Chemical and Biological Engineering, 1931-2026

Recurring pelagic Sargassum overblooms impose substantial environmental and economic burdens on coastal communities across the Atlantic. Existing operations already recover a profitable liquid biostimulant from the pressate fraction of mechanically processed Sargassum. The pressed cake solid fraction remaining after biostimulant extraction, however, constitutes a significant waste stream with associated disposal costs. This thesis presents the first BioSTEAM-based techno-economic model evaluating whether this solid waste stream can be valorized through anaerobic digestion and downstream fermentation, simulating a floating platform processing 15,000 wet metric tonnes per day across two parallel pathways: a methanogenic route producing biomethane, and an acidogenic route producing volatile fatty acids subsequently fermented by Yarrowia lipolytica into crude microbial oil. Both standalone pathways proved economically unfavorable under base-case assumptions. The best-performing methanogenic case using combined peroxide-enzymatic pretreatment yielded a biomethane MSP of $30.92/MMBtu, more than twice the highest regional market reference price. The acidogenic pathway achieved a base-case microbial oil MSP of $15.99/kg, well above the $1.50/kg soybean oil benchmark. In the integrated model, full allocation to the methanogenic route consistently maximized NPV, with positive returns achievable only when feedstock tipping fees, biostimulant revenue, or access to higher-value biomethane markets were included. These results suggest that the pressed cake waste stream can contribute towards economic revenue, but biological performance improvements and waste-management framing of the feedstock remain critical enablers of economic viability.

The precise manipulation of microscale particles and polymer chains in three dimensions remains a central challenge in soft matter engineering and microfluidic control. In this work, a computational framework is developed that extends the hydrodynamic trapping paradigm of the Stokes trap to three dimensions, enabling flow based manipulation of particles and polymer chains. Control inputs are determined through Model Predictive Control (MPC), with the flow field modeled using a Green's function solution to the incompressible Stokes equations. Polymer dynamics are represented by a bead--spring model with finitely extensible nonlinear elastic bonds and steric repulsion.

A key feature of the framework is the use of a simplified predictor within the MPC optimization that accounts only for the imposed flow field and neglects interparticle forces. This introduces a systematic model mismatch between the predicted and true dynamics. Despite this discrepancy, the controller successfully achieves target configurations across a range of tasks, including single particle transport, coordinated multi--particle motion, and the formation of structured polymer geometries such as letters and helices.

The results suggest that accurate control does not require a high fidelity model of the underlying physics. Instead, performance is governed by the ability of the receding horizon architecture to correct prediction errors through repeated feedback. However, as particles approach contact and strong repulsive interactions dominate, the simplified predictor cannot capture sufficient dynamics for convergence. More broadly, these findings indicate that model fidelity can be traded for frequent feedback, provided prediction errors remain bounded over short time intervals. This establishes a scalable extension of the Stokes trap framework to three-dimensional, multi particle, and polymer systems, and identifies the regimes in which simplified predictive models remain effective.

Plastic waste is one of the most pressing environmental challenges of the modern era, yet current waste management strategies are unable to effectively address the rapidly increasing plastic production at scale. Catalytic upcycling of polyolefins over zeolite catalysts provides a promising pathway to mitigate this problem and contribute to a circular economy, but a quantitative understanding of early-stage reaction kinetics is underexplored in the literature. This thesis employs a microreactor-based methodology to investigate the initial kinetics of polyethylene upcycling over zeolites BEA-12.5, FAU-15 and MFI-40 across polymer lengths of 0.5 kDa, 3 kDa, and 4 kDa. A first-order kinetic model with exponential decay was fit to time-resolved solid conversion data to allow for the extraction of rate constants k and deactivation constants kd. Across molecular weights, BEA12.5 achieved the highest raw rate constant, followed by FAU-15 and MFI-40, consistent with the relative framework pore accessibility. Rate constants decreased monotonically with increasing polymer length across frameworks, confirming that polymer chain length is an intrinsic kinetic variable. Normalization by surface Brønsted acid site (SBAS) density inverted framework ordering, with MFI-40 achieving the highest per-site rate constant, suggesting its surface sites are more productive than the other frameworks on a per-site basis. Temperature-dependent experiments with BEA-12.5 and 4 kDa polyethylene at 423 K, 473 K and 523 K yielded an apparent activation energy of Ea = 40.4 kJ / mol. Together, these results offer a kinetically grounded understanding of structure-activity relationships in polyolefin upcycling and provide a foundation for further kinetic analysis, and, ultimately, the rational design of more efficient upcycling systems.

Mesogens are molecules that can access anisotropic, liquid crystalline mesophases with useful optical, magnetic, and electronic properties. Mesogens with a bent shape have restricted rotational freedom along their long axis. Because of this restriction, these mesogens can form a range of low symmetry, smectic ‘Banana’ or ‘B-phases’. Also, under certain conditions, oligomers and dimers synthesized with flexible linkers can adopt a bent intermesogen geometry, allowing them to form B-phases as ‘flexible’ bent core mesogens. In rigid bent-core mesogens and dimers synthesized with a flexible linker, the formation of complex microstructures is triggered both by geometric frustrations from the packing arrangement and by a driving force to escape macroscopic polar order. These features give rise to assemblies with substantial curvature and sometimes supramolecular chirality, even in the absence of point chirality within the chemical structure of the mesogen. Because of their chirality, these specialized microstructures have drawn attention for their potential applications in synthesis and nanofabrication efforts in bio- sensors, optomechanics, and as templates for nanoparticle systems. This work examines the self-assembly of an achiral, asymmetric mesogenic heterodimer which is hypothesized to favor a bent intermesogen geometry. The heterodimer self-assembles into nanocylinders 128 ± 21 nm in diameter. Despite the achirality of the mesogenic subunits themselves, their hierarchical nanocylinder arrangement features planar chirality. To reveal the precise structure of the molecular assembly and affirm the potential of this system for further functionalization, copper-catalyzed azide-alkyne cycloaddition (CuAAC) fluorescent tagging procedures are used to visualize the placement of key functional groups within the nanocylinder assembly. We demonstrate that we can achieve a high degree of long-range order and alignment in the nanocylinder microstructure through seeded nucleation followed by slow, high-temperature isothermal crystallization. These findings explore methods to prescribe nucleation sites and direct the subsequent growth of the nanocylinder microstructures, which is an important step towards exploiting these cylindrical assemblies in device fabrication.

In humans, the composition and abundance of microbial species that reside inside the intestinal tract– termed the gut microbiome– plays an integral role in health. These microbes produce essential nutrients, modulate metabolic pathways, and promote immune system function [1, 2]. As such, gut microbiome dysbiosis– a disturbance from the homeostatic microbiome– is associated with various health complications, such as Irritable Bowel Syndrome, obesity, and Type-II Diabetes [3]. Beyond modulating physical health, the features of the gut microbiome also impact neurological activity and cognitive function via a bi-directional signaling pathway known as the gut-brain axis. Over the past decade, studies of this axis have revealed correlations between microbiome composition and the progression of neurodegenerative diseases such as Alzheimer’s and Parkinson’s, as well as mental illnesses such as anxiety and depression [4]. Here, we use the common fruit fly, Drosophila melanogaster, as a model organism for studying the gut-brain axis. With the aim of not only assessing the impact of the gut microbiome on stereotyped fly behaviors, but also exploring its effect on Drosophila circadian rhythms, we collect and analyze data over a 24-hour imaging window. We utilize high-resolution pose tracking and subsequent behavioral mapping to extract information regarding six stereotyped behaviors: locomotion, fore grooming, hind grooming, wing grooming, proboscis extension, and idle. From this data, we find that the daytime activity level and overall behavioral state of non-virgin female D. melanogaster is less sensitive to microbiome perturbations than that of non-virgin male Drosophila. Among male populations, we observe increased daytime activity levels following antibiotic treatment, which are rescued via monocolonization with Lactobacillus brevis and Lactobacillus plantarum. We also find that the circadian rhythms of female D. melanogaster are stabilized by the presence of a gut microbiome, as, upon microbiome removal, their night-day behavioral difference becomes more defined. This trend is not observed in male populations. Broadly, this project motivates the development and implementation of novel treatment methods for neurological disorders through high-resolution behavioral analysis of the Drosophila melanogaster gut-brain axis.

Semicrystalline polymer networks were synthesized from a butadiene-derived cyclobutene oligomer (B4P1) via thiol-ene photopolymerization to investigate how crosslink density and curing temperature jointly control crystallinity and thermomechanical properties. Networks were formulated across four PETMP:B4P1 molar ratios (z = 0.1–0.4), yielding crosslink densities ranging from 1.25 to 5.00 B4P1 units per junction, and cured either at room temperature (25 °C) or above the networks’ melting point (120 °C). Gel fraction analysis confirmed that higher PETMP concentrations produced more complete network formation, while the lowest crosslinker ratio (z = 0.1) exhibited substantial swelling and network defects. Differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) revealed that increasing crosslink density increases both the glass transition temperature and thermal degradation onset, and that polymerizing the network at 120 °C significantly reduces crystallinity when compared to the room temperature synthesized networks. Dynamic mechanical analysis (DMA) demonstrated that room-temperature networks retain higher rubbery-plateau storage moduli, consistent with trapped crystalline reinforcement, and tan ω profiles suggest a broad crystal-to-rotator phase transition between 50–80 °C. Variable-temperature wide-angle X-ray scattering (VT-WAXS) corroborated the thermal melting behavior observed by DSC. Tensile testing confirmed that room temperature networks exhibit greater stiffness and yield strength, while heated networks display enhanced ductility at higher crosslink densities. Finally, preliminary shape memory experiments demonstrated repeatable thermal programming and recovery, establishing B4P1-based thiol-ene networks as a promising platform for stimuli-responsive smart materials.

Ribosome biogenesis is a fundamental process to cellular function that generates the ribosomes required for protein synthesis. This pathway involves the transcription, processing, maturation, and assembly of ribosomal RNA that ultimately leads to the formation of the ribosomal subunits. Because it is complex, resource-intensive, and energetically demanding, it is normally tightly regulated. In cancer cells, however, oncogenic signaling increases ribosome biogenesis to support rapid cell growth. In this study, we found that although oncogene activation significantly increased nucleolar transcriptional activity, nucleolar rRNA outflux remained largely similar between oncogene-hyperactive and normal states, suggesting that increased transcription alone did not explain how efficiently material progressed through the pathway. This raised the question of whether downstream processing and maturation steps were able to keep pace under oncogenic stress. By comparing newly synthesized pre-rRNA with newly produced ribosomes, we found that oncogene activation increases transcriptional input without producing a proportional increase in mature ribosome output, indicating reduced ribosome biogenesis efficiency. Pulse-chase sequencing following rRNA processing over time revealed that malignant cells progress more slowly specifically through late-stage processing steps. Perturbing factors involved in these late maturation stages selectively reduced ribosome biogenesis in oncogene-hyperactive cells, both in vitro and in vivo. These findings suggest that elevated ribosome biogenesis in cancer is limited by downstream processing capacity, revealing late-stage bottlenecks as selective vulnerabilities that can potentially be utilized as targets for future therapeutics.

Multilayer polymer packaging relies on polyethylene (PE) layers bonded by ethylenemethacrylic acid (EMAA) copolymer tie layers, but the adhesion mechanisms governing interfacial strength are not fully understood. This thesis presents a systematic peel strength study across EMAA compositions varying from 5 to 22 wt% methacrylic acid (MAA) content and sodium neutralization levels from unneutralized to 83%, paired with both low-density polyethylene (LDPE) and high-density polyethylene (HDPE) substrates under two thermal histories (aircooled and quenched). Peel strength decreases monotonically with increasing MAA content for both PE substrates, consistent with increasing polarity mismatch reducing interfacial cocrystallization. A partial recovery in peel strength at 22 wt% MAA is attributed to the glass transition temperature of the EMAA amorphous phase rising through room temperature, producing a mechanically asymmetric bilayer where energy dissipation during peeling increases. Sodium neutralization uniformly reduces adhesion relative to unneutralized EMAA, with an abrupt drop upon introduction of any ionic crosslinks attributed to conformational entropy penalties imposed by ionic aggregates on interfacial chain segments. Aircooled samples consistently exhibit higher peel strength than quenched samples, reflecting more extensive interfacial co-crystallization during slow cooling, while quenched samples provide greater reproducibility for isolating compositional effects. These results establish quantitative, reproducible relationships between EMAA composition, processing conditions, and bilayer adhesion, offering practical guidance for designing multilayer packaging systems with controlled adhesion for both mechanical performance and end-of-life recyclability.

In the last century, rising atmospheric carbon dioxide (CO2) levels have demanded carbon removal technologies in addition to simple emissions cuts. Among these negative emission technologies is direct air capture (DAC), which extracts CO2 directly from ambient air. While DAC is a promising negative emission technology, it is limited by energy-intensive sorbent regeneration that typically involves high temperature or pressure swings. This limitation prompts interest in moisture-swing adsorption (MSA), which relies on changes in humidity to regenerate sorbent. In a moisture-swing DAC system, CO2 is adsorbed at low humidity and desorbed at high humidity. Tangentially, polyethylene terephthalate (PET) plastic is one of the most produced and under-recycled plastics globally, single-handedly contributing to both emissions and waste problems. This work proposes a chemical synthesis process to upcycle post-consumer PET bottles into a quaternary ammonium bicarbonate sorbent for moisture-swing DAC. Functionalization occurs via a four-step reaction sequence: aminolysis, Eschweiler-Clarke methylation, Menshutkin quaternization, and bicarbonate ion exchange. Successful conversion of the products is confirmed at each step via 1H NMR spectroscopy. The carbon capture performance of the upcycled PET-derived sorbent is then evaluated in the moisture-swing DAC system. The sorbent demonstrates reversible CO2 capture and release over five adsorption/desorption cycles. CO2 levels correlated directly with humidity swings from 35% relative humidity (dry) to 70% relative humidity (wet), consistent with the MSA mechanism. In addition, cyclical stability was observed with no measurable capacity loss. Thus, this research serves as a proof of concept for a negative emissions process that simultaneously addresses plastic waste and atmospheric CO2 mitigation.

Current modes of chemical commodity production have brought up a vast array of economic, environmental, and sustainability issues or concerns. However, to bypass many of these current issues of chemical production, microbial biosynthesis that utilizes inexpensive and renewable feedstocks are being studied for production of chemical commodities instead. In previous studies, biosynthetic pathways were discovered that could be introduced into E. coli to produce phenylpropanoids, particularly eugenol, from a glucose and glycerol feed. Modularization of this pathway, meaning breaking up the pathway across multiple strains of E. coli, and allowing bioproduction to occur under this tripartite E. coli coculture found significantly higher yields of eugenol relative to their respective monoculture. To improve upon this discovery of utilizing cocultures to optimize phenylpropanoid production, optogenetics was applied to this consortia system to determine if further optimizations could be accomplished by controlling the growth of each E. coli strain under their respective light conditions. While more experimentation needs to be conducted in order to find optimal light conditions that yield maximal phenylpropanoid titers, there is currently promising data that utilizing optogenetics in these co-cultures can further optimize phenylpropanoid production relative to the control co-culture system. Overall, pending further research, promising data is present towards the applicability of optogenetics in furthering microbial biosynthesis as a sustainable alternative in the production of chemical commodities like phenylpropanoids.

Avian and mammalian lungs share signaling pathways during development and both consist of epithelial tissue in the small airways and alveoli, embedded in mesenchyme. This thesis examines how either oligomycin or blebbistatin treatment affects the stiffness (elastic moduli) of embryonic day 5 chicken lung tissue. Oligomycin treatment was expected to increase elastic moduli, as the drug’s inhibition of oxidative phosphorylation can increase production of reactive oxygen species, promote the expression of extracellular matrix proteins, and cause local inflammation. In contrast, blebbistatin treatment was anticipated to decrease elastic moduli, because its NM II inhibition can weaken cell-cell adhesion and lead to incorrect mitotic spindle orientation during mitosis, increasing tissue permeability. Published experiments have also found that blebbistatin treatment reduced F-actin levels, hindered apical constriction, and prevented the emergence of new secondary bronchi. Combined nanoindentation results from multiple trials testing each treatment found no significant difference in stiffness between conditions. However, oligomycin-treated explants were significantly stiffer than control explants in two of four trials, while control explants were stiffer than blebbistatin-treated explants in one of three trials. As a supplement to stiffness measurements, immunofluorescence stained blebbistatin-treated explants had significantly lower mean levels of nuclei, F-actin, and fibronectin per focal plane, as well as significantly lower maximum levels of F-actin, E-cadherin, and fibronectin per focal plane compared to control explants. Additionally, image analysis before and after blebbistatin incubation revealed that significantly fewer secondary bronchi emerged in blebbistatin-treated explants than in control explants, aligning with previous experiments. This study can serve to enhance understanding of pulmonary compliance in disease states, such as stiffening in Acute Respiratory Distress Syndrome or excessive compliance in COPD. This understanding can be applied to the design and implementation of medical ventilation devices to reduce the risk of lung injury and lower the annual number of premature deaths—over 4 million as of 2024—in the United States linked to lung diseases.

The development of cancer leads to dysregulated metabolic fluxes in the human body. Comprehending the altered metabolism in a cancerous microenvironment is beneficial for trying to find cell-type specific vulnerabilities that enable researchers to create new therapeutics that target tumorous cells while sparing non-tumorous ones. Thus far, studying metabolism at a single cell resolution has not been feasible. This engineering thesis aims to solve this issue through the usage of MALDI imaging mass spectrometry (IMS) and Xenium to inspect, at a near single cell-adjacent resolution (10 μm), the distribution of metabolites and the expression of metabolic genes in two epithelial cancer models, pancreatic adenocarcinoma and HR+/HER2− breast cancer. The observation and quantification of the Warburg effect through both MALDI IMS and Xenium, as well as the discovery of specific phospholipids that co-located with cancerous cells in both carcinomas was achieved. Similarly, the overall conservation of central carbon metabolism across the two epithelial cancers, irrespective of their origin, was noted, posing the possibility of finding more generalized metabolic vulnerabilities in different carcinomas. Although Xenium revealed that cancerous cells are the most transcriptionally active, further research into the normalization, control-quality and cellular segmentation techniques of this spatial technology needs to be done.

Triple negative breast cancer is an aggressive subtype of breast cancer, and its cancer progression is influenced by various microenvironmental factors, including diet and fluid shear stress. While dietary conditions like high-fat diets have been shown to influence cell metabolism, research on their interaction with physical forces such as flow remains limited. This project aims to investigate how dietary conditions affect breast cancer morphology under varying flow conditions in a microfluidic platform. MDA-MB- 231 cells were cultured under baseline and high-fat conditions and were seeded into a microfluidic chip, which ran varying flow rates using a monitored syringe pump. Images of cells in the chip were captured under a microscope for cell morphology analysis. Cell characteristics, such as cell roundness and area, were analyzed to evaluate morphological changes between both varying flow rates and the two dietary conditions. In both conditions, increasing flow rate increased the cell roundness, with more significance in the higher flow rates with increased shear stress. The effect of flow rate was more pronounced in the high-fat condition, where significant increases in cell roundness was observed at lower flow rates compared to baseline. Cell area also showed a similar pattern, with high-fat condition having a significant change in median cell area at a high flow rate, which was not seen in the baseline condition. Both findings suggest that cells cultured under high-fat conditions are more likely to be sensitive to shear stress, inducing morphological changes related to adhesion, which in turn may impact migration and cancer progression. This highlights the importance of studying cancer in the context of both metabolic and mechanical factors.

Consolidated Bioprocessing (CBP) represents a promising platform for lignocellulosic biofuel production, yet its kinetic efficiency is fundamentally constrained by the structural recalcitrance of highly branched hemicellulose networks. To overcome these depolymerization bottlenecks, this study investigates the biochemical boundaries and cooperative dynamics of novel cellulosomal glycoside hydrolases derived from the thermophilic bacterium Acetivibrio thermocellus. Specifically, three distinct multi-modular architectures were selected for functional characterization: Clo1313 2857 (a GH43 10 α-L-arabinofuranosidase), Clo1313 2216 (a GH43 4 endo-α-1,5-L-arabinanase), and Clo1313 2795 (a multidomain GH30/GH43 20 construct). Recombinant targets were successfully expressed in Escherichia coli and isolated via immobilized metal affinity chromatography. Extensive profiling established isolated operating windows, with the GH43 catalytic cores exhibiting robust thermophilic activity between 60 ◦C and 70 ◦C at mildly acidic to neutral pH (5.0–7.0). Substrate specificity and depolymerization modalities were subsequently validated through colorimetric reducing-sugar assays and High-Performance Liquid Chromatography (HPLC). Clo1313 2857 demonstrated strict exo-debranching activity on wheat arabinoxylan (WAX), yielding exclusive L-arabinose monomers. Clo1313 2216 exhibited targeted endo-acting cleavage on sugar beet arabinan (SBA), accumulating higher-order oligosaccharide intermediates and displaying a pronounced non-linear kinetic lag phase at low enzyme concentrations (<0.015 μM). Building upon these isolated kinetic parameters, the cooperative dynamics of the hemicellulases were evaluated through systematic combinatorial screening. Combinatorial assays on SBA identified a highly active synergistic pairing between the endo-arabinanase Clo1313 2216 and the exo-acting AtAbf43C (Clo1313 2794), yielding a 1.91-fold increase in specific activity (U/mg). Further stoichiometric validation via the method of continuous variation defined a mathematically optimal 2 μmol Clo1313 2216: 3 μmol AtAbf43C formulation. Kinetic analysis confirmed that this synergy is driven by mechanistic compensation. The exo-enzyme cleaves the oligosaccharides generated by the endo-enzyme, alleviating the non-productive lag phase and yielding a 15-fold recovery in specific activity at low concentrations of Clo1313 2216. Ultimately, the precise biochemical characterization of these domains, coupled with empirical evidence of stoichiometric synergism, provides a foundation for engineering architectures necessary for robust monomer release during CBP fermentation.

The chemical termination of the diamond (100) surface plays a critical role in determining the performance of near-surface nitrogen-vacancy (NV) centers, which are promising platforms for emerging quantum sensing and computing technology. Oxygen-termination of the diamond (100) surface produces favorable conditions for NV center applications. On this surface, carbonyl (C=O) and (C–O–C) terminations coexist. The carbonyl termination likely introduces magnetic noise and is an undesirable termination. These two states’ relative stability at finite temperatures and near defects remains unclear. In this work, molecular dynamics (MD) simulations with enhanced sampling techniques, including on-the-fly probability enhanced sampling (OPES), are used to construct free-energy surfaces for monolayer coverage of oxygen on diamond (100). For ideal surfaces, ether is favored at low temperatures, with carbonyl becoming increasingly stable as temperature rises. Free-energy differences between states predict approximately 90% ether coverage at 300 K, consistent with prior studies. In contrast, substitution defects locally favor carbonyl at 600 K and perturb oxygen bonding environments over distances exceeding 12 ˚A.

Antisense oligonucleotides (ASOs) are an established therapeutic modality with a growing preclinical pipeline. Translating a transcript into a viable ASO, however, requires more than sequence complementarity: mRNA folds into secondary and tertiary structures that render nominally complementary sites base-paired, buried, or only transiently accessible. Existing in silico tools capture secondary structure or single static three-dimensional folds but do not report time-dependent accessibility or binding behaviour in the presence of an ASO. This thesis develops a coarse-grained molecular dynamics (CGMD) pipeline that narrows these gaps. Target structures are obtained from the Protein Data Bank or predicted with NuFold and simulated in LAMMPS under the single-interaction-site (SIS) RNA force field of Nguyen, Hori, and Thirumalai [1] with the folding and binding implementation of Aierken and Joseph [2]. Candidate sites are identified by intersecting secondary-structure single-strandedness with per-residue solvent-accessible surface area, and ASO–RNA association is scored from the trajectory as a comparative apparent dissociation constant under a centre-of-mass distance framework. The pipeline is applied to a 1YMO-derived hTR pseudoknot hairpin and extended, without per-target retuning, to six additional PDB-derived RNAs. Across ten ASO variants at fixed loading, apparent Kd spans roughly an order of magnitude; extended-complementarity designs register the highest apparent Kd despite suppressing loop-core fluctuation most strongly, and both loop-core RMSF and apparent Kd depend non-monotonically on ASO loading. The integrated ranking reorders candidates relative to a complementarity-only shortlist—the pipeline’s principal contribution. Apparent Kd is reported as a comparative rather than an absolute quantity, several comparisons are single-replicate, and cross-target comparisons use differing binding criteria and box volumes; these caveats attach to the absolute values, not to the reordering itself.

CRISPR-based diagnostics (CRISPR-Dx) are a growing technology for sensitive and specific nucleic acid detection without the accessibility restrictions of qPCR. However, there is currently no platform that achieves highly multiplexed CRISPR-Dx without the need for specialized equipment, preventing efficient deployment in point-of-care or point-of-need settings. Here, we make significant progress toward addressing this gap in the field: we design, develop, and test several iterations of a microfluidic chip capable of 10-plex CRISPR-Dx with lateral flow (LF) integrated for equipment-free result readout. We confirm the efficacy of our device in facilitating stop-and-start fluid flow and on-chip mixing in colorimetric flow control tests. To make our device compatible with biological diagnostics, we also develop novel processes for lyophilized reagent storage and post-reaction fluid dilution. We make strides toward the validation of our device and workflow in a CRISPRDx setting, successfully detecting synthetic tuberculosis (TB) and monkeypox (Mpox) DNA targets to achieve a new level of combined multiplexing and equipment-free accessibility in CRISPR-Dx.

This study evaluates how the representation of thermal generator start-up and shutdown behavior in capacity expansion modeling influences future power system investment decisions. Greater renewable energy penetration has increased operational variability in power systems because renewable output is not fully controllable, requiring generators to respond to more dynamic conditions. Although the transitional behavior of thermal generators is often simplified in capacity expansion models, the variability introduced by renewable resources calls for an examination of whether this simplified representation remains adequate for accurate grid planning. To address this question, a deterministic mixed-integer linear programming model is developed that simultaneously solves capacity expansion, unit commitment, and economic dispatch problems to minimize the total cost of a small utility-scale power system over a 20-year planning horizon. A baseline model formulation with simplified start-up and shutdown representation is first constructed, followed by a detailed transition formulation that incorporates path-indexed, multi-hour start-up and shutdown trajectories for thermal generating technologies. Comparing the two formulations isolates the effect of increased transition resolution on investment decisions, dispatch outcomes, cycling behavior, carbon emissions, and total system cost. The results show that the detailed transition formulation remains broadly consistent enough to be a suitable substitute for the baseline model, but it produces meaningful differences in operational and planning outcomes. In particular, the baseline formulation appears to overstate thermal flexibility, underestimate operational friction, and distort the resulting investment portfolio when compared with the detailed formulation. Overall, the findings demonstrate that start-up and shutdown behavior is an important modeling choice in capacity expansion analysis, and that greater transition detail can improve the credibility of long-term power system planning results.

A large fraction of the active pharmaceutical ingredients in drug development pipelines suffer from low solubility in aqueous solutions. Addressing this issue and improving drug bioavailability for patients requires innovative materials and chemical formulations. Amorphous solid dispersions (ASDs) have emerged as one of these innovative materials, where the insoluble drug is dispersed in its amorphous form within a polymer matrix. ASDs have been effectively utilized in industry for drug delivery. However, there is a lack of understanding of how polymer and drug chemistry affects the ASD dissolution process, which hinders the design and formulation of ASDs. Previous studies have narrowly focused on specific polymer-drug combinations. On a broader scale, the field lacks a phenomenological model for how varying the interactions between ASD components affects dissolution. Therefore, in this work, we have aimed to provide such a model through molecular dynamics simulations in a semi-infinite slab geometry, where dissolution occurs along the long axis. We built the system using a modified Lennard-Jones potential called the Ashbaugh-Hatch potential. In our simulations, we varied a large number of Lennard-Jones interaction strength parameters (ε values) and length scale parameters (σ values). We found that size mismatch between drug molecules and monomers greatly hinders drug transport. We also found that polymer hydrophilicity and drug size were the main factors governing polymer and drug transport properties, drug release, and water penetration into the ASD. This study serves as a guide for which parameters are most important to tune during ASD development. We offer a qualitative perspective on how polymer hydrophilicity and drug size, along with other less critical parameters, should be chosen to achieve desired ASD dissolution behavior.

Expansion microscopy (ExM) assumes biological specimens and the hydrogel expand isotropically, but distortions are often observed and are typically corrected in two dimensions, without addressing their three-dimensional structure or physical origin. Here, I treat the hydrogel as an independent material system and develop a quantitative framework to model its intrinsic deformation during expansion in 3D. Photoactivatable fluorescent fiducial beads were embedded throughout polyacrylamide gels and imaged pre- and post-expansion. I built a 3D registration pipeline based on polyharmonic spline modeling to reconstruct the full volumetric deformation field. Unlike existing slice-by-slice 2D correction methods, this framework captures out-of-plane distortions invisible to planar analysis and yields a continuous mathematical description of the deformation field. The results show that non-uniform expansion arises from intrinsic hydrogel properties rather than biological sample effects, allowing material-driven distortion to be quantitatively separated from specimen-specific contributions. Quantifying this baseline behavior enables separation of material-driven distortion from specimen-specific contributions and provides a framework for engineering ExM protocols with uniform expansion.