are using multiscale physics to model and investigate this complex interplay in the tumor microenvironment. In recent work, we have elucidated how surface topology and immune modifications of a nanoparticle impact cellular recognition. These findings provide important insights for treatment planning and personalization. For instance, the present framework can be employed to understand disease mechanisms and inform optimal therapeutic design given patient biomarkers. Planned work includes integrating the cellular-level model with our signaling kinetics and tissue-level tumor models to study the impact on tumor progression. We will also perform verification, validation, and uncertainty quantification using clinical data to enable well-informed, personalized decision-making.

22. — Abstract Retracted —

Xuelin Yang, De-en Jiang, Vikash Khokhar

23. How do Antifreeze Proteins Inhibit Ice Recrystallization?

Jeongmoon Choi, Amish J. Patel

Recrystallization is a phenomenon in which larger ice crystals grow at the expense of smaller ones. Both theory and experiments have shown that during recrystallization, the cube of the average crystal size grows linearly with time, r̅³ = kt. However, in the presence of antifreeze proteins (AFPs), experiments have shown that ice recrystallization can be retarded and even arrested altogether above a critical AFP concentration, c^∗. In this study, we propose a theoretical model explaining how AFPs suppress ice recrystallization and how their molecular properties influence c^∗. We also interrogate how experimental parameters, such as sucrose/saltconcentration, temperature, and the initial crystal size distribution, influence c^∗. Our findings offer insights into the relationship between AFP properties and c^∗, which could guide the designof novel materials for ice recrystallization inhibition.

24. Machine Learning Accelerated Radiative Transfer Modeling in CFD Fire Simulations

Mitchell Daneker, Lu Lu, Xiaoyi Lu

Radiative heat transfer is the dominant heat transfer mode in many fire scenarios, yet its solution procedure is often a major performance bottleneck of Computational Fluid Dynamics (CFD) simulations. Conventional solvers for the governing radiative transfer equation (RTE) are computationally expensive and can suffer from numerical artifacts. This paper presents a machine learning-accelerated RTE solver based on Deep Operator Networks (DeepONet) to address these limitations. The DeepONet is trained as a neural surrogate to learn the underlying solution operator of the RTE, directly mapping input functions, such as absorption coefficient and temperature, to the output radiative intensity solution. This pre-trained DeepONet model is then integrated into the open-source CFD code FireFOAM, replacing the baseline finite-volume discrete-ordinate method (fvDOM) solver. The performance of this hybrid ML-CFD framework is evaluated in transient simulations of canonical fire scenarios, including two-dimensional (2D) and three-dimensional (3D) pool fires and a 3D wall fire. Results demonstrate that the DeepONet solver reproduces the radiation solutions with significant computational speed-up. Moreover, when using the DeepONet solver at high resolution, the ray effect is mitigated.

25. 3D Printing of Bicontinuous Nanoparticle-Stabilized Emulsion Gels via Co-Solvent Removal

Phillip R. Iaccarino, Damilola Lawal, Jordan R. Raney, Daeyeon Lee, Kathleen J, Stebe

Hierarchical microporous materials are found throughout nature. For example, bones provide structural support throughout the body of many animals, with a foam-like microporous internal structure that allows for a strong but lightweight material while also promoting cell growth and self-healing mechanisms. These properties exist due to the bone’s functional hierarchical structure which includes distinct structural pathways for fluid, a microporous solid supporting structure. Recapitulating such complex structures in synthetic systems is a significant materials challenge. Bicontinuous emulsion gels represent an important advance in this direction. Bicontinuous emulsion gels consist of interpenetrating channels of two immiscible fluids stabilized with nanoparticles. These materials are readily processed to form microporous scaffolds with unique transport and optical properties. However, current fabrication methods typically limit their macroscale morphology to sheets, fibers, and microparticles, limiting applications that call for complex morphologies; the ability to generate large scale bicontinuous materials with arbitrary morphologies would be a significant advance.

We present the fabrication of bicontinuous emulsion gels via 3D printing. We develop a precursor ink, consisting of a one-phase miscible ternary liquid mixture and fumed silica nanoparticles, which phase separates into a bicontinuous emulsion gel via co-solvent evaporation. Fumed silica provides the requisite rheological profile to support direct ink write (DIW) extrusion and stabilizes the bicontinuous channels through interfacial jamming and bulk gelation mechanisms. We combine top-down DIW additive manufacturing to control the material’s morphology on the centimeter scale in three dimensions with bottom-up assembly mechanisms that generate and stabilize the bicontinuous microstructure on submicrometer scales. These biphasic, hierarchical materials are exciting in a variety of applications, such as biomedical implants, tissue engineering, filtration systems, and more.

26. Gas-Bubble Encapsulating Microcapsules (GEMs)

CK Yeh, Daeyeon Lee

Microcapsules that can respond under hydrostatic pressure would open a new avenue of application in ultrasound- or impact-induced release of therapeutic agents. While microcapsules that are designed to release cargo under uniaxial compressive loadings have been developed, they are filled with liquid, rendering them insensitive to hydrostatic pressure. To overcome this limitation, bubbles can be introduced into the core of the capsule that can impart hydrostatic pressure sensitivity. Although microcapsules that contain bubbles have been reported, the methods for their fabrication make it difficult to control the size of the bubble and microcapsule. In this work, we use microfluidics to create uniform microcapsules while also utilizing osmosis-induced-cavitation to convert microcapsules into gas-encapsulating microcapsules (GEMs). Following GEMs formation under high salt concentrations, the size of the bubble can be modulated by placing the GEMs back into relatively lower salt concentrations. GEMs with a well-defined thickness to diameter ratio (t/D) and volume fraction of gas are subjected to a known hydrostatic pressure through the means of a drop tower. These data establish a relationship between how the thickness to diameter ratio and volume fraction of bubble influence the rupture pressure and drug release, helping to establish the foundation for a new class of on-demand therapeutics that take advantage of pressure inputs.

27. Spatially Varying Alignment of Block Copolymers Using Nonuniform Magnetic Fields

Haotian Long, Chinedum Osuji

Self-assembly of block copolymers is broadly considered an attractive means of generating nanoscale structures and patterns in nanotechnology and compelling for creating functional materials and devices. Here, we introduce an approach to create spatially varying alignment in a cylindrical block copolymer using structured, nonuniform magnetic fields. These fields are generated by using a ferromagnetic block to distort an applied uniform magnetic field. We show that the resulting alignment can be tuned by changing the strength of the applied field and the geometry of the ferromagnetic material, which corresponds well with our theoretical calculations and simulations. We also show that this strategy could be a generic way to structure the alignment of a variety of soft materials, including homopolymers, block copolymers, and potentially nanocomposites. We speculate that this can be an approach towards precise control of the soft materials at the nanoscale and creating complex nanostructures, including 3D morphology.

28. Quantifying Water Channels in Fluorine-Free Proton-Conducting Polymers

Lily Wang, Karen I. Winey, Stephen Kronenberger, Amalie L. Frischknecht, Arthi Jayaraman

Commercially available perfluorosulfonic acid polymers are commonly used as proton exchange membranes (PEMs) due to their favorable mechanical and electrochemical properties. Because of the high cost and toxicity associated with fluorine-containing polymers, alternative hydrocarbon-based PEMs are of interest. Well-defined nanoscale water channels lined with sulfonic acid groups are a key feature of these materials that we recently reproduced in a series of fluorine-free polymers. Previously, we used all-atom molecular dynamics (MD) simulations to reveal details about the water channels in a range of hydration levels. In this work, we quantitatively compare the nature of water channels found by MD with experimental small-angle X-ray scattering (SAXS) data. We reconstructed three-dimensional real-space nanoscale morphologies from SAXS data using Gaussian random fields. We developed methods to characterize the water channels within these reconstructions, specifically calculating the characteristic lengths, water clustering, tortuosity, channel width distributions, and fractal dimensions. Our results show good agreement between all-atom MD simulations, suggesting that this work can be used to reliably reconstruct and analyze nanoscale morphologies in phase-separated soft materials. 

29. Low-Temperature Stability of the Double Gyroid Morphology in Ultrahigh-χ Low-N Multiblock Copolymers

Margaret Brown, Karen I. Winey, Viola A. Burlein, Sharin Rashid, Stefan Mecking

In previous work, the double gyroid morphology was reported in multiblock copolymers with strictly alternating linear alkyl blocks of 12 to 23 carbons and sodium or lithium sulfonated blocks which contain 4 carbons (PES4Mx). The bicontinuous morphology showed enhanced ionic conductivity compared to layers or hexagonally packed cylinders, but it was not accessible below ~100 °C due to the crystallization of the alkyl block. In this work, the sulfonated block is modified by adding an additional carbon (PES5Nax) to increase its flexibility and disrupt crystallization. This modification successfully inhibits polymer recrystallization upon cooling which, for some values of x, results in the double gyroid morphology at room temperature. This morphology remains stable at room temperature for over 1 year when x = 18. Recrystallization on cooling is only seen for polymers where x ≥ 18. Fitting of disordered X-ray scattering data indicates an ultrahigh-χ for this class of materials, similar to what was previously observed in PES4Li12.  This highlights the importance of considering packing frustration in both blocks when designing ion-containing multiblock copolymers with network morphologies.

30. Diffusion of Bottlebrush Polymer Grafted Nanoparticles: The Role of Matrix Molecular Weight

Ben Indeglia, Karen I. Winey, Jensen N. Sevening, Robert J. Hickey

Nanoparticle diffusion is a valuable heuristic for structure-property relations in polymer nanocomposites. Nanoparticle diffusion in a polymer matrix is known to be sensitive to surface chemistry and size of the nanoparticles, and previous studies have focused on bare nanoparticles in athermal and attractive melts, or nanoparticles with linear grafted chain architecture. This investigation explores the effect of non-linear polymer grafted nanoparticles (PGNs) on nanoparticle diffusion. We report the diffusion coefficients of nanoparticles grafted with bottlebrush polymers in melts of linear polymers. These bottlebrush polymer grafted nanoparticles (PGNs) are comprised of large silica NPs (radius 80 nm) grafted with various backbone degrees of polymerization and fixed polystyrene side chains (4.2 kDa) to give bottlebrush molecular weights of 80 to 388 kDa. The polymer matrix is polystyrene (100 or 425 kDa), and nanoparticle diffusion was studied at 170 °C. After annealing a trilayer sample, time-of-flight secondary ion mass spectroscopy (ToF-SIMS) measured the cross-sectional PGN concentration profiles and extracted diffusion coefficients. The PGNs with high bottlebrush molecular weights relative to the matrix polymer display core-shell diffusion behavior, wherein the shell thickness is comparable to the grafted bottlebrush thickness. Surprisingly, when the bottlebrush molecular weight is smaller than the matrix polymer, the PGNs diffuse 10 – 100 times faster than predicted, which we attribute to a local decrease in viscosity associated with disentanglement of the matrix polymer near the large nanoparticles.

1. Development and Validation of HIV Infectious Molecular Clones with Diverse Neutralization Profiles

Jiajie Li, Katherine Bar, Ryan Krause, Himanshu Garg, Jake Robinson, Noah Beratan, Hanah Schrader, Rebecca Lynch

Human immunodeficiency virus (HIV) remains a major global health challenge, and  broadly neutralizing antibodies (bnAbs) show great promise in combating infection.  bnAbs bind to the HIV envelope (Env) glycoprotein on the surface of virions and impede  viral entry into target cells. In our previous study, Panels of relevant Envs and bnAbs,  which were identified from our and published human clinical trials, were screened and  resulted in the selection of seven Envs representing a spectrum of sensitivity and  resistance to a bnAb pair (8ANC195 and N49). 

To characterize respective neutralization sensitivities and replication fitness of seven  Envs, this project focuses on engineering Env-chimeric infectious molecular clones  (IMCs) of replication-competent viruses to serve as research accessible resources. The  well-characterized HIV-1 NL4-3 plasmid was selected as the HIV backbone. Through In-Fusion Assembly, the env region was seamlessly replaced between the EcoRI site  (within vpr, creating overhang across pol-vif-vpr) and the XhoI site (within nef, creating  overhang across nef-3’LTR). This design prevents frame shift disruptions in overlapping  genes and regulatory elements, preserving full replication competence. IMCs’ functionality was validated by infectivity titration through TZM-bl cells first, in which all  IMCs were successfully infectious. Second, we validated replication fitness by in vitro infection of human primary CD4⁺ T cells, and through AlphaLISA detection of p24 antigen (capsid protein), we confirmed robust viral replication in host HIV target cells.  Last, neutralization assays comparing HEK293T-derived IMCs (original production) and  CD4⁺ T cell-passaged IMCs (natural infection) confirmed that Env-specific neutralization  phenotypes from selection criteria were maintained in IMCs. Replication experiments in  primary human CD4+ T cells and humanized mice to test relative replication fitness are  ongoing. This study demonstrates the utility of a scalable, reliable molecular platform for  generating replication-competent HIV clones. Engineered HIV IMCs with unique Envs  that maintain sensitivity/resistance profiles to clinically used bnAbs provide an essential  tool for downstream in vitro and in vivo studies to test combinatorial approaches to bnAb strategies and to highlight synergistic methods for managing HIV.

2. Engineering split Cas13bt3 for inducible RNA editing

Devin Golla, Sherry Gao, Tyler Daniel, Richard Phillips

Precision gene editing technologies such as base and prime editing can enable specific and programmable insertions, deletions, or substitutions into targeted genomic loci. Recent development of these technologies has opened up exciting new applications to model and treat complex diseases involving multiple genes, which has motivated a need for compact, modular systems capable of expressing multiple guide RNAs (gRNAs) from a single transcript. Solutions like the drive-and-process (DAP) array, a modular architecture composed of alternating gRNA and tRNA units, work by utilizing the cell’s endogenous tRNA processing machinery to cleave each tRNA from the array and release individual gRNAs, enabling simultaneous editing at multiple loci; however, the array’s efficiency may vary between organisms, limiting its breadth of application. We hypothesize that the specific tRNA sequence chosen for the DAP array may in part dictate the system’s efficiency across organisms, and have tested and chosen top tRNA candidates shown to enable 2-loci base editing at previously validated gene sites in N2A (mouse) cells. We are in the process of validating the editing efficiency of these chosen tRNAs in 4-loci DAP arrays, and are working to elucidate how the choice of tRNA impacts the extent of endogenous tRNA processing. Along with our collaborators, we aim to utilize our eventual mouse-compatible DAP array to simultaneously edit a selection of loci known to co-occur in glioblastoma, in order to create a novel mouse model of the disease.

3. Heterogeneous Escape From X Chromosome Inactivation in Lymphocytes Revealed by Quantitative Allele-Specific FISH

Peyton Collins, Bomyi Lim, Montserrat Anguera

Females often mount stronger immune responses than males, resulting in better outcomes from  infection but also a higher risk of autoimmunity. Because the X chromosome encodes many  immune-related genes, differences in gene dosage may underlie these sex biases. However, in  therian mammals, dosage balance is accomplished by X chromosome inactivation (XCI), which hinders transcription from all but one randomly chosen X chromosome, through a process  regulated by the long noncoding RNA Xist. Notably, in mouse B and T cells, the recruitment of  Xist to the Xi is stimulation-dependent and heterogenous. Genes that continue to be transcribed  from the Xi, called “escapees”, have the potential to accomplish the imbalance that multiple X  chromosomes portend. The mechanism by which genes escape is poorly understood, but the stimulation-dependent heterogeneity of Xist localization suggests that escape from XCI may  behave similarly. 

Immune effector genes themselves are expressed heterogeneously across lymphocyte populations  in response to stimulation. Whether escapees follow a similar pattern of heterogeneous regulation  remains unknown, since most studies rely on bulk RNA-seq when reporting gene expression  from the Xi. This work intends to fill this gap in the field by applying sequential RNA and DNA  fluorescent in situ hybridizations (FISH) to measure allele-specific nascent transcription in naïve  and stimulated lymphocytes. Analysis of two constitutive escapees in thousands of cells, the demethylase Kdm6a and the RNA helicase Ddx3x, revealed that both the intensity and frequency  of escape increase in response to stimulation in splenic B220+ and T cells. Although escape in  both naïve and stimulated lymphocytes accounts for a significant portion of total gene  expression, it occurs in only a subset of cells, suggesting heterogenous regulation. This approach  opens the door to investigating immune-related escapees and their role in sex-biased immunity.

4. Understanding the Effects and Mechanisms of Cytokines and Death Signals on CARM5 T Inflammatory Response, Cell Fitness, and Tumor Control via Multiplex Base Editing

Shimin Liu, John Scholler, Carl June

Chimeric antigen receptor (CAR) T cell therapy has shown remarkable success in hematologic malignancies but remains limited in efficacy against solid tumors. CAR M5 T cells, a new-generation MSLN-CAR T cells, targeting mesothelin expressed in various solid tumors, are being evaluated in clinical trials. To study the effects of various cytokines on CAR M5 T cells, the previous experiments showed that while individual drug neutralization of tumor necrosis factor alpha (TNFα), interferon gamma (IFNγ), or IL1β provided partial protection against lung inflammation, combined blockade of all three cytokines significantly improved survival in human mesothelin knock-in (huMSLN-KI) NSG mouse model. The same combination of blockade also slowed CAR M5 T cell expansion and, in a wt NSG mouse model, dramatically reduced tumor control by CAR M5 T cells. In this study, we aimed to explore how specific cytokine-receptor or combination of TNFα and IFNγ signaling pathways, influence toxic response, function, persistence, and tumor control efficacy of CAR M5 T cell. Using ABE8e and TadCBEd base editors, we screened multiple gRNAs for best knockouts of the genes involved in the cytokine signaling (TNFα, TNFαR1, TNFαR2, IFNγ, and IFNγR). In in vitro killing assay, we found IFNg-KO on CAR M5 T cells significantly impairs their tumor control ability against AsPC1 cells. In the AsPC1-wt NSG mouse model, TNFaR1 knockout (KO) showed a trend toward better tumor control compared to TNFaR2 KO. We also targeted death signaling genes (Fas, FasL, DR3, DR4, DR5, and TRAIL) to study CAR M5 T cell cytotoxic mechanisms mediated by the death receptor pathway and to enhance their persistence. We firstly observed the inverse expression of TRAIL and DR5 on CAR M5 T cells upon activation. Through Incucyte assay, we found that TRAIL-KO on both CAR M5 T+IL18 cells and CAR119+IL18 cells largely decrease their killing ability against Jeko1 WT cells. Consistent with in vitro killing assay, we found in Nalm6-NSG WT mice that TRAIL-KO on CAR119+IL18 cells impairs their killing ability. This study established a base-editing platform to investigate how cytokine and cell-death pathways intrinsically regulate CAR T-cell activity and found that IFNγ and TRAIL are essential for e<ective tumor killing. Further studies will uncover the e<ects of knockouts of death receptors to improve CAR M5 T cells in persistence and tumor control ability.

5. Effect of pendent group placement on the ion transport properties of single ion conducting polymers

Aubry Hymel, Karen I. Winey, Ethan Wagner, Will Gutekunst

Polymer single-ion conductors (SICs) are promising candidates for mechanically robust electrolytes for electrochemical devices due to their reduced flammability hazards and improved charge transference compared to typical organic solvents. However, the ionic conductivities of SICs are orders of magnitude below those necessary for commercial application because of their slow backbone dynamics and large energy barriers for ion transport. Previously, Paren et al. studied a SIC with a polyethylene-like backbone and a Li-neutralized phenyl sulfonate pendant on every 5th carbon (p5PhSA-Li) that sequesters ions into percolated aggregates and demonstrates improved ionic conductivities. We compare the morphology and electrochemical properties of a precise stereo- and sequence-controlled polymer synthesized by the Gutekunst group with vicinal Li-neutralized sulfonated phenyl pendants every 7th and 8th carbons (p8[PhSA-Li]₂) to p5PhSA-Li and commercial lithium polystyrene sulfonate (>95% sulfonated) to investigate the effect of pendent group spacing on ion transport since this parameter changes the ion content and the chain packing arrangements available. Dielectric relaxation spectroscopy and small angle X-ray scattering results show that ion content directly affects the ionic conductivity and inter-aggregate spacing, but electrode polarization analysis shows a non-monotonic trend in free ion concentration and free ion mobility. These results suggest that polymer pendent group placement affects the local aggregate structure and the availability and mobility of ions participating in conduction. The study of these analogous materials aids in understanding how ionic group packing influenced by pendent group placement can enhance SICs’ local ion transport performance. 

6. Fourier Transform Infrared Spectroscopy of Water Dynamics in Hydrocarbon Proton Exchange Membranes

Lindsay Jones, Karen I. Winey, Daniel Vigil, Amaile L. Frischknecht

The effect of sulfonation (Y) on the morphology and proton conductivity of a linear polyethylene with pendant sulfonated phenyl groups on every fifth carbon (p5PhSH-Y) was recently investigated. Because the local water environment of the sulfonate groups is key to understanding conductivity and ion dynamics in the system, we explore the water-polymer interactions using Fourier Transform Infrared Spectroscopy and perform both experiments and simulations as a function of sulfonation levels and relative humidities. The OD stretch peak was deconvoluted with a fitting function containing three Gaussian peaks, representing bulk-like, tightly bound, and other water. The fractional peak areas of the OD stretch peak correspond to the relative amounts of these water types and reflect the nature and strength of the hydrogen bonding between water and the sulfonic acid anion. This additional insight into the water environment will help to design optimized proton exchange membranes.  

7. How can Nanoparticle Adsorption be Tuned via Grafting Density of Polymer Brushes?

Katie Sun, Karen I. Winey, Russell J. Composto, Zixuan Lin

Polyelectrolyte brushes enable the design of stimuli-responsive surfaces for applications in sensing, separations, and interfacial engineering. Their behavior is primarily dictated by three structural parameters—dispersity, molar mass, and grafting density. Grafting density, in particular, directly influences brush conformation, osmotic pressure, and the accessibility of chemical moieties at the interface. Low-grafting-density brushes provide architectures advantageous for post-fabrication functionalization, such as conjugation platforms and biosensing interfaces. Meanwhile, high grafting density brushes are commonly utilized in anti-fouling coatings or membrane filtration. In this lightning talk, I focus on amine-terminated poly(2-vinylpyridine) (P2VP-NH₂, 40 kg/mol) brushes and the impact of varying environmental pH and brush grafting density (σ = 0.10, 0.15, and 0.17 chains/nm²) on the adsorption of spherical citrate-coated gold nanoparticles (AuNPs). Specifically, the P2VP-NH₂ brushes were immersed for seven days in mixed-diameter solutions at either pH 4.0 or 6.2 containing equal picomolar concentrations of 13- and 23-nm diameter AuNPs. The size distribution of adsorbed AuNPs were extracted from helium ion microscopy (HIM) images, with the pH-dependent and selective adsorption trends elaborated upon in my poster. These findings clarify the relationship between grafting density and pH and highlight strategies for engineering next-generation responsive interfaces.

8. Liquid Crystals in Spatially Varying Magnetic Fields with Antagonistic Anchoring

Shrey Patel, Jennifer Wilcox, Hélène Pilorgé, Greg Cooney, Makenna Damhorst

Aviation currently accounts for approximately 3.3% of U.S. carbon dioxide emissions, and its share could double by 2050 without major intervention. While sustainable aviation fuels (SAFs) represent a key route toward decarbonizing the aviation industry, their production costs remain two to ten times higher than those of conventional jet fuel. This study presents a combined techno-economic analysis (TEA) and life-cycle assessment (LCA) of using Fischer-Tropsch (FT) synthesis integrated with biomass gasification and direct air capture (DAC) to produce not just carbon-neutral, but carbon-negative SAF. 

Sustainably sourced biomass is gasified in an oxy-fired gasifierto produce hydrogen, and the produced carbon dioxide is captured and sequestered underground yielding net-negative emissions. Direct Air Capture captures atmospheric carbon dioxide which is converted to carbon monoxide via reverse water-gas shift (RWGS) and reacted with renewable hydrogen from the gasification step for Fischer-Tropsch synthesis, followed by hydrocracking and isomerization steps yielding carbon-negative SAF as the final product.

Carbon intensities of -17 to -26 kg CO2 per gallon of SAF can be achieved with this process configuration. Modeling for California’s policy framework shows that the estimated production cost of $13-16 per gallon can be reduced substantially through policy incentives such as 45Q, 45V, and LCFS, bringing the ecective cost of a carbon-neutral blend with conventional jet fuel to $3.5-4.5 per gallon, approaching parity with conventional fossil-derived fuel.

The produced carbon-negative SAF can be blended with conventional Jet A to obtain a carbon-neutral Jet fuel mix, further reducing the fuel cost as a near-term strategy. Other pathways using alternative hydrogen and carbon dioxide sources are also analyzed for comparison. This analysis demonstrates that integrating biomass conversion, carbon capture, and fuel synthesis can deliver a practical, near-term pathway to low-carbon aviation fuels, providing a bridge between today’s infrastructure and a fully decarbonized aviation sector.

9. Phase Behavior of Thin-Film Metal Oxides

Shiqiang Wang, Aleksandra Vojvodic, Yingjie Shi

ABO_x-type (A = Ce, La; B = Fe, Mn, V, Cr; x = 2–4) oxides exhibit tunable redox and catalytic properties, making them promising candidates for energy conversion and environmental applications. However, the factors controlling their thin-film phase behavior remain poorly understood, as these systems involve a combinatorial interplay of composition, stoichiometry, and dimensional confinement. In this study, we employ density functional theory (DFT) to investigate how cation substitution, defect, and strain effects govern the structural stability of representative ABO_x oxides. By understanding the thermodynamics of transformation between bulk and thin-film phases, we reveal the leading factors that stabilize perovskite phases relative to their fluorite counterparts. This work provides atomistic insight into the structure–stability relationships of ABO_x oxides. There, we establish design principles for developing robust oxide catalysts and adaptive thin-film materials.

10. Distribution of local active sites in the electrocatalytic behavior of multi-metallic 2D carbides

Bo Spinnler, Anupma Thakur, Brian Cecil Wyatt, Babak Anasori

Recent studies have successfully synthesized multi-metallic MXenes, a type of 2D layered carbide material which can contain between 2 and 9 different transition metals. MXenes have garnered great interest as electrochemical catalysts due to the tunability of MXenes chemistry combined with their metal-like electrical conductivity and hydrophilicity. However, the effect of introducing multiple metallic elements into the MXene structure on electrocatalysis has not been well-studied experimentally nor computationally. This work uses the electrochemical hydrogen evolution reaction (HER) as a model reaction to measure the catalytic activity of several multi-metallic MXene compositions. We report that (MoTiVW)C₃T_x is the best-performing multi-metallic MXene out of the experimentally screened compositions for HER. Additionally, density functional theory (DFT) calculations were performed to better understand the catalytic behavior of the 4TM MXenes. This work applies a statistical approach to best model the catalytic activity of multi-metal MXenes and finds that the catalytic activity is strongly related to the probability distribution of local active sites present on the MXene surface, as well as the theoretical overpotential of each local active site. With a statistical approach, there is agreement HER activity trends between experiment and computation for the various compositions of MXenes screened. This work re-emphasizes the tunability of MXenes and introduces the surface site distribution and local active sites as a path forward to improve the catalytic activity of multi-metal MXenes which may also be tuned for other electrochemical reactions.

11. Decoding Galectin–Galactose Interactions through Molecular Simulations

Rahul Ramaraju, Amish Patel

Galectin–glycan interactions play a central role in cellular recognition, immune regulation, and cancer progression. While their biological importance is well established, quantifying the thermodynamics of binding at atomistic resolution remains challenging. Isothermal titration calorimetry (ITC) provides ΔH, K and estimates of ΔG experimentally, but molecular simulations must capture the same behavior through accurate free-energy landscapes.

In this work, we use enhanced sampling (umbrella sampling) to characterize the binding free energy surface of galectin-galactose. Simple distance-based collective variables (CVs) reveal significant hysteresis and inconsistencies between binding and unbinding profiles, indicating missing internal degrees of freedom. By decomposing the potential energy into protein-sugar electrostatic interactions and solvation effects, we find that most residual hysteresis arises from intramolecular rearrangements within the protein and sugar, rather than solvation differences. 

Together, these results highlight that achieving a hysteresis-free free-energy profile is only the first step; accurately connecting computation to experiment requires integrating intramolecular flexibility, electrostatics, solvation, and restraint corrections into a unified thermodynamic framework.

12. Heterogeneous Ice Nucleation on Textured Surfaces

Mian Qin, Amish Patel

Heterogeneous ice nucleation (HIN) plays a vital role in atmospheric and cryobiological processes, yet the influence of surface topology remains poorly understood due to its coupling with surface chemistry. In this study, we employ the Phase-Field Model and Interfacial Dynamics to investigate how surface texture—represented by planar surfaces decorated with one or more cylindrical pillars—affects HIN. Contrary to conventional expectations, ice nuclei rarely form axisymmetrically around the pillars; instead, asymmetric structures are typically more stable. To interpret these observations, we derive analytical expressions that reduce the original three-dimensional problem to an equivalent two-dimensional formulation, enabling clearer physical insight. Furthermore, for single-pillar systems, we find that several system properties, including the critical temperature for heterogeneous nucleation, exhibit a sigmoidal dependence symmetric with respect to log r (the pillar radius). These findings establish a foundation for understanding how complex surface topologies influence ice nucleation and suggest the existence of underlying mathematical principles governing such symmetry.

13. Towards Biophysics-Driven Digital Twins for Personalized Immunotherapy

Stephanie Monson, Ravi Radhakrishnan, Bomyi Lim, Jina Ko

Digital twins are data-driven virtual models of real-world systems used to analyze and predict system performance in industries ranging from healthcare to manufacturing. In biomedicine, a digital twin of a patient’s tumor programmed with diagnostic data can be used to predict response to immunotherapy options. This prediction is otherwise challenging due to the heterogeneous and patient-specific nature of cancer. According to a 2023 National Academies report, an integral research requirement for digital twins is the development of models which are multiscale, mechanistic, and fit-for-purpose. Recent studies on cancer mechanisms have uncovered a complex, multiscale relationship between tumor-derived exosomes, immune modulation, tumor progression, and therapeutic resistance. Thus, an effective tumor digital twin must incorporate mechanistic models of these phenomena.

In my lightning talk and poster, I demonstrate how we are using multiscale physics to model and investigate this complex interplay in the tumor microenvironment. In recent work, we have elucidated how surface topology and immune modifications of a nanoparticle impact cellular recognition. These findings provide important insights for treatment planning and personalization. For instance, the present framework can be employed to understand disease mechanisms and inform optimal therapeutic design given patient biomarkers. Planned work includes integrating the cellular-level model with our signaling kinetics and tissue-level tumor models to study the impact on tumor progression. We will also perform verification, validation, and uncertainty quantification clinical data to enable well-informed, personalized decision-making.

14. — Abstract Retracted —

Xuelin Yang, De-en Jiang, Vikash Khokhar

15. Gas-Bubble Encapsulating Microcapsules (GEMs)

CK Yeh, Daeyeon Lee

Microcapsules that can respond under hydrostatic pressure would open a new avenue of application in ultrasound- or impact-induced release of therapeutic agents. While microcapsules that are designed to release cargo under uniaxial compressive loadings have been developed, they are filled with liquid, rendering them insensitive to hydrostatic pressure. To overcome this limitation, bubbles can be introduced into the core of the capsule that can impart hydrostatic pressure sensitivity. Although microcapsules that contain bubbles have been reported, the methods for their fabrication make it difficult to control the size of the bubble and microcapsule. In this work, we use microfluidics to create uniform microcapsules while also utilizing osmosis-induced-cavitation to convert microcapsules into gas-encapsulating microcapsules (GEMs). Following GEMs formation under high salt concentrations, the size of the bubble can be modulated by placing the GEMs back into relatively lower salt concentrations. GEMs with a well-defined thickness to diameter ratio (t/D) and volume fraction of gas are subjected to a known hydrostatic pressure through the means of a drop tower. These data establish a relationship between how the thickness to diameter ratio and volume fraction of bubble influence the rupture pressure and drug release, helping to establish the foundation for a new class of on-demand therapeutics that take advantage of pressure inputs.

16. Distribution of local active sites in the electrocatalytic behavior of multi-metallic 2D carbides

Haotian Long, Chinedum Osuji, Brian Cecil Wyatt, Babak Anasori

Self-assembly of block copolymers is a compelling way to create nanoscale patterns for new technologies, but precisely controlling how these patterns are organized remains a challenge. In this talk, I’ll introduce our approach to guide this spatially varying self-assembly using structured magnetic fields.

We use a ferromagnetic material to warp a uniform magnetic field, creating a non-uniform field that acts as a blueprint to direct the alignment of the block copolymer nanostructures. I will briefly show how we can tune this alignment by changing the field strength and the magnet’s geometry, with results that match our theoretical predictions.

This three-minute talk will give you the key highlights. To see our detailed results, simulations, and discuss how this strategy could be a generic way to structure a variety of soft materials, please visit my poster after the talks!

17. Quantifying Water Channels in Fluorine-Free Proton-Conducting Polymers

Lily Wang, Karen I. Winey, Stephen Kronenberger, Amalie L. Frischknecht, Arthi Jayaraman

Commercially available perfluorosulfonic acid polymers are commonly used as proton exchange membranes (PEMs) due to their favorable mechanical and electrochemical properties. Because of the high cost and toxicity associated with fluorine-containing polymers, alternative hydrocarbon-based PEMs are of interest. Well-defined nanoscale water channels lined with sulfonic acid groups are a key feature of these materials that we recently reproduced in a series of fluorine-free polymers. Previously, we used all-atom molecular dynamics (MD) simulations to reveal details about the water channels in a range of hydration levels. In this work, we quantitatively compare the nature of water channels found by MD with experimental small-angle X-ray scattering (SAXS) data. We reconstructed three-dimensional real-space nanoscale morphologies from SAXS data using Gaussian random fields. We developed methods to characterize the water channels within these reconstructions, specifically calculating the characteristic lengths, water clustering, tortuosity, channel width distributions, and fractal dimensions. Our results show good agreement between all-atom MD simulations, suggesting that this work can be used to reliably reconstruct and analyze nanoscale morphologies in phase-separated soft materials. 

18. Pendant Groups in the Majority Domain of Aliphatic Multiblock Copolymers Prohibit the Double Gyroid Morphology

Margaret Brown, Karen I. Winey, Viola A. Burlein, Benjamin T. Ferko, Anne Saumer, Stefan Mecking

The double gyroid morphology (DG) is associated with improved performance such as enhanced mechanical properties and ionic conductivities. We previously observed this morphology in strictly alternating multiblock copolymers comprised of a short sodium-sulfonated polar block containing 4 carbons and a linear alkyl block of precisely 12 to 23 carbons (PES4Nax). These materials crystallize into layers at room temperature and transition to the DG upon melting, suggesting that the crystallization of the alkyl blocks prevents the DG from forming. In this work, multiblock copolymers (PES4Nax–R) were synthesized to disrupt crystallinity by either (1) adding an isopropyl group to the center of each alkyl block or (2) replacing 20% of the linear alkyl blocks with branched alkyls. These modifications to the alkyl block successfully reduce or prevent crystallization. In PES4Nax, the DG was observed above Tm when the volume fraction of the polar block ranged from 0.27 to 0.41. Although the PES4Nax–R materials are in the same composition range (0.29 to 0.39), X-ray scattering experiments show the DG is absent. We interpret this result by considering the amount of chain stretching required to form the DG using a medial packing model. Compared to PES4Nax, PES4Nax–R significantly increases the degree of chain stretching required to form the DG. These results highlight the importance of chain stretching in the majority domain on the stability of the DG in multiblock copolymers.

19. Morphological Impacts and Diffusive Properties of Organic Solvents in Self-Assembled Nanoporous Membranes

Jack Granite, Chinedum Osuji, Kathleen Stebe

Membrane technologies hold promise to achieve highly selective separations, reducing energy costs in some cases by a factor of 2–3 compared to thermal methods. Conventional polymer membranes suffer from tradeoffs in their permeance (flux normalized by operating pressure) and selectivity due to broad distributions in pores sizes, which limits their industrial implementation. Self-assembled membranes address this issue− they have monodisperse pores of one transport limiting dimension due to the governance of pore formation by the thermodynamics of self-assembly. To date, these novel membranes have been heavily explored for their capability in aqueous nanofiltration and ion separations, and have shown promising performance compared to conventional polymer membranes. However, organic solvent separations remain mostly unexplored in these systems. In this work, we reveal the critical role of polymerization sites in amphiphilic monomer design and highlight key materials design principles for stability of these membranes in industrially relevant organic solvents. Additionally, we investigate the sorption of organic species in these membranes and show that uptake depends on solvent polarity, since these membranes are charged. Finally, through simple h-cell experiments coupled with gas chromatography, we elucidate the diffusive properties of organic species in these membranes and highlight selective separations based on both size-exclusion and chemical affinity