Search

The environmental remediation of water and air sources is a challenging field, due to widespread contamination, pollutant variety, and efficiency of existing treatment options. Remediation of certain pollutants, such as per- and polyfluorinated chemicals (PFAS), polychlorinated biphenyls (PCBs), volatile organic compounds (VOCs), and viral aerosols, has shown promise via adsorptive, degradative, or filtration-based processes, each with their respective advantages and limitations. Membrane technology is one option that has exhibited considerable pollutant removal from environmental sources via reverse osmosis (RO) filtration, but is limited by low permeabilities and lack of pollutant detoxification. With the support of NIEHS and NSF, the functionalization of microfiltration (MF) membrane systems with high water/air permeabilities was investigated to increase remediation efficiency, decrease operation cost, and create a flexible platform that could adapt to different pollutant environments. Depending on the intended pollutant, membranes were functionalized with stimuli-responsive polymers, biological components, and catalytic nanoparticles to incorporate adsorptive and/or degradative properties. For enhancing the remediation of perfluorooctanoic acid (PFOA), adsorptive capabilities were incorporated in flat-sheet polyvinylidene fluoride (PVDF) membranes via functionalization with poly-N-isopropylacrylamide (PNIPAm), a temperature-responsive polymer that can exhibit both hydrophilic and hydrophobic properties below and above its lower critical solution temperature (LCST) of 32°C, respectively. Over 5 cycles of convective flow, PNIPAm-functionalized PVDF membranes showed stable PFOA adsorption of ~1800 μg/m2 of membrane above the LCST (hydrophobic expression) with 50-60% desorption of PFOA below the LCST (hydrophilic expression), displaying the temperature swing adsorption/desorption capabilities of the membrane/polymer matrix. Membrane systems with bimetallic nanoparticle catalysts, which are limited by low mass transfer between the pollutant and catalyst active site, benefited from the incorporation of stimuli-responsive polymer hydrogels as well. The addition of PNIPAm into a Fe/Pd-functionalized membrane matrix allowed for increased interaction between the pollutant, such as PCB-1, and catalyst site above the LCST, resulting in first-order kSA values of 0.28 and 1.36 L/m2/g at 25 ºC and 45 ºC, respectively. This functionalization was additionally applied to highly-scalable hollow fiber membranes, which yielded a 30% and 420% increase in kSA above 32 °C for the dye methyl orange and the VOC trichloroethylene, respectively. As a cost-effective option for increasing degradation rates, precise heating of the catalytic membrane domain was also studied via the palladium nanoparticle’s plasmonic properties. This work extended to viral aerosols by utilizing enzyme-functionalized membranes for the denaturation of SARS-CoV-2 spike glycoproteins. PVDF membranes provided a protection factor of 540 ± 380 for coronavirus-sized particle, above the Occupational Safety and Health Administration’s standard of 10 for N95 masks. SARS-CoV-2 spike glycoprotein on the surface of virus-sized particles was denatured in 30 s by subtilisin enzyme-functionalized membranes with 0.02-0.2% water content on the membrane surface. For the environmental remediation of both water and air, stimuli-responsive functionalized membranes exhibit a promising field for high-efficiency capture and detoxification of harmful pollutants.

As the development of energy storage technologies for renewable energy sources continues to garner attention, vanadium redox flow batteries (VRFBs) have emerged as promising source for grid-scale energy storage. A major component of the VRFB is the ion exchange membrane (IEM), as it serves to physically separate the two liquid electrolytes and also directly impacts the battery’s capacity, efficiency, and cost. The current industry standard IEM, Nafion™, not only accounts for a large portion of the cost of the VRFB but also suffers from poor proton selectivity. Herein, we have synthesized sulfonated poly(ether ether ketone) (SPEEK) composite membranes containing amine functionalized softwood Kraft lignin. The lignin amination was performed via epoxidation of the hydroxyl groups of the lignin, followed by amination. The addition of amine groups was verified using 31P, 1H, 13C Nuclear Magnetic Resonance (NMR), as well as Fourier-transform infrared spectroscopy (FTIR), as well as elemental analysis. After amination, the lignin was combined with SPEEK, and membranes containing 5 and 15 mass % aminated lignin were fabricated via solution casting. Notably, the aminated lignin composite membrane showed reduced vanadyl ion permeability compared to the neat SPEEK membrane. Transmission electron microscopy (TEM) showed a reduction in lignin aggregation and more uniform dispersion of the aminated lignin compared to its nonaminated counterpart. The ion exchange capacity (IEC) and equilibrium water uptake (EWU) for these ionomers were evaluated and compared to the neat SPEEK membrane, as well as SPEEK membranes containing unmodified lignin. Results from this analysis showed that ionomers containing aminated lignin exhibited lower IEC values compared to the neat SPEEK membrane but significantly higher EWU compared to unmodified lignin composite membrane. Combining the increase in the EWU, decrease in IEC, and the dispersion state of the lignin, the aminated lignin was postulated to interact with the sulfonic groups of the SPEEK altering membrane properties compared to the unmodified lignin sample. Overall, results from this work demonstrated how chemical functionalization of lignin can be used to tune the SPEEK–lignin composite membrane properties, providing a pathway forward for the development of more cost-effective ionomer composites containing polymer from renewable feedstocks.

Cell therapy is a technology that relies on replacing diseased or dysfunctional cells with healthy functioning ones. One of the cells used for such advanced therapies are stem cells, owing to their ability to differentiate into specific cells required for repairing damaged or defective tissues or cells. The majority of cell-based products are intended to transiently persist in the patient, secreting factors which then allow the patient’s body to heal; in these products, the cells are subsequently eliminated from the body. Furthermore, unique manufacturing platforms, in addition to novel commercialization strategies, will be required to create a successful, sustainable cell therapy industry. Currently in cell manufacturing companies, need to produce adherent cells, that bind to a solid surface such as tissue culture- treated plastic. In some cases, the growth surface needs to be treated or coated with a matrix (such as natural proteins) to facilitate cell adhesion. Cells can be grown on large flat surfaces or on spherical microcarriers for suspension?based culture in bioreactors. However, those proteins leading to batch-to-batch variability and concerns regarding contamination and affects the cell therapeutic potency. This thesis aims to develop techniques based on the merger of novel materials approaches to manipulate cell microenvironments in culture. For controlling cell-cell interaction and cell-soluble factor interactions, layer-by-layer deposition of ionic biopolymers were developed (Heparin as a negative charge polymer, collagen and Poly-L-Lysin are used as positive charge polymers). In addition, techniques were developed to control the viability of cells under hypoxic conditions within collagen hydrogels by controlling the three-dimensional properties of hydrogels and oxygen delivery (Perfluorocarbon-based oxygen carriers). In addition, to control cell-soluble factor interactions, Metal-Organic-Frameworks (MOFs) nanocarriers application was discussed as a carrier for delivering interferon-gamma (IFN-γ) to cells.

In a world where clean water predictability is challenged by both global warming and the contamination of our natural resources, it is our responsibility to advance water separation technologies into more efficient processes and to lessen their environmental footprint. Pore and surface functionalization of membranes can enhance the separation of ions from water and even help overcome intrinsic challenges that some current separation technologies possess. If we begin with a complete fundamental understanding of the nanoscale interactions that occur in the process of separating ions from water, then we can engineer new functionalized composite membranes to provide alternative processes keeping life cycle aspects, lowering energy consumption, and keeping minimum supply needs in consideration. This research evaluated the synthesis and performance of three new membrane platforms: a positively charged nanofiltration (NF) membrane, a dual-functional NF and adsorptive membrane, and a set of adsorptive/ion exchange membranes. The application of these technologies include: the recovery of valuable lanthanide ions (rare earth elements) from water, and the removal of organic and inorganic ions and pollutants, such as per- and polyfluoroalkyl substances (PFAS) and arsenic, among other ions being studied. A comprehensive study that involved the synthesis of the membrane technologies, thorough materials characterization, analysis of the separation performance, and modeling of the solute transport through these membranes, was performed for each one of these three materials/applications. First, positively charged nanofiltration membranes were created by adding commercially available polyallylamine hydrochloride (PAH, primary amine) to the conventional interfacial polymerization technique. The NF membranes with added PAH had high and stable lanthanides retention rates, with values around 99.3% in mixtures with high ionic strength (100 mM, equivalent to ~6000 ppm), 99.3% rejection at 85% water recovery (and high Na+/La3+ selectivity, with 0% Na+ rejection starting at 65% recovery), and both constant lanthanum rejection and permeate flux even at pH 2.7. The Donnan steric pore model with dielectric exclusion elucidated the transport mechanism of lanthanides and sodium, proving the potential of highly selective separation with acceptable water permeability using positively charged NF membranes. Second, the rejections of ionogenic compounds by nanofiltration (NF) membranes at pH values near the membrane isoelectric point were compared to the size-transfer-dependent and mass-transfer-dependent modeled rejection rates of these compounds in an ionized state. This provided new insight on the transport mechanism of PFAS through the NF layer. The synthesis of an additional support barrier with thermo-responsive adsorption/desorption properties (quantified by water permeance variation) was proposed. This allowed for enhanced separations of PFAS and preventing other compounds, such as NOM, to enter the backing adsorption domain. The synthesis was made possible by successfully adding an NF layer on top of a poly-N-isopropylacrylamide (PNIPAm) pore-functionalized microfiltration support structure. The support layer adsorbed organics (178 mg perfluorooctanoic acid (PFOA) adsorbed/m2), and the simultaneous exclusion (50-70%) from the NF layer allows separations at a high water flux of 100 L/m2-h at 7 bar. Finally, three types of adsorptive polymers comprised of strong/weak acid, strong base, and iron-chitosan complex groups were synthesized in the pores and on the surface of microfiltration (MF) membranes. These were tested for removal of organic and inorganic cations and anions from water, including arsenic, per- and polyfluoroalkyl substances (PFAS), and calcium (hardness). An increase in the relative adsorption capacity by up to eight times when compared to beads and crushed-down resins, as well as up to two orders of magnitude faster adsorption kinetics were found. This research includes an analysis of the simple synthesis approach, consideration for how to apply the results of this research on a larger scale, and comparison against commercial adsorptive/ion exchange technologies, ultimately quantifying the gap in adsorption capacities between adsorptive membranes and adsorptive resins.

Membrane fouling and sustainable production of membranes are two of the most important challenges in membrane science. Chapter 1 of this dissertation introduces these challenges and covers previous research that has been done in those areas. Additionally, the objectives of this dissertation are stated. In chapter 2, a study was done in which polyethersulfone ultrafiltration membranes were functionalized with copper oxide nanoparticles and zwitterions for the purpose of reducing fouling on the membrane surface. Characterization confirmed the attachment of each component to the membranes. Dead-end filtration was performed with water and bovine serum albumin solution and showed that the functionalization reduced flux decline caused by organic fouling. The copper oxide nanoparticles were stable, with minimal copper leached. In chapters 3 and 4, two different green solvents, Cyrene and gamma-valerolactone, were investigated for fabricating polysulfone ultrafiltration membranes. The solvents are potential replacements for toxic solvents, such as n-methylpyrrolidone, that are commonly used in membrane fabrication. Ultrafiltration membranes were successfully fabricated, characterized, and tested by dead-end filtration. For both solvents, casting membranes into coagulation baths with mixtures of water and ethanol led to membranes with better flux and rejection than casting membranes into coagulation baths of pure water or pure ethanol. The mechanisms of how coagulation bath composition influences membrane morphology were analyzed. In chapter 5, dissipative particle dynamics simulations were performed to test the effect of solvent viscosity on the dynamics of phase inversion during membrane formation. Solvent viscosity impacted the speed of phase inversion and final degree of agglomeration of polymer. In chapter 6, conclusions are drawn and recommendations for future work are given.

This thesis project describes the modification and evaluation of a new affinity membrane for rapid chromatographic purification of non-antibody proteins. The affinity membrane utilizes Im7/CL7 coupling technology developed by Dr. Vassylyev's lab at the University of Alabama Birmingham (UAB), licensed by TriAltus Bioscience. The behavior of the membrane was evaluated using purified CL7-tagged Cas9 as my model protein for static and dynamic binding capacity analysis. Chapter one provides an overview on biopharmaceutical drug production process development. I discuss how protein drugs are produced, isolated, and purified from the cell supernatant after upstream phases are completed. Despite increasing demands for biologics, the isolation and purification of proteins from the cell supernatant, especially for non-antibody proteins where there is no universal affinity technology for purification, present major challenges for drug development and manufacturing. I highlight how new developments in downstream chromatography separation process, combined with multiple improvements in upstream manufacturing already in motion, will result in faster and higher volume production. Chapter two describes the development of Im7 affinity membranes. I discuss the synthesis of the membranes using direct modification of a regenerated cellulose base membrane and the measurement of binding capacities for CL7 and Cas9 proteins. For synthesis, N, N’-disuccinimidyl carbonate was coupled to the membranes. Unreacted succinimide carbonyl groups were coupled to primary amines groups of diamino-dipropylamine spacers. The unreacted primary amine was reacted with iodoacetyl chloride. Im7 was covalently coupled to the membranes. Ligand incorporation was evaluated with UV and FT-IR spectroscopy analyses. The ligand density was measured to be 38 mg Im7/mL. The static binding capacities (SBC) were measured to be 15 mg CL7/mL and 30 mg CL7-Cas9 /mL. Chapter three describes the development of Im7 affinity membrane using a polymer grafting approach and evaluation for the purification of Cas9. Surface-initiated atom transfer radical polymerization (ATRP) was used to graft 2-hydroxyethyl acrylate (HEA) from the membrane pores, growing polymer tentacles containing hydroxyl groups that were reacted with iodoacetyl chloride. Im7 was coupled covalently to the membranes through reaction with the iodoacetyl groups, as confirmed with spectroscopy analysis. The ligand density was measured to be 50 mg Im7/mL. For capturing analysis with purified proteins, the SBC were measured to be 20 mg CL7/mL and 50 mg CL7-Cas9 /mL. Dynamic binding capacities at 10% breakthrough (DBC) were measured to be 14.5 mg CL7/mL and 22 mg/mL CL7-Cas9. Chapter four presents an overview of the upstream and downstream work done by Dr. Vassylyev's lab at UAB that includes cell culturing of the proteins and the development of the resin beads used for the purification of Cas9. The 4B resin ligand density was measured to be 35 mg/mL Im7. The SBC was measured to be 66.5 mg/mL CL7. The 6B resin ligand density was measured to be 18 mg/mL Im7. Its SBC was measured to be 38.8 mg/mL of CL7, and it had a DBC of ~8 mg/mL for purifying CL7-Cas9 from lysate. Overall, Im7 affinity membranes produced by surface-initiated ATRP obtained better performance results than the membranes developed by the direct modification approach. Grafting HEA polymer chains from the membrane pores compensated for the smaller surface area than resin beads. It produced a higher Im7 ligand density compared to the TriAltus affinity resin beads and higher CL7-Cas9 capacity. Dynamic protein binding capacities remained constant after multiple runs of elution and regeneration. Therefore, the membranes can be reused for multiple runs without any production loss. Dynamic protein binding capacities for the membranes do not depend on flow rate and were higher compared to TriAltus resin beads.

Tangential flow filtration has many advantages for bioreactor harvesting as the permeate could be introduced directly to the subsequent capture step, the process is easy to scale up, and fouling of the filter is limited by the cross flow. However, membrane fouling has limited its widespread use. This is particularly problematic given the high cell densities encountered today. Here a reverse asymmetric commercial membrane, BioOptimal™ MF-SL (Asahi Kasei), where the more open surface faces the feed stream, and the tighter barrier layer faces the permeate stream, has been investigated for bioreactor harvesting. The open surface contains pores up to 40 µm in diameter, while the tighter barrier layer has an average pore size of 0.4 µm. The filtration performance, including fouling analysis conducted in this dissertation, involves using different feed streams, comparison of the filter performance with other filters possessing different membrane structures, mathematical modeling to predict the flux and fouling, fouling visualization using confocal laser scanning microscopy, and fouling identification using liquid chromatography-mass spectrometry. For the feed streams studies, filtration of yeast suspensions and Chinese hamster ovary cell culture has been conducted under various conditions. The yeast cells are trapped in the open pore structure, while CHO cells are more externally deposited. The membrane stabilizes an internal porous cake that acts as a depth filter. This stabilized cake layer removes particulate matter that fouls the barrier layer, protecting the fine pores from the large aggregates. As filtration continues, a cake layer forms on the membrane surface. Resistance-in-series model has been developed to describe the permeate flux during tangential flow filtration. The model contains three fitted parameters, which can easily be determined from constant pressure normal flow filtration experiments and total recycle constant flux tangential flow filtration experiments. The model can be used to estimate the filter's capacity for a given feed stream. Our results suggest that using a reverse asymmetric membrane could avoid severe flux decline associated with fouling of the barrier during bioreactor harvesting. Laser scanning confocal microscopy is used to observe the location of particle entrapment. The throughput of the reverse asymmetric membrane is significantly greater than the symmetric membranes. The membrane stabilizes an internal high permeability cake that acts as a depth filter. Confocal imaging helps visualize the secondary membrane directly by staining the DNA and membrane proteins using fluorescent dyes. Host cell proteins are the most challenging impurities for downstream purification processes. In order to investigate the fouling during cell clarification, HCPs in the bioreactor, harvest, and backwash are identified and quantified using different methods. A dataset is established using the identified HCPs and used to train the deep learning model. The model predicts unknown HCPs on fouling potential with an accuracy of 76%. The dataset of identified HCPs in this study provides insights into the characterization of membrane fouling, membrane selection, and process development. This approach could be used to screen cell lines or hosts to select those with reduced HCP profiles or identify HCPs that are problematic and difficult to remove.

Traditional radiochemistry approaches for the detection of trace-level alpha-emitting radioisotopes in water require lengthy offsite sample preparations and do not lend themselves to rapid quantification. Therefore, a novel platform is needed that combines onsite purification, concentration, and isotopic screening with a fieldable detection system. My dissertation research objective was to develop novel reactive thin polymer films and thin film composite membranes for the selective separation of uranium from environmental water followed by direct isotopic analysis by alpha spectroscopy. Chapter 1 reviews progress made on uranium separation from aqueous matrices and discusses methods used for the determination of isotopic composition. Chapter 2 describes the development of reactive polymer films for the concentration of uranium from circumneutral pH solutions for spectroscopic analyses. These films were prepared by grafting uranium-selective polymers from polyethersulfone (PES) films via UV-initiated polymerization, and by introducing uranium-selective functional groups to polyacrylonitrile (PAN) films by chemical reaction. Ellipsometry was used to study poly(phosphoric acid 2-hydroxyethyl methacrylate ester) film growth kinetics on PES films. X-ray photoelectron spectroscopy of modified PAN films revealed the conversion of nitrile groups to amidoxime groups to be as high as 40% and showed that the extent and depth of reaction could be varied precisely. Static uptake experiments with solutions of depleted uranium spiked with 233U were conducted to determine uranium binding capacities and kinetics of the modified polymer films at different pH values from 4 to 8. Sorption isotherm data were fitted to the Langmuir model, and the highest sorption capacities of 1.09×10-2 ± 1.03×10-3 mmol/m2 and 1.02×10-2 ± 3.00×10-3 mmol/m2 were obtained at pH 6 for modified PAN (M-PAN) and PES (M-PES) films. Capacities at pH 4 and 8 were lower and could be explained by differences in sorption mechanisms. Uranium batch uptake kinetics followed a pseudo-second order rate model. Equilibrium uptake was attained within 3 h for M-PAN film and 1 h for M-PES films. Alpha spectroscopy pulse height spectra were analyzed to study the role of selective layer film thickness on peak energy resolution. Full width at half maximum values from 20 to 41 keV were recorded for M-PAN film and from 26 to 45 keV for M-PES film. Whereas uranium uptake increased with selective layer film thickness and varied with polymer chemistry/extent of modification, the peak energy resolution was independent of layer thickness and polymer chemistry within the experimental measurement uncertainties. Results discussed in this chapter were used to guide the development of thin-film composite membrane-based detection methods for the rapid, fieldable analysis of radionuclides in water for nuclear forensics investigations and environmental studies. Chapter 3 describes the synthesis and characterization of polyamidoxime membranes for isolation and concentration of uranium from aqueous matrices, including high-salinity seawater. The aim was to develop a field portable screening method for the rapid quantification of isotopic distribution by alpha spectroscopy. Membranes with varying degree of modification were prepared by chemical conversion of nitrile groups to amidoxime groups on the surface of polyacrylonitrile ultrafiltration (UFPAN) membranes. Attenuated total reflectance Fourier-transform infrared spectroscopy was used to analyze changes in surface chemistry. Flow through filtration experiments conducted using deionized (DI) water and simulated seawater solutions indicated that the modified membrane was effective in capturing more than 95% of the uranium in the solution prior to breakthrough even in the presence of salt ions. Batch uptake experiments were conducted and compared with the flow through experimental data to elucidate on likely binding mechanisms. Alpha spectra of uranium loaded membranes were analyzed, and the effects of solution matrix and degree of modification on peak energy resolution were studied. Peak energy resolution of 24 ± 2 keV and 32 ± 6 keV full width at half maximum (FWHM) were obtained by loading uranium from DI and seawater solutions onto modified membranes. Full width at 10% maximum of the same spectra were calculated to be 63 ± 9 keV and 160 ± 34 keV to quantify differences seen in peak tailing. Calculations performed based on the results show that it would take less than 3 h of analysis time to screen a sample provided enough volume of solution are available. This chapter offers a facile method to prepare polyamidoxime-based membranes for uranium separation and concentration at circumneutral pH values, enabling the rapid, onsite screening of unknown samples. Chapter 4 presents the application of reactive membranes for the isolation and concentration of uranium from circumneutral pH solutions by ultrafiltration. The reactive membranes were prepared by grafting a uranium-selective polymer layer via ultraviolet-initiated radical polymerization from the surfaces of polyethersulfone ultrafiltration membrane supports. Dynamic uranium binding capacity measurements were conducted using the reactive membranes housed in an inline filter column, and column breakthrough data were fitted to theoretical models. The experimental data were best described by the Thomas model, indicating that uranium sorption was a reaction rate-limited process. Fitted model parameter values were compared with the results from batch experiments, where similar reaction rate constants were obtained for loading from solutions at pH 4 and 6. Maximum uranium binding capacity of the reactive membrane decreased in the presence of multiple competing ions, from 7.8 ± 0.3 mg/g in deionized water to 3.6 ± 0.2 mg/g in simulated seawater at pH 6. Alpha spectroscopy pulse height spectra of uranium-loaded reactive membranes were analyzed. Peak energy resolutions measured as full width at half maximum of 70 ± 8 keV and 65 ± 5 keV were obtained from samples loaded with uranium from DI and seawater solutions. The results of this study provide an approach for a rapid, on field screening of liquid samples to complement existing techniques and accelerate sample analyses. This work offers a thin-film platform for selective separation of uranium from seawater and rapid analysis of radionuclides by alpha spectroscopy. The proposed materials were characterized and contacted with uranium-contaminated solutions using batch and flowthrough filtration methods to demonstrate feasibility for field portable applications.

The development of rapid screening tools for special nuclear materials remains a crucial focus for nonproliferation efforts. Traditional approaches for the analysis of trace-level Pu isotopes in water requires tedious and time-consuming sample preparation steps that do not lend well to expeditious screening. Therefore, a novel analytical method that combines both Pu concentration and source preparation into a single detection system would make for an invaluable tool for nuclear security applications. Extractive membranes absorbers can help to fulfill this role as they are capable of concentrating Pu to detectable limits while subsequently serving as alpha spectrometry sample sources. In Chapter 1, I discuss the standard separation and sample preparation procedures for alpha spectrometry to screen waterborne Pu at concentrations below 10-12 M. Pu-extractive thin films and membranes are reviewed, and I detail methods used to develop the extractive copolymers and functional membranes presented throughout the dissertation. Chapter 2 describes my initial studies into the development of thin polymer-ligand films for Pu concentration and direct alpha spectrometry analysis. Submicron thin films were prepared through spin coating combinations of polystyrene (PS) with dibenzoylmethane, thenoyltrifluoroacetone and di(2-ethylhexyl) phosphoric acid (HDEHP) onto Si wafers. Of the three ligands tested, only films containing HDEHP showed significant Pu recovery at pH 2.3 and 6.3. Alpha spectrometry peak resolutions that rivaled conventional electrodeposited sources were obtained for the PS-HDEHP films over film thicknesses ranging from 30 to 250 nm. In an uptake study with a solution composed of greater than 90% Pu(V), it was observed that Pu recovery would increase over ten-fold when the HDEHP loading in the film was doubled. X-ray photoelectron spectroscopy (XPS) revealed that the ligand bloomed to the film surface in the samples with higher loadings of HDEHP. The XPS analysis also revealed that HDEHP was stripped from the films after contact with water at a circumneutral pH, but the films with higher HDEHP loadings retained some ligand even after a 12 h soak in water. Chapter 3 explores the fabrication of extractive thin-film composite (e-TFC) membranes for the rapid concentration and isotopic screening of Pu from a spiked water source. The e-TFC membranes were composed of a Pu-extractive copolymer spin coated onto an ultrafiltration membrane support. Phosphonate and phosphoric acid monomers were tested for their ability to create copolymers suitable for casting and extracting waterborne Pu at trace levels. These monomers were polymerized with either methyl methacrylate or 4-methylstyrene through bulk radical polymerization. In batch screening studies, it was observed that copolymers containing ethylene glycol methacrylate phosphate (EGMP) achieved the highest 242Pu recoveries from solutions at pH 4.2 and 6.8. A copolymer containing 5 % (w/w) EGMP to 4-methylstyrene exhibited a 242Pu distribution coefficient of 92.7 ± 18.5 L/kg at pH 6.8, and the copolymer was suitable for casting when dissolved in toluene. e-TFC membranes were prepared with the EGMP-based copolymer and tested in a direct filtration study. The e-TFC membranes were able to extract up to 10.2 ± 4.2 % Pu after filtering 10 mL of a solution bearing 4.53 Bq/mL 242Pu at pH 6.8. A 242Pu peak energy resolution of 71.7 ± 8.7 keV full-width at half-maximum was obtained for the e-TFC membranes and allowed for the distinction between 242Pu from 238Pu/241Am also present in the solution. Chapter 4 details the development of Pu-extractive membranes through non-solvent induced phase separation (NIPS). An innovative grafting procedure was performed to create a copolymer consisting of EGMP anchored to polyvinylidene (PVDF), which then can be used as a dope additive to cast PVDF membranes with Pu binding sites. Permeability studies were performed with membranes containing 10 wt% PVDF-g-EGMP, and the NIPS casting conditions were varied to study their effect on membrane permeability. Direct filtration studies showed that the membranes recovered up to 59.9 ± 3.0 % and 19.3 ± 3.5 % 238Pu from deionized and synthetic seawater solutions, respectively, after filtering 10 mL of solution at a concentration of ~0.5 Bq/mL 238Pu. Peak energy tailing was observed in the alpha spectra for the copolymer membranes, and SEM-EDS analysis indicated that PVDF-g-EGMP was distributed throughout the membrane and attributed to the loss in spectral resolution. The studies in my dissertation demonstrate that functional membranes for the detection and screening of waterborne Pu can be developed through relatively simple and effective measures. I hope that these studies will further aid the development of rapid and fieldable Pu detection systems and aid the advancement of nuclear security endeavors.

Virus filters are single-use devices that use a size-based separation process. In virus filters, contaminating virus particles are retained while the therapeutic molecules pass through the membrane pores. Virus filters are an essential component of the overall virus clearance strategy. Sections 1 and 2 of this dissertation provide an introduction and extensive review of monoclonal antibody (mAb) process development, where virus filtration is pivotal. In section 3, prefiltration studies were performed with an industrially relevant IgG1 type mAb using adsorptive and size-exclusion-based prefilters with different mechanisms of action. This mAb has an isoelectric point range of 7.1 to 8.0 and a molecular weight (MW) of 148 kDa. Decoupled prefiltration and virus filtration studies were conducted. We attempted to elute bound species from the membrane to identify them. Permeate fractions from the prefilters were introduced as feed fractions to a Planova BioEX (Asahi Kasei Medical, Tokyo, Japan) commercial virus filter for flux decay studies. Prefiltration and virus filtration studies were performed at different pH and ionic strength buffer conditions. By adjusting buffer conditions, and choosing prefilters with an appropriate mechanism of action, increased selectivity for foulant capture resulting in improved flux behavior during virus filtration could be achieved. Extensive characterization was also performed for the various filtration fractions to determine molecular species that increase fouling propensity in the virus filter and the efficacy of the different prefilters at removing these species. In section 4, prefiltration and flux decay studies on a Viresolve Pro (MilliporeSigma, Billerica, MA) as well as the Planova BioEX virus filter was performed with another industrially relevant mAb with an isoelectric point range of 5.95 - 6.55. The impact of excipients on mAb fouling behavior was determined. The impact of buffer pH was also evaluated with one pH condition below the isoelectric point (pI) of the mAb and another pH condition above the mAb pI. Decoupled prefiltration was performed to evaluate the impact of different types of prefilters on the filterability of this mAb. The pharmaceutical analysis system PA800 plus (SCIEX, Redwood City, CA) was also used to characterize the various mAb fractions from prefiltration and virus filtration. Dynamic light scattering (particle size analysis), size exclusion chromatography, SDS PAGE, capillary electrophoresis, and MALDI mass spectrometry were used for characterization. In section 5, a new technique of fractionating close molecular weight biomolecules was evaluated for virus clearance. The technique is known as internally staged ultrafiltration (ISUF), where layers of ultrafiltration membranes operate in stages to fractionate biomolecules based on differences in isoelectric points. The membranes of interest were the Pall Omega PES 300 kDa molecular weight cutoff (MWCO) flat sheet membrane, Pall Omega PES 100 kDa MWCO membrane, Millipore Ultracel 100kDa MWCO, and the Millipore Ultracel 30kDa MWCO. Virus clearance studies were performed using internally staged ultrafiltration membranes in skin and backing configurations. Section 6 is an overall conclusion for this work showing major findings and identifying areas for future study.

The specificity and efficiency with which enzymes catalyze selective chemical reactions far exceeds the performance of traditional heterogeneous catalysts that are predominant in industrial applications such as conversion of commodity chemicals to value-added products, fuel cells, and petroleum refinement. Moreover, biocatalysts exhibit exceptionally high product turnover at ambient conditions with little health and environmental burden. These advantageous qualities have led to the prolific use of enzyme catalysis in pharmaceutical, detergents, and food preservation industries wherein their use has greatly reduced waste generation, Unfortunately, the full slate of benefits that enzymes can impart to a broader range of chemical processes is severely hindered by lack of enzyme reusability and/or denaturation of native enzyme structure in industrially relevant conditions, i.e., high temperatures, turbulent flow regimes, non-aqueous solvents. Enzyme immobilization—the interfacing of biocatalysts with materials possessing properties of interest—is a well-studied strategy to address the shortcomings of free enzyme catalysis. Notably, immobilized enzymes demonstrate improved operability in non-physiological reaction conditions, lengthened shelf lives, and can be separated from reaction mixtures for subsequent reuse. Enzyme immobilization itself is however not a universal solution to the shortcomings that characteristic of free enzyme catalysis. This is because the immobilization of target enzymes often incurs mass transfer limitations, additional costs of materials, and loss of biocatalyst activity that can be prohibitive in the viable implementation. Thus, facile enzyme immobilization techniques require intensive investigation into the interactions between enzymes and carrier/scaffold materials that are not only determinant in the feasibility of immobilized enzymes applications but are also enlightening to the ongoing efforts to establish meaningful relationships between enzyme structure and function that will preserve extended biofunctionality in a variety of synthetic environments. Herein, we hypothesize that a strategy encompassing both laboratory-scale and computational work should be used to more fully understand the chemical and physical interactions that occur between an immobilized enzyme and different immobilization supports and thereby critically assess the interface between the two can be controlled to develop improved, user-controlled strategies for robust, efficient immobilized biocatalysis. To demonstrate our hypothesis, we have implemented our strategy on two material systems that are promising in future enzyme immobilization applications. The first material is a hyaluronic acid (HYA)-based polymer network that we have chemically modified to tailor its physicochemical properties for use as an enzyme immobilization platform for biomedical applications. HYA-based hydrogels were synthesized and characterized via techniques like Fourier transform infrared (FTIR) spectroscopy, thermogravimetric (TG) analysis, nuclear magnetic resonance (NMR) spectroscopy, among others; laboratory findings were corroborated with atomistic simulation of our confirmed hydrogel products to understand self-assembly of such polymer networks and how differences in our synthesis protocols impacted polymer properties that are critical in the design of an immobilized enzyme system. The second category of materials that we have investigated is metal-organic framework (MOF)-based composites zeolitic imidazolate framework-8 (ZIF-8) and aluminum (Al)-based MIL-160. The two model MOFs were considered due to their modular, tunable nature which presents a strong opportunity to engineer the interfacial surface at which enzymes interact with such materials and, subsequently, rationally implement optimization strategies that preserve the native structure and function of immobilized enzymes for biocatalytic implementations. Specifically, we have investigated how user-controllable MOF properties—i.e., internal cage structure, hydrophobicity, porosity—impact MOF performance in gas separations and how such properties can be integrated efficiently with active biocatalysts. Our complex and synergistic findings demonstrate not only the feasibility but rather the necessity of well-designed, user-tailored material interfaces in improving the performance of immobilized enzymes for robust industrial-scale catalysis applications. We expect that the realization of such high-performance technical enzymes will allow for further improvements in the sustainability and economic viability of chemical processes spanning a range of technical sectors.

As interest in the modularization and intensification of chemical processes continues to grow, more research must be directed towards the modeling and analysis of intensified process units. Intensified process units such as membrane reactors pose unique challenges pertaining to design and operation that have not been fully addressed in the reported literature. This work aims to address the design and control challenges caused by the integration of phenomena and the loss of degrees of freedom (DOF) that occur in the intensification of modular membrane reactor units. First, a novel first-principles approach for modeling membrane reactors is developed using the AVEVA Process Simulation Platform’s equation-oriented capabilities. The produced model allows for the simulation of generalized membrane reactors under nonisothermal and countercurrent operation for the first time. This model is then applied to generate an operability input-output mapping to study how operating points translate to overall unit performance. This work demonstrates how operability analyses can be used to identify areas of improvement in membrane reactor design, other than just using operability mapping studies to identify optimal input conditions for process operations. Next, a novel approach to designing membrane reactor units is proposed. This approach consists of designing smaller modules based on specific phenomena such as heat exchange, reactions, and mass transport and combining them in series to produce the final modular membrane-based unit. This module-based approach to designing membrane reactors is then assessed using a process operability analysis to maximize the operability index, as a way of quantifying the operational performance of intensified processes. This work demonstrates that by designing membrane reactors in this way, the operability of the original membrane reactor design can be significantly enhanced, translating to an improvement in achievability for a potential control structure implementation. Although the demonstrated novel module-based design approach to membrane reactors could improve the operability index of membrane reactor systems, the computational time to determine such an optimal design made this class of design problems intractable to solve in a reasonable amount of time. So lastly, this work proposes a set of design heuristics for this new module-based design approach for membrane reactors. These heuristics are used in combination with a genetic algorithm to produce a novel, two-staged algorithm for the design and control of membrane reactor systems. The proposed algorithm leads to a reduction in computational time by about 2 orders of magnitude while also improving the operability index of the original membrane reactor design by 21%.

The objective of this research was to test the feasibility of using forward osmosis (FO) with polyelectrolyte draw solutions to recover water from bioreactor mixed liquors. When combined with an Anaerobic Osmotic Membrane Bioreactor (AnOMBR), such a system could process fecal and food waste from astronauts aboard the International Space Station (ISS) and reclaim important nutrients and water. This project focused on measuring the obtainable water recovery rates from bioreactor effluent, and the identifying challenges associated with the operation. AnOMBRs feature several advantages over aerobic bioreactors, and non-osmotic anaerobic membrane bioreactors (AnMBR). Anaerobic bioreactors avoid the significant energy costs of aerobic bioreactors. Conventional AnMBRs use microfiltration or ultrafiltration with an applied pressure. AnOMBRs use FO membranes that do not allow low molecular weight organics to pass, and do not require an applied pressure. AnOMBRs may be an option for water and nutrient recovery in space if they can attain high water flux and reverse solute flux selectivity (RSFS), which quantifies the volume of permeated water per gram of draw solute that has diffused from the draw solution into the bioreactor. To obtain a high RSFS, poly(acrylic acid) and poly(acrylic acid-co-maleic acid) were evaluated as draw solutes. The large size of the polymer and the high osmotic pressure they can provide make them highly advantageous. Water flux was measured in a direct flow system using wastewater from a municipal wastewater treatment plant and draw solutions prepared with two polyelectrolytes at different concentrations. The direct flow tests displayed a high initial flux (>10 L/m2/h) that decreased rapidly as solids accumulated on the feed side of the membrane. A test with deionized water as the feed revealed an exceedingly small mass of polyelectrolyte crossover from the draw solution to the feed; thus, RSFS was 80, which was higher than the target value of 20 for viable operation. Crossflow filtration experiments demonstrated that steady state flux above 2 L/m2/h could be maintained for 70 h following an initial flux decline due to formation of a foulant cake layer. This study established that FO could be feasible for regenerative water purification from bioreactor digesters. By utilizing a polyelectrolyte draw solute with exceedingly high RSFS, we expect to overcome the need for draw solute replenishment. This would be a major step towards sustainable operation in long-duration space missions lasting 30 months or more.

The motivation of this work comes from one of the major problems of emerging non-aqueous flow battery (NAFB) that a separator or membrane which facilitates conductivity and blocks redox species crossover does not exist. Although many aspects of principles can be mirrored from mature fuel cell and aqueous flow battery, it is found that some well-defined membrane properties in aqueous systems such as swelling, transport and interactions are different in non-aqueous solvents to some extent. However, the approach of this work does follow the way perfluorosulfonate ion exchange membrane (PFSA) facilitated development of fuel cell and aqueous flow battery in the past. The aim of the work is not to identify or suggest one or one type of membrane that fits the non-aqueous flow battery. This is quite impossible based on the extremely diverse scope of published non-aqueous redox species. Thus, effort to understand and establish fundamentals of transport and interactions within membrane is the goal of the work to bridge the polymer design/synthesis and practical flow battery development is the goal of this work. The highlight of this work includes observation of high ionic conductivity of tetraalkylammonium form PFSA of high organic solvent swelling as well as, solvation and interactions among the cation, organic solvent and membrane, low swelling but high ionic conductivity of asymmetric organic cation as a charge carrier for non-aqueous flow battery from an interaction perspective. In terms of techniques, advanced Fourier-transform infrared spectroscopy (FTIR) and Nuclear Magnetic Resonance (NMR) methods beyond simple material characterization are designed in order to probe interactions and solvation. Although the methodologies involved here have long been used to study aqueous or non-aqueous electrolytes dating back to the 50s and 60s, this work aims to push these methods to another level inside of the membrane, a non-homogeneous and non-aqueous environment. Local information is separated from bulk property to understand transport dynamics and reveal membrane properties.

As substantial developments were achieved in nanotechnology and polymer engineering, especially in the last few decades, the use of membranes and membrane-based procedures was found to be expanding into more and more research and development areas; including biological engineering, life sciences and biomedical engineering. Not only have they been the main focus of meaningful research, but they have also been the main pieces of the solutions to very thorny problems encountered within a wide range of applications from microfluidics to water treatment, thanks to their versatility, cost-effectiveness and biocompatibility, when compared to conventional separation techniques. To celebrate and embrace these qualities, the current research focuses on several impactful membrane-based approaches including reverse electrodialysis (RED) (Chapters 2 and 3), electrodeionization (EDI) (Chapter 4) and hemodialysis (HD) (Chapter 5) and their implementations for biomedical engineering applications. Moreover, this dissertation aims to offer a helpful reference for novel technologies including the development of an external hollow fiber membrane salt cartridge to boost the performance of a reverse electrodialysis cell, and a unique activated ion exchange technology that could be adapted into selective electrolyte removal in various ion-exchange based aqueous treatment applications. As we continue to thrive to solve the mysteries created by fluid-solid surface interactions, membrane technologies will remain to be a strong tool to navigate the humanity in his endless journey.

Porous polymer membrane filters are widely used in separation and filtration process. Micro- and ultra-filtration membranes are commonly used in biopharmaceutical applications such as filtering viruses and separating proteins from a carrier solution. The formation of these membrane filters via phase inversion is a complex and interconnected process where varying casting conditions can have a wide variety of effects on the final membrane morphol- ogy. Tailoring membrane filters for specific performance characteristics is a tedious and time consuming process. The time and length scales of membrane formation make it extremely difficult to experimentally observe membrane formation. Modeling the membrane formation process allows one to slow down time and closely observe the formation of complex pore net- works. This allows new understanding and visual representations of the effects of different casting conditions and the resulting pore networks that form. This dissertation presents two separate models for two different membrane formation processes - thermally induced phase separation (TIPS) and non-solvent induced phase sepa- ration (NIPS). The Phase-Field method is employed to model the mesoscopic morphological structures that emerge during membrane formation. The Cahn-Hilliard equation is used to capture the diffusive nature of membrane formation and the Flory-Huggins free energy of mixing describes the thermodynamic equilibriums of the polymer solutions. Large-scale two- and three-dimensional simulations are used for capturing the resulting pore morphology under various casting conditions. The model for TIPS evaluated different casting conditions for the membrane forming system PVDF/DPC (polyvinylidene fluoride/diphenyl carbonate). The effects of casting surface temperature, thermal conductivity, quench rate, and polymer concentration were in- vestigated. Isotropic thermal quenches were used to analyzed pore size and interconnectivity while varying the quench rate and polymer concentration. Anisotropic thermal quenches in which a constant temperature cooling surface is imposed at the top surface were used to an-alyze the pore size and formation of a dense pore region near the casting surface. Different polymer concentrations, casting surface temperatures, and polymer thermal conductivities were used to characterize their effects on pore morphology. The formation of a dense pore region occurs near the top surface and the thickness of this dense pore region is dependent on the previous mentioned casting conditions. The model for NIPS evaluated the effects of coagulation bath composition and polymer concentration on final membrane morphology for PES/NMP/Water (polyethersulfone/n- methyl-2-pyrrolidone/water). These simulations were then compared to handcast mem- branes of a similar membrane forming system for comparison the simulated structures. The NIPS model evaluated coagulation bath composition and polymer concentration and the resulting morphologies. Two and three dimensional simulations were used to look at cross section and top surface morphology respectively. The concentration of NMP in the co- agulation bath was varied to verify the predictive capabilities of the model. The resulting simulations of cross sections and top surfaces had good agreement with handcast membranes when NMP is added to the coagulation bath.

With the looming dominance of poorly water-soluble chemical entities within the pharmaceutical pipeline, the pharmaceutical industry has leaned on the use of supersaturating drug delivery systems (SDDSs) to achieve efficacious concentrations within the gastrointestinal fluids. SDDSs aim to achieve concentrations in solutions greater than the solubility of the lowest energy crystalline form. However, the generation of supersaturated solutions of active pharmaceutical ingredients (APIs) creates a strong crystallization potential, which is undesirable. In product development, supersaturating products often fail in Phase I and Phase II clinical trials due to poor oral bioavailability and a lack of in vivo efficacy. Pre-clinical testing exists as a filter for product failures; regardless, product failures still occur in the clinical stage of product development leading to delays or cancellations of novel treatments. This disconnect between pre-clinical performance testing and in-vivo outcomes highlights the lack of understanding of the in-vivo oral absorption process of APIs from supersaturated solutions. The pharmaceutical industry would greatly benefit from identifying critical factors associated with the safe and efficacious delivery of pharmaceutical products to patients in need. Currently, in-vitro dissolution testing remains the most commonly used technique to evaluate product performance. One salient variable to dissolution testing is the dissolution media itself. Maintenance of supersaturation is believed to be part of the complex mechanism by which SDDSs enhance the oral bioavailability of poorly soluble drugs. Studies have shown that the dissolution of SDDS is highly dependent on the media used. The implications of endogenous materials such as bile acids and phospholipids on oral absorption have been studied and accounted for in commercial formulations of simulated gastrointestinal (GI) fluids. However, the impact of mucin, the main glycoprotein in mucus, in solution has been neglected, which results in studying product performance without fully characterizing all the variables. Mucin is the main component of mucus gels lining the entirety of the gastrointestinal tract, and significant quantities are expected to be found within GI fluids. Several recent studies have shown the impact of mucin on the permeation of lipophilic compounds, indicating the potential for interactions with such molecules. Therefore, it is expected mucin may stabilize supersaturated solutions of poorly water-soluble compounds via similar interactions. Overall, this dissertation aims to probe the impact of mucin as a solution component capable of stabilizing supersaturation: The inclusion of mucin into in-vitro media to achieve an improved biorelevance is explored and evaluated as a potential reason as to why the in-vitro dissolution characteristics of ketoconazole do not match that within in-vivo systems. The sensitivity of the in-vitro examination method to seed crystals is also evaluated as a potential confounding factor in in-vitro examinations. A 17beta-Estradiol (E2) and polysorbate 80 (PS80) micelle system was used to probe the increases in permeability to the diffusion of nano-scale particles through the aqueous boundary layer (ABL). Coupled with diffusion coefficient determination of a model nano-scale species via NMR, a mathematical model was developed to accurately predict the permeability increases due to increases in effective diffusivity due to the nano-scale shuttling of E2 through the ABL. The ability of mucin to stabilize supersaturated solutions is evaluated in the context of its ability to reduce the thermodynamic activity, or the driving force for crystallization, of several model drugs: 17beta-estradiol, nifedipine, and bicalutamide. The flux of these model drug compounds is evaluated in the absence and presence of mucin using horizontal diffusion cells. A decrease in the flux of API in the systems containing mucin relative to the systems in the absence of mucin indicates a reduction of thermodynamic drug activity.

Capitalizing on byproducts of industrial and agricultural economies is among the utmost goals of sustainability. Of particular interest for commercial upgrading is lignin, a phenolic biopolymer found in the cell walls of plants which is the second most abundant biopolymer on Earth after cellulose. Due to its heterogeneous structure, deconstructing lignin to selected small molecules for use as chemicals or advanced materials has been elusive. This work capitalizes on a “bottom up” approach to the synthesis of lignin oligomers of known bond chemistry to better understand their interfacial interactions. The potential pharmacological mechanism of lignin deconstruction components and their toxicological effect on biological systems is currently relatively unexplored. Herein, the interaction of three lignin-derived small molecules (lignin dimers with varying chemical functionality) with lipid bilayers (model cell membranes) was investigated via quartz crystal microbalance with dissipation monitoring (QCM-D) studies of binding and differential scanning calorimetry (DSC) measurements of the change in lipid bilayer phase behavior as a function of dimer concentration. Our results demonstrate that minor differences in structure of lignin molecules have a significant impact on their ability to penetrate into model cell membranes. To show that the lignin oligomers under investigation have the potential to impart surfaces with lignin-like properties (e.g. pharmacological and toxicological properties), a hydrophobic lignin dimer which was previously shown to interacts strongly with model cell membrane was chemically modified to covalently attach to mesoporous silica nanoparticles (MSNPs). The ability of lignin dimer-functionalized particles to interact with and disrupt lipid bilayers was compared to MSNPs functionalized with eugenol, a natural aromatic pharmaceutical found in cloves. While eugenol-grafted particles showed evidence of weak interactions with the bilayer, dimer-grafted particles with the same concentration caused considerable lipid mass loss associated with disruption of up to ~93% of the membrane, suggesting higher biocidal activity. The structure of lignin suggests that it is not only a potential therapeutic and pharmacological substance, but also a promising source of aromatic chemicals. Developing methods to selectively separate and purify lignin oligomers from the complex lignin depolymerization mixture remains an ongoing challenge. Binding of β-cyclodextrin (CD) (a truncated cone-shaped molecule with a hydrophilic exterior and a hydrophobic cavity) to three lignin dimers with different chemical structures was studied using Isothermal Titration Calorimetry (ITC). The thermodynamic parameters (K, ΔH, ΔS, and ΔG) of formation of dimer:β-CD complexes were measured and compared. The results demonstrated that differences in the thermodynamics of CD-lignin interactions can be used to develop selective separation strategies for lignin-derived small molecules. For example, binding with β-CD of a dimer containing a bb bond is entropically driven (due to hydrophobic effects) while favorable enthalpy of interactions drives binding of a dimer with a different bond (bO4). A surface modification technique was also proposed to attach β-CD directly to mesoporous silica nanoparticles, with a goal of using the silica particles for selective capture of lignin-derived small molecules. In sum, this work established structure−function relationships for well-defined lignin derivatives at biologically relevant surfaces. A strategy to create lignin-conjugated silica surfaces was developed and it was shown that bio-inspired materials comprising lignin small molecules have the potential to serve as a platform for novel antimicrobial coatings and therapeutic carriers. Moreover, the potential of β-cyclodextrin to selectively separate lignin oligomers from lignin deconstruction mixture was studied, and a strategy to create immobilized cyclodextrin-functionalized surfaces was proposed.

Virus filtration is an integral part of the downstream purification of mammalian cell culture-derived biotherapeutics to assure the viral safety of the products. Virus filtration membranes remove viruses based on a size-exclusion mechanism. Commercial parvovirus filers possess unique membrane structure and are designed to remove smaller non-enveloped parvoviruses with size 18-26 nm. However, some filters face issues, such as pre-mature fouling, the decline of filtrate flux, and reduction in virus retention. This doctoral dissertation focused on identifying the factors that influence the filtrate flux and the virus retention capability of commercial virus filters. The effects of solution pH and ionic strength, buffer type, protein molecules, and filter properties on virus filters' performance are investigated. In Chapter 2, virus filtration was performed for an Fc-fusion protein using three commercial virus filters in solutions with varying buffer type and salt concentrations. A combined pore blockage and cake filtration model successfully described the three filters' fouling behavior in all the solution conditions. The model indicates that the blockage of membrane pores by larger particles (e.g., virus and protein aggregates) occurs in the initial stage of the filtration, followed by a cake filtration where particles accumulate on the membrane surface. Both the pore blockage rate and the rate of increase of protein layer resistance over blocked pores are affected by the solution conditions and membrane properties. In Chapter 3, an Food and Drug Administration (FDA) recommended model virus, minute virus of mice (MVM) was spiked into the solution of three different protein molecules under various solution conditions, then filtered with three commercial virus filters. The log reduction values (LRV) were determined to systematically evaluate the effects of solution pH, salt concentration, buffer species, protein molecules' properties, cumulative viral loads, and the process interruption on the virus retention capability of the virus filters. The results showed that some virus filter is more prone to virus breakthrough than the others, both during the filtration and after a process interruption. In some cases, a higher flux is associated with a more significant LRV decline. The diffusion interaction parameter, ??, is an indicator of colloidal and thermal stability of proteins in solution and a negative ?? represents the presence of attractive intermolecular forces. The ?? of a monoclonal antibody was determined in various solution conditions and was found to be affected by the pH, ionic strength, and the buffer species. Besides, a more negative ?? value is associated with a lower average relative flux of a commercial virus filter. These results indicate that ?? can be potentially used as an indicator of virus filter performance. The comparison between a two-layer membrane and a single-layer module of the same membrane was made to investigate the cause for their difference in virus retention. Fluorescent labeling of the virus and the protein was done; confocal imaging of the membranes after filtration was done as well to visualize the location of the captured virus and protein within the membrane.