JP7797436B2 - Electrospun nanofiber hybrid felt - Google Patents

Description

The present invention relates generally to compositions for use in biological and chemical separations and other applications. More specifically, the present invention relates to hybrid felts fabricated from electrospun nanofibers having high permeability and high capacity.

Microfiber and nanofiber membranes or "felts" have a variety of uses in both biological and industrial applications. For example, felts are useful for providing textile reinforcement, protective clothing, catalytic media, agricultural applications, sensors for environmental, medical, and military surveillance, biomedical applications (e.g., bioseparations, tissue engineering, and wound dressings), electronic applications (e.g., capacitors, transistors, and diodes), and space applications (e.g., support structures for solar sails and space mirrors). Microfiber and nanofiber felts are particularly advantageous for purifying biological materials such as proteins, nucleic acids, carbohydrates, bacteria, viruses, and cells. They are useful in all fluid applications, both liquid and gaseous.

The biopharmaceutical therapeutics industry is expanding as an increasing number of biologics are approved for sale. In addition, biologically-based diagnostic tools are being used extensively to perform high-throughput, highly sensitive diagnostic tests for a variety of disease states. For both therapeutic and diagnostic purposes, biological materials (e.g., recombinant proteins, monoclonal antibodies, viral vaccines, and nucleic acids) must be efficiently produced, purified, and used.

Traditional purification methods include, for example, the use of packed microbeads for adsorption/chromatography, ultrafiltration, and precipitation/crystallization to separate desired biological materials from by-products and other contaminants. While these traditional separation methods produce results suitable for biological applications, they are limited in terms of yield, processing time, and purity. These limitations are primarily due to the slow diffusion rates of relatively large biomolecules, which restrict the ability of the material being purified (i.e., the "target material") to reach available binding sites deep within the separation matrix. Furthermore, these systems can only be used for a limited number of cycles, and some can only be used once.

Ion exchange (IE) and hydrophobic interaction (HI) adsorption/chromatography are two of the more robust conventional separation techniques widely used for the separation of biological materials. While these are generally less efficient overall than specific affinity-based separation techniques, such as antibody-based separations, they are still useful for purifying many target substances from unwanted by-products and impurities when separation conditions are carefully selected.

While affinity-based adsorption/chromatography can be more efficient than IE and HI, they are generally difficult and expensive to produce due to the complexity of generating and purifying biological ligands, such as monoclonal antibodies and nucleic acids. Such ligands are often highly sensitive to environmental conditions (e.g., temperature, pH, ionic strength, etc.) and can easily degrade, thus disrupting the affinity interactions required for adsorption. Furthermore, disrupting binding interactions can be difficult without using harsh conditions that can reduce biological activity and therefore the utility of the target substance and/or the reusability of the purification medium.

Membranes useful for purifying biological materials have been described (see, e.g., Bioprocessing for Value-Added Products from Renewable Resources, Shang-Tian Yang, Ed., Chapter 7). Recently, membrane adsorption/chromatography using nanometer-diameter fibers assembled into mats (i.e., "nanofiber felts") with controlled thickness has shown considerable promise for use in bioseparations (Todd J. Menkhaus, et al., "Chapter 3: Applications of Electrospun Nanofiber Membranes for Bioseparations, ^" in Handbook of Membrane Research, Stephan V. Gorley, Ed.). Such nanofiber felts are superior to microfiber felts because pore size, affinity properties, and other performance criteria can be tightly controlled.

While previously described single-component nanofiber felts have provided promising results, they are often less efficient than would be desirable in terms of felt stability and material and time requirements. This is especially true when the target substance is present at low concentrations in the starting material to be purified and is enriched with contaminants and/or synthetic by-products. Thus, there is a need to improve felt stability and the efficiency of biological product purification. The embodiments disclosed below fulfill that need.

The following simplified summary provides a basic understanding of some aspects of the claimed subject matter. This summary is not an extensive overview, and is not intended to identify key/critical elements or delineate the scope of the claimed subject matter. Its purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is presented below.

In one embodiment, the invention is an electrospun hybrid nanofiber felt formed from composite nanofibers and single-component nanofibers. The composite nanofibers can be electrospun from a mixture of derivatized cellulose and a first non-cellulosic polymer, and the single-component nanofibers can be electrospun from a second non-cellulosic polymer, which can be the same as or different from the first non-cellulosic polymer. Typically, the first and second non-cellulosic polymers are differentially removable from the nanofiber felt. This means that conditions exist (using solvent or heat, or a combination of solvent and heat) under which one of the non-cellulosic polymers is removed to a greater extent than the other (e.g., by a difference in removability of 10% or more, e.g., 20% or 50%).

The derivatized cellulose in the composite nanofibers can be an organic ester of cellulose, an inorganic ester of cellulose, or an alkyl cellulose. The organic ester of cellulose can be cellulose acetate, cellulose triacetate, or cellulose proprionate.

When the derivatized cellulose is an inorganic ester, it can be cellulose nitrate or cellulose sulfate, and when it is an alkyl cellulose, it can be hydroxyethyl cellulose or carboxymethyl cellulose.

The first non-cellulosic polymer can be a synthetic or natural polymer such as a vinyl polymer, a polyether, an acrylic polymer, a polyester, a polycarbonate, a polyurethane, a polysaccharide (e.g., starch or chitin), a polyamide (e.g., a protein or gelatin), a polylactide, a polyglycolide, or a copolymer thereof.

In one embodiment, the second non-cellulosic polymer forming the single component nanofibers is a synthetic polymer such as a vinyl polymer, polyamide, polyimide, polyester, or copolymers thereof.

In another embodiment, the present invention is an electrospinning method for producing the above-described nanofiber felt, which can be described by the following steps: a) separately preparing a composite polymer spin dope and a single-component polymer spin dope; b) placing the spin dopes in two different spinnerets; c) applying a voltage to each spin dope using electrodes; d) separately electrospinning the composite and single-component nanofibers from the spinnerets; and e) collecting the solidified nanofibers as a randomly overlapping or partially aligned nanofiber felt.

The as-formed nanofiber felt can be further processed by regenerating (i.e., converting back to cellulose) the derivatized cellulose in the composite nanofibers. The method of making a nanofiber felt may also include the additional step of removing some or all of the non-cellulosic first polymer from the composite nanofibers. Alternatively, or in addition to these steps, the method of making a nanofiber felt may also include the step of surface functionalizing one or more of the polymer nanofibers in the nanofiber felt. Such surface functionalization may involve the attachment of an affinity ligand having a specific affinity for a particular target molecule to be purified from the fluid.

In another embodiment, the present invention is a method for purifying a biomolecule from a fluid, which can be described by the following steps: a) preparing a nanofiber felt according to the method described above; b) flowing the fluid through the nanofiber felt; and c) recovering the biomolecule from the nanofiber felt.

Other aspects of the present invention are found elsewhere in this specification.

FIG. 1 is a schematic diagram showing an electrospinning process.

The present invention generally relates to hybrid felts composed of electrospun nanofibers for use in, for example, chemical and biological separations. Hybrid nanofiber felts have high separation capabilities and provide reproducibility at high flow rates and pressures over multiple cycles. Such nanofiber felts exhibit a complex, interconnected three-dimensional porous structure and a relatively large surface area.

In particular, hybrid nanofiber felts are composed of more than one polymer type (i.e., they are "hybrid" felts).

The felts of the present invention are composed of more than one polymer type (i.e., are "hybrid" felts). This includes hybrid felts made from a combination of single-component nanofibers and "composite" nanofibers (e.g., nanofibers made from a mixture of two or more materials) into a "hybrid" felt. For "composite" nanofibers, the "backbone polymer" is derivatized cellulose, and the first non-cellulosic polymer can be removed from the fiber/felt by exposure to high temperature, a chemical solvent, or both high temperature and a chemical solvent. In some embodiments, removal of the first non-cellulosic polymer simultaneously converts the derivatized cellulose back to cellulose, i.e., the cellulose is "regenerated."

The nanofibers in the felt of the present invention are produced using the electrospinning technique, which refers to the production of fibers based on extruding a polymer "spin dope" by exposing it to an electrostatic field, thereby stretching an extruded polymer "jet" into nanofibers.

These and other exemplary aspects of the present invention are described in further detail below.

DEFINITIONS In the following description, several terms are used generically. The following non-limiting definitions will provide a clear and consistent understanding of the specification and claims, including the exemplary scope to be given such terms.

The terms "one," "a," or "an," when used in this disclosure, mean "at least one" or "one or more" unless otherwise indicated.

As used herein, the term "invention" or "present invention" is not intended to be limiting and is not intended to refer to any single embodiment of a particular invention, but rather encompasses all possible embodiments described in the specification and claims.

As used herein, the term "permeance" refers to the flux of fluid through a nanofiber felt per unit pressure drop per unit thickness of the felt. A permeability greater than 500 L/(min· ^m2 · ¹⁰⁵ Pa) is considered "high."

The term "flux" refers to the rate at which a fluid flows through a nanofiber felt per unit time and per unit area of surface exposed to the flow.

As used herein, the term "capacity" refers to the amount of product bound per unit of adsorbent. Protein adsorption capacity is considered "high" when it exceeds 100 mg of protein/g of adsorbent.

As used herein, the terms "membrane," "felt," and "mat" are used interchangeably and refer to a nonwoven or randomly layered collection of fibers.

As used herein, the term "nanofiber felt" refers to a collection of nanofibers arranged in a substantially planar configuration, but may also include microfibers added for strength, improved flux, etc.

As used herein, the term "microfiber" refers to fibers having a diameter greater than 1.0 micrometer, generally a diameter of 1.0 micrometer or greater and 1.0 millimeter or less.

As used herein, the term "nanofiber" refers to a fiber having a diameter of less than 1.0 micrometer, typically between 10 nanometers and 1.0 micrometers, for example, between 200 nm and 600 nm.

As used herein, the term "hybrid nanofiber felt" refers to a nonwoven or randomly layered collection of fibers made from at least two types of polymers, in which monocomponent or bicomponent fibers are combined with at least one other monocomponent fiber or at least one other bicomponent fiber.

As used herein, the term "single-component nanofiber" refers to a nanofiber made from a single polymer.

As used herein, the term "single-component nanofiber felt" refers to the assembly of many single-component nanofibers into a nonwoven or randomly overlapping mass of fibers.

As used herein, the term "composite nanofiber" refers to a nanofiber made from at least two different polymers.

As used herein, the term "gently elevated temperature" refers to a temperature above 24°C and below 110°C.

As used herein, the term "differentially removable" means that when the hybrid nanofiber felt is composed of at least two non-cellulosic polymers, the conditions (exposure to elevated temperatures and/or solvents) can be selected such that one of the non-cellulosic polymers is removed to a greater extent than the other non-cellulosic polymer (by at least 10%, up to 100% vs. 0%).

As used herein, the term "solvent" refers to any single-component liquid or mixture of liquids capable of dissolving one or more components of the nanofiber felt.

As used herein, the term "spin dope" refers to the polymer solution used in the electrospinning process.

As used herein, the term "electrospinning" refers to the application of electric forces to a spin dope to form nanofibers.

As used herein, the term "thermally stable" means that the polymer does not degrade in the temperature range of 50-110°C.

As used herein, the term "chemically stable" means that the polymer is not soluble in solvents such as water or commonly used organic solvents (e.g., alcohols and hydrocarbons) and mixtures thereof.

Derivatized Cellulose Cellulose is a structural component found in the cell walls of plants and algae. It is also secreted by some bacteria. Therefore, it is the most abundant organic compound on Earth. It is derived from D-glucose units linked into linear polymers via β(1-4) glycosidic bonds. For biological and industrial applications, cellulose is purified from plants, wood pulp, or cotton and converted into many useful materials, including paper, cellophane, rayon, and biofuels. The utility of cellulose can be largely attributed to its physical properties: it is odorless, hydrophilic, relatively insoluble, has very low nonspecific binding properties, and is biodegradable.

While cellulose-based separation media offer many advantages, they are unfortunately chemically unstable (i.e., they decompose) in strong acids and bases. Furthermore, dissolving cellulose requires the use of specialized solvent mixtures, such as N-methylmorpholine-N-oxide (NMMO) and water or lithium chloride and N,N-dimethylacetamide. This limits the use of cellulose-based media to operations that do not require the harsh regeneration conditions often required in the biopharmaceutical industry to meet stringent cleanliness regulations mandated by the FDA.

Cellulose fibers are traditionally produced by wet spinning and contain pre-existing cellulose derivatives, since it is very difficult to electrospin cellulose directly from a solution or melt. To prepare cellulose nanofibers, considerable research effort has been devoted to electrospinning cellulose derivatives, such as cellulose acetate. Unlike cellulose, cellulose acetate is soluble in many commonly used solvents, such as acetone. Cellulose acetate can be electrospun into nanofibers, and regenerated cellulose nanofibers can be produced by subjecting the nanofibers to a post-spinning treatment of hydrolysis/deacetylation.

Therefore, in the practice of the present invention, one of the polymers in the hybrid nanofiber felt is derivatized cellulose. Cellulose can be readily derivatized using well-known methods to convert the -OH groups of individual glucose units to other moieties with higher or lower reactivity, various charges, etc. Such derivatized cellulose species exhibit improved stability when exposed to solvents and other desirable physical properties. Many cellulose derivatives are readily available commercially. Exemplary derivatized cellulose species include, for example, organic esters (cellulose acetate, cellulose triacetate, cellulose propionate, cellulose acetate propionate, cellulose acetate butyrate), inorganic esters (cellulose nitrate, cellulose sulfate), and alkyl celluloses (hydroxyethyl cellulose, carboxymethyl cellulose).

The hybrid nanofiber felt of the present invention typically comprises a majority (i.e., 51% or more) by mass of derivatized cellulose, for example, more than 60% or more than 70%.

While the majority by mass of the hybrid nanofiber felt is derivatized cellulose, incorporating additional types of fibers into the felt can provide functionality required for the felt's intended use. Thus, having additional fibers in the felt is desirable because it can provide increased mechanical strength to the felt, allow for the incorporation of multiple functionalities into the felt, and provide stability and other aspects of the manufacturing process, as described elsewhere herein. Indeed, the inventors have unexpectedly discovered that the inclusion of even a small proportion of non-cellulosic polymers in the hybrid nanofiber felt can improve the electrospinning process and also enable the finished product to be suitable for a variety of biological and industrial applications, particularly when the hybrid nanofiber felt is comprised of both composite and single-component nanofibers.

Synthetic polymer nanofibers (e.g., those produced from vinyl and acrylic polymers) offer a wide range of chemical functionality for bioseparations and other applications. By combining different polymer units, the surface chemistry of the resulting fiber can be controlled as part of the electrospinning process, imparting functionality directly to the resulting nanofiber. Alternatively, and similar to conventional micrometer-scale fibers, the surface functionality of polymer nanofibers can be chemically modified after electrospinning to meet specific functionality requirements for various bioseparations applications (described below). Functionalization chemistries are well known in the polymer art. They also tolerate the rigorous cleaning regimes typically associated with bioprocessing. Exemplary functionalization chemistries are also described in further detail elsewhere herein.

Synthetic carbon-based adsorption media and filtration membranes are often much more chemically robust than cellulosic media and can therefore be used when strong acids and bases are required to clean the separation media between uses. Furthermore, hybrid nanofibers containing both cellulosic and non-cellulosic polymers (e.g., polyacrylonitrile and polyvinyl alcohol) exhibit much larger specific surface areas and higher mechanical strengths compared to single-component cellulose or single-component synthetic polymer nanofibers. Thus, there is an observable synergistic effect when composite nanofibers contain both cellulose and non-cellulosic polymers.

Many polymers have been successfully electrospun into nanofibers, including (1) thermoplastic homopolymers such as vinyl polymers, acrylic polymers, polyamides, polyesters, polyethers, and polycarbonates; (2) thermoplastic copolymers such as vinyl-co-vinyl polymers, acrylic-co-acrylic copolymers, and vinyl-co-acrylic polymers; (3) elastomeric polymers such as triblock copolymer elastomers, polyurethane elastomers, and ethylene-propylene-diene elastomers; (4) high-performance polymers such as polyimides and aromatic polyamides; (5) liquid crystalline polymers such as poly(p-phenylene terephthalamide) and polyaramid; (6) textile polymers such as polyethylene terephthalate and polyacrylonitrile; (7) conductive polymers such as polyaniline; and (8) biocompatible polymers (i.e., "biopolymers") such as polycaprolactone, polylactide, chitosan, and polyglycolide. As noted above, polymers can also be copolymers of two or more of the above-listed polymer types.

Examples of additional polymers that can be added to hybrid nanofiber felts include polyacrylonitrile (PAN), polyimides, polyamides (nylon 6, nylon 6,6, nylon 6,10, etc.), polyesters (polyethylene terephthalate, etc.), and copolymers thereof electrospun as single-component nanofibers.

Composite Nanofibers In one embodiment of the present invention, hybrid nanofiber felts comprise composite nanofibers. This is in part because the electrospinning process used to produce nanofibers from single-component derivatized cellulose solutions can be unstable, resulting in low yields, low efficiency (long times and many interruptions), and low-quality nanofibers (broad size distribution, weak, etc.) with only one chemical functionality. Therefore, to efficiently produce large quantities of high-quality nanofiber felts with multiple functionalities, it may be necessary to combine the cellulose derivative with a non-cellulosic polymer that stabilizes the electrospinning process.

The non-cellulosic polymer of the composite nanofibers of the present invention may comprise a synthetic carbon-based polymer that is removable from the nanofiber felt by exposure to elevated temperature and/or solvent. Exposure of the nanofiber felt to a mixture of solvents or a combination of elevated temperature and solvent can be simultaneous or sequential. The presence of a non-cellulosic polymer in the electrospinning process also improves nanofiber stability and other aspects of the method described elsewhere herein.

Synthetic polymer nanofibers (e.g., those produced from vinyl and acrylic polymers) offer a wide range of chemical functionality for bioseparation applications. By combining different polymer units, the surface chemistry of the resulting fiber can be controlled as part of the electrospinning process, and functionality is imparted directly to the resulting nanofiber. Alternatively, and similar to conventional micrometer-scale fibers, the surface functionality of polymer nanofibers can be chemically modified after electrospinning to meet the specific functionality requirements for various bioseparation applications (discussed below). Synthetic polymer nanofibers offer an astounding range of potential functionalization chemistries, lending them to a wide variety of uses. Such functionalization chemistries are well known in the polymer arts. They also tolerate the rigorous cleaning regimes typically associated with bioprocessing.

Many polymers have been successfully electrospun into nanofibers, including (1) thermoplastic homopolymers such as vinyl polymers, acrylic polymers, polyamides, polyesters, polyethers, and polycarbonates; (2) thermoplastic copolymers such as vinyl-co-vinyl polymers, acrylic-co-acrylic copolymers, and vinyl-co-acrylic polymers; (3) elastomeric polymers such as triblock copolymer elastomers, polyurethane elastomers, and ethylene-propylene-diene elastomers; (4) high-performance polymers such as polyimides and aromatic polyamides; (5) liquid crystal polymers such as poly-p-phenylene terephthalamide and polyaramid; (6) textile polymers such as polyethylene terephthalate and polyacrylonitrile; (7) conductive polymers such as polyaniline; and (8) biocompatible polymers such as polycaprolactone, polylactide, and polyglycolide.

Exemplary non-cellulosic polymers for making composite nanofibers include, for example, polyethylene oxide, poly(vinylpyrrolidone), poly(vinyl acetate), poly(vinyl alcohol), polysaccharides (chitin, starch, etc.), polystyrene, and poly(methyl methacrylate).

The non-cellulosic polymer is typically present in the composite nanofiber in an amount of 49% by weight or less, e.g., 30% by weight, 25% by weight, etc.

Electrospinning is a technique that uses only electric force to drive the spinning process and produce polymer fibers from a solution or melt. Unlike traditional spinning techniques (e.g., solution spinning and melt spinning), which can produce fibers with diameters in the micrometer range (approximately 5-25 μm), electrospinning can produce fibers with diameters in the nanometer range. Electrospun polymer nanofibers have many surprising properties, including small fiber diameters and the associated large specific surface area, high degree of polymer orientation, and resulting excellent mechanical properties. Furthermore, felts made from electrospun polymer nanofibers exhibit controlled pore size compared to nanofibers produced using other fabrication techniques. Unlike nanorods, nanotubes, and nanowires, which are mostly produced by synthetic methods, electrospun nanofibers are produced by a "nanofabrication method," resulting in low-cost nanofibers that are relatively easy to assemble and process into applications.

In general, the formation of nanofibers is a delicate and complex balance of three main forces involved in the electrospinning process, including electric forces, surface tension, and viscoelastic forces. Of these three forces, electric forces always promote the formation of products with the largest surface area. Surface tension forces always promote the formation of products with the smallest surface area. Viscoelastic forces are the forces that change significantly with solvent evaporation and are the main reason that prevent the electrospun jet/filament from breaking into droplets. When electric forces dominate, viscoelastic forces act in opposition to the electric forces. When surface tension forces dominate, viscoelastic forces act in opposition to the surface tension forces.

Theoretically, the smallest nanofibers can be formed under two conditions: (1) when the excess charge density carried by the electrospinning jet is high, and (2) when the time is long enough to prevent capillary rupture of the jet/filament, and the viscoelastic force is high enough to prevent capillary rupture of the jet/filament, but low enough to allow the electric force to effectively stretch the jet. Under condition (1), it has been found that adding a soluble electrolyte to the spin dope (e.g., adding a strong electrolyte such as NaCl to an aqueous polyethylene oxide solution) can significantly increase the excess charge density carried by the jet, resulting in the formation of nanofibers with smaller diameters. However, this method also has negative effects, such as (a) a decrease in flow rate and the resulting decrease in nanofiber productivity, and (b) contamination of the prepared nanofibers with the electrolyte. It can be difficult to remove the electrolyte without sacrificing the nanofiber properties.

Condition (2) requires further understanding of jet solidification. In general, jet solidification is closely related to the volatility of the solvent. If the solvent volatility is too high, the time to effectively draw the electrospun jet/filament is short. Therefore, fibers with a relatively large diameter are obtained. If the solvent volatility is too low, the electrospun jet/filament may break into small droplets upon drawing. Therefore, beads and/or beaded fibers are obtained.

The electrospinning process generally involves three steps: (1) initiation of the electrospinning jet/filament and its extension along a linear trajectory; (2) development of a bending instability in the jet and further elongation (which allows the jet to become highly elongated, following looped and spiral paths); and (3) solidification of the jet by solvent evaporation or cooling (which results in the formation of nanofibers). Figure 1 shows a schematic diagram of the electrospinning process (Hao Fong, In Polymeric Nanostructures and Their Applications, Volume 2: Applications: Chapter 11, Electrospun Polymer, Ceramic, Carbon/Graphite Nanofibers and Their Applications, Hari S. Nalwa Editor, American Scientific Publishers, Los Angeles, CA (ISBN: 1-58883-070-5), 2007, pp. 451-474).

An exemplary electrospinning process can be generally described as follows:

Step 1: As shown in Figure 1, a spin dope (e.g., a polymer solution) is placed in a vessel equipped with a spinneret (1), and high voltage DC (2), typically in the range of 5 to 40 kilovolts, is applied to the solution through electrodes (e.g., copper wires) (3). An electrically grounded collector (4) is placed at a certain distance (known as the gap distance) (5) from the spinneret. The gap distance can range from a few centimeters to one meter. When the electrostatic field reaches a critical value and the electrical forces overcome the surface tension and viscoelastic forces, a jet/filament is ejected and travels in a straight line for a certain distance (known as the jet length).

Step 2: The jet then begins to bend, forming a helical loop. This phenomenon is called the "bending (or whipping) instability." Typically, the bending instability causes the jet to extend in length by more than 10,000 times in a very short time (less than 50 ms). Therefore, the extension rate during bending instability is extremely high (up to 1,000,000 s ^-1 ). This extremely fast extension rate effectively stretches the polymer chains and tightly aligns them along the nanofiber axis.

Step 3: The jet solidifies either by evaporation of the solvent or when the melt cools below the solid-liquid transition temperature. The longer the solidification time, the longer the jet can be. The solidification time is related to many factors, including the solvent vapor pressure, the solvent diffusivity, the volume charge density carried by the jet, and the strength of the applied electrostatic field.

Optional Post-Electrospinning Treatment After the collected nanofibers have been solidified, there are several steps that can be taken to "customize" the nanofibers for a particular use. Exemplary additional steps are described below.

a. Removal of the First Non-Cellulosic Polymer In some cases, one or more of the polymers present in the composite nanofibers, particularly the non-cellulosic polymer, can be removed using high heat and/or solvent(s). Removal of the first non-cellulosic polymer adds surface area and improves the porosity of the remaining cellulosic polymer. This is because, after removal of the non-cellulosic polymer, the cellulosic polymer is left with size-controlled "pores" where the non-cellulosic polymer previously occupied space. This additional "void space" provides a larger surface area to the resulting nanofiber felt, which can, for example, increase adsorptive binding capacity for separations, improve size-based separation selectivity, and improve throughput from the additional porosity. Removal of the non-cellulosic polymer eliminates the opportunity for multiple functionalities (present in the composite nanofibers) to be directly present on the remaining cellulosic polymer nanofibers.

b. Cellulose Regeneration After preparing the "as electrospun" nanofibers, the derivatized cellulose can be converted to cellulose by a regeneration process. Regenerated cellulose has the same properties as the previously described pure native cellulose. The regeneration process is completed by contacting the nanofibers containing the derivatized cellulose with, for example, a strong base (e.g., sodium hydroxide) or other solvent. After the regeneration reaction to convert to cellulose, the nanofibers can be washed to remove excess solvent used during the process.

c.Surface functionalization
After the hybrid nanofiber felt is prepared, the fiber surface can be functionalized, non-limiting examples of which include the addition of ion-exchange groups such as weak or strong acids and bases (e.g., carboxylic acids and amines), hydrophobic groups such as phenolic compounds, and affinity ligands such as antibodies or enzyme substrates.

For use in bioseparations, the hybrid nanofiber felts of the present invention are ideally biologically inert, meaning they should prevent nonspecific binding of insoluble solids such as cells and cell debris, as well as undesirable interactions with proteins, sugars, nucleic acids, viruses, and other soluble components present in many biologically derived systems.

Furthermore, nanofiber felts for use in bioseparations should exhibit several qualities: (1) small diameter fibers, allowing for the maximum amount of specific surface area (this criterion is most important for adsorption processes and less important for strictly size-based separations, as discussed below); (2) well-controlled, narrow pore size distribution among fibers, allowing for even flow rate distribution in adsorption applications and tight size cutoffs for size-based separations; (3) the fibers should have excellent mechanical and chemical stability to withstand potentially high operating pressures and harsh cleaning conditions; and (4) the fibers should have a well-defined, spatially consistent size and chemical composition.

For adsorption processes where macromolecular products such as proteins, nucleic acids, and viruses are primary targets, the extremely large specific surface area associated with nanofiber felts provides a vast number of potential binding sites for biosorptive separations. Nanofibers can be modified to contain a staggering number of binding sites, and adsorption occurs almost exclusively on the fiber surface, making them readily available and eliminating the need for interior diffusion of relatively large target molecules. When using traditional porous resin beads, interior diffusion can often limit the capacity of many adsorption processes for bioproducts. Furthermore, because nanofiber membranes can be fabricated with many different chemistries, adsorption ligands can be tailored to meet specific separation needs (e.g., ionicity, hydrophobicity, and affinity). In some cases, ligands can be incorporated into the nanofibers from the source material during electrospinning, or the surface can be chemically modified to provide the desired adsorbent after nanofiber production.

Two of the most important characteristics of the separation process are that (1) flow passes through the micropores and macropores of the felt (as opposed to densely packed resin beads), and (2) adsorption occurs at the surface of the fiber, eliminating the need for internal diffusion. These factors reduce concerns about high pressure drop at high flow rates and eliminate the slow intraparticle diffusion required for adsorption within resin beads. The binding capacity of currently available adsorbent felts for biomolecules is similar in size to resin beads, yet they can be operated at process flow rates more than 10 times faster than packed beds. These factors result in much faster process times and potentially higher binding levels for purifying valuable biological products. This is particularly desirable for large biomolecules (molecular weights greater than 250 kDa and/or hydrodynamic diameters between 20 and 300 nm), which are extremely difficult to purify using packed beds due to severe mass transfer limitations within the pores of resin beads.

The surface of the nanofiber felt of the present invention can be modified to provide ion exchange and hydrophobic interaction chemistry. Simple chemical modifications, such as sulfonation of polystyrene fibers with sulfuric acid, have been used to generate cation exchange media. Grafting, atom transfer radical polymerization (ATRP), and plasma treatment have been used to create ion exchange surface functional groups and three-dimensional tethers from various polymer substrates, including polypropylene, polyvinylidene difluoride, polysulfone, and others. Phenyl and butyl groups can also be introduced as hydrophobic interaction ligands. Often, the surface of polymer membranes must be further modified to increase hydrophilicity to prevent nonspecific binding. This has been achieved by incorporating poly(ethylene glycol) and other polyols onto the surface.

The ion exchange capacity of hybrid nanofiber felt can also be improved by introducing, for example, diethylaminoethyl (DEAE) groups as weak anion exchange ligands or carboxylic acids as weak cation exchange ligands.

d. Surface Functionalization with Antibacterial Substances In one embodiment of the present invention, the non-cellulosic polymer is polyacrylonitrile (PAN). PAN fibrous membranes have been widely adopted in filtration due to their thermal stability, high mechanical properties, and chemical resistance. Electrospun PAN nanofiber felts have been particularly important due to their small fiber diameter and associated large specific surface area, as well as their capacity to control pore size between nanofibers and incorporate antibacterial agents at the nanoscale. Antibacterial functionalized nanofiber felts have attracted increasing attention due to concerns about the quality and processing costs of purified water and/or filtered air. Water and air filters (especially those operated in dark, humid conditions) are constantly susceptible to attack from environmental microorganisms. Microorganisms (e.g., bacteria) that can be easily captured by filters rapidly multiply, forming biofilms. Microbial buildup on the filter surface impairs the quality of purified water and/or filtered air. Furthermore, their buildup negatively impacts water and/or air flow.

Furthermore, biofilm-contaminated filters are difficult to clean. High pressure is usually required during operation, which increases costs. In the reported method, antimicrobial agents (such as N-halamines or silver ions/nanoparticles) are directly incorporated into the spin dope, thus distributing the antimicrobial molecules/particles throughout the nanofibers (Xinbo Sun, Lifeng Zhang, Zhengbing Cao, Ying Deng, Li Liu, Hao Fong, and Yuyu Sun. "Electrospun Composite Nanofiber Fabrics Containing Uniformly Dispersed Antimicrobial Agents as an Innovative Type of Polymeric Materials with Superior Anti-Infective Efficacy". ACS Applied Materials and Interfaces, 2(4), 952-956, 2010).

However, this often leads to processing issues, primarily because high antibacterial agent content can severely affect the electrospinning process and/or impair the properties of the resulting nanofibers. A potential solution to these issues is to introduce antibacterial functionality onto the nanofiber surface after the nanofibers are produced (Lifeng Zhang, Jie Luo, Todd J. Menkhaus, Hemanthram Varadaraju, Yuyu Sun, and Hao Fong. "Antimicrobial Nano-fibrous Membranes Developed from Electrospun Polyacrylonitrile Nanofibers". Journal of Membrane Science, 369, 499-505, 2011).

It is known that the nitrile (-C≡N) group in PAN can be chemically converted to an amidoxime (-C(NH ₂ )=NOH) group. The amidoxime group can coordinate with a wide range of metal ions, including silver ions, and can reduce the coordinated silver ions to silver nanoparticles. Both silver ions and silver nanoparticles are antibacterial agents with high antibacterial effects.

e. Other Examples A promising alternative to packed bed chromatography and other separation techniques is the use of the hybrid nanofiber felts of the present invention as selective adsorption membranes. This style of adsorption utilizes the nanofiber felt as a support for the ligands used in the selective adsorption process.

Selective adsorption involves "active" surface functionalization of hybrid nanofiber felts, allowing for the direct capture (adsorption) of target substances. Such modification is simplified when hybrid nanofiber felts contain chemical moieties that are relatively easy to chemically modify to provide adsorption sites on their surfaces.

Unlike modifying nanofiber surfaces for ion-exchange and hydrophobic interaction functionality, incorporating affinity ligands onto nanofibers can be more challenging. This method often requires first modifying the surface to create coupling sites for ligand immobilization, followed by binding of the ligand to the active sites. Importantly, both the initial surface modification and the ligand coupling should be robust so as not to leach out during processing.

In some cases, the simple carboxyl groups of grafted methacrylic acid to the surface can serve as active coupling sites by creating covalent amide bonds between the functionalized carboxyl groups and exposed amine groups on the protein ligand. Similarly, oxidation of cellulose (if properly controlled) can provide aldehyde groups on the fiber surface that can form covalent bonds to primary amines of proteins (including Protein A and Protein G), particularly through the amino acid lysine. In other cases, surface functionalization with common affinity dyes (e.g., Cibacron Blue, which can bind several proteins) can be directly coupled to cellulose nanofibers.

More elaborately, bioactive moieties for protein ligand immobilization can be incorporated into the nanofiber backbone during nanofelt construction. One example of this is the use of polyethylene glycol (PEG) as a block copolymer with poly(D,L-lactide) (PDLLA). After electrospinning, the glycol can be coupled with biocytin (which can interact with streptavidin fusion proteins) to create affinity nanofibers. Similarly, a polycaprolactone (PCL) and poly(D,L-lactic-co-glycolic acid)-b-PEG-NH2 (PLGA-b-PEF-NH2) diblock copolymer containing surface-aminated nanofibers can be created for protein coupling using homobifunctional coupling agents. Finally, in some cases, inherent active sites associated with some nanofiber matrices can be utilized. For example, concanavalin A (an affinity tag for lectins associated with glycol-protein and/or other glycol conjugates) has been successfully coupled to chitosan-based nanofibers.

Other techniques for attaching specific ligands to cellulosic compounds and/or synthetic polymers are known in the chemical arts.

Size-Based Separation As an orthogonal purification mechanism to adsorption, size-based separation is also routinely used in downstream bioprocessing. Depth filtration and microfiltration are common operations used in fermentation broth clarification, removing cells (approximately 1-20 μm) and cell debris (0.1-1 μm) from bioreactor slurries. Membrane-based nanofiltration is used for virus exclusion and/or purification of virus particles between 20 and 200 nm, while ultrafiltration is commonly used for protein concentration and purification. In each case, several properties of the separation media are desirable. First, a well-defined size cutoff is desirable for tightly controlled separations. Second, highly porous materials are necessary for high-throughput processing without excessive pressure requirements to minimize operation time and/or membrane area requirements. Third, chemical and physical robustness is desirable for the harsh cleaning conditions and operation under moderate pressures. Nanofiber felts present surprisingly good opportunities as advanced size-based separation media because they can be produced inexpensively in large quantities from mechanically and chemically strong fibers with well-controlled pore sizes within the fibers (or as hollow fibers). Polymer nanofibers generally exhibit minimal nonspecific binding but can suffer from lower chemical robustness than carbon and ceramic fibers. Ceramic fibers suffer from brittleness and the potential for significant nonspecific adsorption of biomass/bioparticles with attendant fouling, but can withstand harsh regeneration conditions.

To date, nanofiber meshes for size-based separation have primarily been applied to the isolation of nanometer- and micrometer-scale biological particles (or surrogates) via depth filtration mechanisms. The increased specific surface area of nanofibers within the filter mat provides a more tortuous path, increasing the chances of capturing desired particles from solution while maintaining high porosity. Polymer, carbon, and ceramic nanofibers have all been evaluated, and all have been successful in separating desired particle sizes from mixtures while maintaining high flux rates. Specifically, electrospun nanofibers made from polyvinylidene fluoride (PVDF) and nylon 6 were able to remove polystyrene particles between 0.5 and 10 μm. Ceramic nanofiber meshes have perhaps been most widely used. One example shows that the combination of large titanatate nanofibers and smaller boehmite nanofibers was capable of very high flux rates (1000 L/m2 h) at a relatively low driving force (20 kPa), removing virtually all particles larger than 60 nm from solution. It should be noted that many applications of micro- and nano-depth filtration also rely on the chemical adsorption of particles to surfaces, and that nanofibers can be easily manufactured to specifically adsorb desired impurities.

Nanofelt Construction/Configuration By utilizing fibers with diameters in the submicron to nanometer range (1-1000 nm, referred to as "nanofiber" felts), the available surface area within a given bed volume for potential binding will be greatly increased by two orders of magnitude. By controlling the pore size of the nanofiber felt, the pressure drop and hydrodynamic flow properties can also be controlled, making it as efficient as microfiber felts.

Furthermore, the pore sizes of the fibers in the felt typically have a tight pore size distribution (more than 90% of the nanofibers fall in the 100 nm to 500 nm range), preventing channeling and retaining only species above the desired size cutoff for filtration operations. Finally, nanofiber felts are typically mechanically strong enough to operate under conditions of potentially large pressure drops (up to 100 psi) and high flow rates (flux values exceeding 30 L/(min·m ² )), and chemically robust enough to withstand potentially harsh cleaning regimes (often involving strong acids, bases, and organic solvents) without failure.

In one embodiment, the nanofiber felt is comprised of composite nanofibers (one derivatized cellulose polymer plus one non-cellulose polymer) and single component nanofibers (non-cellulosic polymers). However, as noted above, the hybrid nanofelts of the present invention can be formed from various combinations of polymers and nanofibers. Examples of these include, for example:
A composite nanofiber felt, in which all of the nanofibers in the felt consist of a single type of composite nanofiber made from a co-extruded blend of a backbone polymer and a first non-cellulosic polymer.
- A nanofiber felt consisting of at least two different single-component nanofibers. - A nanofiber felt consisting of at least one single-component nanofiber and at least one composite nanofiber.

In addition to the nanofelt configurations described above, the nanofelts of the present invention may also contain microfibers to add stability, strength, and to tailor other physical properties of the felt for use in specific applications. Compared to single-component nanofiber felts, the hybrid nanofiber felts of the present invention exhibit the following exemplary improved properties:

[Example]
[Example 1]
Preparation of Prior Art Cellulose Acetate Monocomponent Nanofiber Felts Cellulose acetate monocomponent nanofiber felts were fabricated as described in "Handbook of Membrane Research, Chapter 3, Applications of Electrospun Nanofiber Membranes for Bioseparations," Todd J. Menkhaus, et al., Nova Science Publishers, Inc., edited by Stephan V. Gorley. Cellulose acetate (average molecular weight approximately 30,000 g/mol), NaOH, NaCl, acetone, N,N-dimethylacetamide (DMAc), and N,N-dimethylformamide (DMF) were purchased from Sigma-Aldrich Co. (Milwaukee, WI). 2-(Diethylamino)ethyl chloride hydrochloride (DAECH) with 98% purity was purchased from Alfa Aesar Co. (Ward Hill, MA).

A 15% (mass fraction) solution of cellulose acetate in acetone/DMAc (2/1 mass ratio) was prepared at room temperature. The solution was added to a syringe. The electrospinning setup included a high-voltage power supply and a laboratory-produced roller. During electrospinning, a positive voltage of 15 kV was applied to the needle, and a flow rate of 1.0 mL/h was maintained using a syringe pump. The cellulose acetate nanofibers were collected as a randomly layered felt on an electrically grounded aluminum foil covering the roller. A heat lamp was used to dry the nanofiber felt during electrospinning, and after electrospinning, the felt was further dried in a vacuum oven. Overall, the electrospinning process was relatively unstable and was frequently interrupted at approximately 2-h intervals. The collected cellulose acetate nanofiber felt had a thickness of approximately 225 μm and a mass per unit area of approximately 60 g/ ^m² .

The as-electrospun cellulose acetate nanofiber felt was first hydrolyzed/deacetylated by immersion in a 0.05M NaOH aqueous solution for 24 hours. The product, referred to as regenerated cellulose nanofiber felt, was then rinsed three times in distilled water and dried in a vacuum oven at 60°C. The sample was then immersed in a 15% (mass fraction) DAECH aqueous solution for 10 minutes and subsequently dried at 60°C. The sample was then immersed in a 0.5M NaOH aqueous solution at 90°C for 10 minutes. The sample was then rinsed three times in distilled water and dried at 60°C to obtain DEAE anion-exchanged cellulose nanofiber felt.

[Example 2]
Preparation of hybrid nanofiber felts of CA/PEO composite nanofibers and PAN single-component nanofibers Cellulose acetate (CA), polyethylene oxide (PEO), chloroform (CHCl ₃ ), dimethylformamide (DMF), polyacrylonitrile (PAN), and diethylaminoethyl chloride were purchased from Sigma-Aldrich Co. (Milwaukee, WI).

PAN and CA+PEO spin dopes were prepared separately. Briefly, for the preparation of PAN spin dope, PAN was dissolved in DMF to form a solution. For the CA+PEO spin dope, CA plus PEO was prepared in _CHCl3 /DMF with diethylaminoethyl chloride.

During the electrospinning process, two syringes filled with PAN or CA+PEO spin dope were placed on opposite sides of a laboratory-produced roller. Overall, the electrospinning process was highly stable and sustainable for long periods (>48 hours), and electrospun hybrid nanofibrous mats (self-supporting or supported on medical cotton gauze) consisting of CA+PEO composite nanofibers and PAN nanofibers were collected on an electrically grounded aluminum foil covering the roller.

The as-electrospun CA+PEO+PAN hybrid nanofibrous mat was then annealed for 24 hours to complete the phase dispersion of CA and PEO. The mat was then hydrolyzed/deacetylated by immersing it in an aqueous NaOH solution for 24 hours. The resulting hybrid nanofibrous mat, consisting of regenerated cellulose nanofibers and PAN nanofibers, was rinsed with distilled water and dried.

[Example 3]
Preparation of hybrid nanofiber felts of CA/PVP composite nanofibers and PAN single-component nanofibers Cellulose acetate (CA), polyvinylpyrrolidone (PVP), chloroform (CHCl ₃ ), dimethylformamide (DMF), polyacrylonitrile (PAN), and diethylaminoethyl chloride were purchased from Sigma-Aldrich Co. (Milwaukee, WI).

PAN and CA+PVP spin dopes were prepared separately. Briefly, for the preparation of PAN spin dope, PAN was dissolved in DMF to form a solution. For the CA+PEO spin dope, CA plus PVP was prepared in _CHCl3 /DMF with diethylaminoethyl chloride.

During the electrospinning process, two syringes containing either PAN or CA+PVP spin dope were placed on opposite sides of a laboratory-produced roller. Overall, the electrospinning process was highly stable and sustainable for long periods (>48 hours), and electrospun hybrid nanofibrous mats (self-supported or supported on medical cotton gauze) consisting of CA+PVP composite nanofibers and PAN nanofibers were collected on an electrically grounded aluminum foil covering the roller.

The as-electrospun CA+PVP+PAN hybrid nanofibrous mat was then annealed for 24 hours to complete the phase dispersion of CA and PVP. The mat was then hydrolyzed/deacetylated by immersing it in an aqueous NaOH solution for 24 hours. This process also dissolves/removes the PVP from the mat. The resulting hybrid nanofibrous mat, consisting of regenerated cellulose nanofibers and PAN nanofibers, was rinsed with distilled water and dried.

[Example 4]
Preparation of hybrid nanofiber felts of CA/PEO composite nanofibers and nylon 6 single-component nanofibers Cellulose acetate (CA), polyethylene oxide (PEO), chloroform (CHCl ₃ ), hexafluoroisopropanol (HFIP), nylon 6, and diethylaminoethyl chloride were purchased from Sigma-Aldrich Co. (Milwaukee, WI).

Nylon 6 and CA+PEO spin dopes were prepared separately. Briefly, for the preparation of nylon 6 spin dope, the polymer was dissolved in HFIP. For the CA+PEO spin dope, CA plus PEO was prepared in _CHCl3 /DMF with diethylaminoethyl chloride.

During the electrospinning process, two syringes containing either nylon 6 or CA+PEO spin dope were placed on opposite sides of a laboratory-produced roller. Overall, the electrospinning process was highly stable and sustainable for long periods (>48 hours), and electrospun hybrid nanofibrous mats (freestanding or supported on medical cotton gauze) consisting of CA+PEO composite nanofibers and nylon 6 nanofibers were collected on an electrically grounded aluminum foil covering the roller.

The as-electrospun CA+PEO+nylon 6 hybrid nanofibrous mat was then annealed for 24 hours to complete the phase dispersion of CA and PEO. The mat was then hydrolyzed/deacetylated by immersing it in an aqueous NaOH solution for 24 hours. This process also dissolves/removes the PEO from the mat. The resulting hybrid nanofibrous mat, consisting of regenerated cellulose nanofibers and nylon 6 nanofibers, was rinsed with distilled water and dried.

[Example 5]
Preparation of hybrid nanofiber felts of CA/PVP composite nanofibers and nylon 6 single-component nanofibers Cellulose acetate (CA), polyvinylpyrrolidone (PVP), chloroform (CHCl ₃ ), hexafluoroisopropanol (HFIP), nylon 6, and diethylaminoethyl chloride were purchased from Sigma-Aldrich Co. (Milwaukee, WI).

Nylon 6 and CA+PVP spin dopes were prepared separately. Briefly, for the preparation of nylon 6 spin dope, nylon 6 was dissolved in HFIP to form a solution. For the CA+PEO spin dope, CA plus PVP was prepared in _CHCl3 /DMF with diethylaminoethyl chloride.

During the electrospinning process, two syringes containing either nylon 6 or CA+PVP spin dope were placed on opposite sides of a laboratory-produced roller. Overall, the electrospinning process was highly stable and sustainable for long periods (>48 hours), and electrospun hybrid nanofibrous mats (self-supported or supported on medical cotton gauze) consisting of CA+PVP composite nanofibers and nylon 6 nanofibers were collected on an electrically grounded aluminum foil covering the roller.

The as-electrospun CA+PVP+nylon 6 hybrid nanofibrous mat was then annealed for 24 hours to complete the phase dispersion of CA and PVP. The mat was then hydrolyzed/deacetylated by immersing it in an aqueous NaOH solution for 24 hours. This process also dissolves/removes the PVP from the mat. The resulting hybrid nanofibrous mat, consisting of regenerated cellulose nanofibers and nylon 6 nanofibers, was rinsed with distilled water and dried.

[Example 6]
Performance Evaluation of Single-Component and Hybrid Nanofiber Felts Batch adsorption, dynamic adsorption, flow distribution, and permeability studies using single-component nanofiber mats (Example 1) and hybrid nanofiber mats (Examples 2-6) were compared to commercially available regenerated cellulose adsorbent membranes and cotton balls, which underwent the same post-electrospinning treatments as the single-component and hybrid nanofiber felts before testing.

Batch adsorption experiments were completed to determine Langmuir equilibrium adsorption isotherms. For batch analysis, single-component felt, hybrid felt, commercial cellulose membrane, and cotton balls were rinsed with buffer, cut into individual pieces of approximately 1 ^cm² , and weighed. For each medium, 10 individual pieces (approximately 100 mg) were then placed in a 15 milliliter (mL) centrifuge tube. A stock solution of the target protein was prepared at 2.0 mg/mL by mixing a known mass of lyophilized protein with buffer. The appropriate combination of stock solution and buffer was added to each test tube containing the cut felt pieces, commercial cellulose membrane, or cotton balls, resulting in a final liquid volume of 14 mL and initial protein concentrations between 0.0 mg/mL and 2.0 mg/mL protein. A 1.0 mL liquid sample was immediately withdrawn from each of the different initial protein concentrations and measured by UV absorbance at 280 nm. The samples were then placed in an end-over-end mixer rotating at approximately 40 revolutions per minute (rpm). After mixing for a minimum of 24 hours, the liquid from each sample was removed, and the protein concentration was determined by UV-280 nm absorbance using a Genesys 10 UV spectrophotometer purchased from Thermo Electron Corporation (Madison, WI). By difference, the protein adsorbed to the felt could be calculated. Test tubes with 2.0 mg protein/mL and no felt were also prepared to assess the potential for protein adsorption to the test tube surface. No adsorption to the test tube surface was observed. Similarly, controls were monitored to assess the possibility of leached chemicals that could contribute to the UV-280 nm absorbance and nonspecific binding of protein to the underivatized membrane (after regeneration with NaOH). No leaching or nonspecific binding was observed for any of the samples. Langmuir adsorption isotherms were then generated, and modeling constants (Q _max and K _d ) were determined by least-squares regression fit to the following equation:

where Q is the equilibrium concentration of protein adsorbed onto the membrane (mg/g), _Qmax is the asymptotic maximum saturation capacity, C is the liquid-phase equilibrium concentration of protein (mg/mL), and _Kd is the desorption constant (mg/mL). A high Q and a low _Kd indicate an efficient adsorption process. Following adsorption analysis, the liquid was decanted, and the felt was washed with 14 mL of buffer for 1 hour with mixing. After removing the wash, a buffer containing 1 M NaCl was added to the test tube, mixed for 1 hour, and the liquid was sampled for protein concentration by UV-280 nm absorbance. The elution percentage was calculated based on the amount of binding observed during the adsorption period studied.

The results, shown in Table 2 below, indicate that the electrospun hybrid nanofiber felt had the highest capacity, followed by the single-component cellulose nanofiber felt. Both nanofiber-based adsorbent materials had significantly higher saturation capacities than either of the commercially available cellulose-based adsorbent media at all equilibrium liquid phase concentrations. The regenerated cellulose microfiber felt and cotton ball exhibited the lowest binding capacities. The high specific surface area combined with the unique nanofiber felt morphology results in increased binding capacity. In all cases, quantitative elution of proteins was achieved.

The permeability of buffers through single-component and hybrid nanofiber felts and commercial regenerated cellulose adsorbent membranes was measured. Higher permeability values indicate higher throughput (faster processing times) and/or the ability to operate at lower pressures, both of which provide significant benefits to the manufacturing process. A small-scale "coin" membrane adsorbent holder was utilized for all experiments. The unit allowed for an effective filtration area of approximately 1.5 ^cm² and was sealed with an O-ring to prevent leakage. First, the pressure drop of the system alone, with the membrane holder in place but no membrane present, was evaluated at flow rates ranging from 2.0 mL/min to 30.0 mL/min. Next, layers of nanofiber felt or commercial membranes were sequentially added to the unit while measuring the pressure drop at various flow rates. For each felt/membrane, 1, 3, 5, 7, and 9 layers were evaluated. The permeability of the felt/membrane at each flow rate was calculated by subtracting the system pressure drop from the pressure drop measured with the felt/membrane in place. A minimum of five flow rates and corresponding pressure readings were taken for each of the different layer numbers.

As shown in Table 3 below, the permeability of buffer through the hybrid nanofiber felt stack was significantly higher than that of the single-component felt, which in turn was at least five times higher than that of the corresponding commercially available regenerated cellulose sample. Also, for comparison, a 15 cm packed bed of Sepharose Fast Flow was reported by the manufacturer to have a permeability of approximately 7 L/(min·m ² ·10 ⁵ Pa), similar to that of functionalized commercial membranes but much lower than that of nanofelt.

System dispersion analysis was performed on hybrid and single-component nanofiber felts and a commercially available regenerated cellulose adsorption membrane to determine the degree of axial mixing with different numbers of layers in place. Lower axial mixing (better flow dispersion) is desirable to minimize channeling and premature leakage during the adsorption process. The same configuration used for the permeability analysis was used for the system dispersion study, except the flow rate was maintained at 1.0 mL/min throughout the process. After equilibrating the felt/membrane stack with buffer, a solution of 1% (by volume) acetone in buffer was added to the system. Online absorbance at UV-280 nm was monitored, and the resulting curves were analyzed to calculate the Peclet (Pe) number by least-squares fit of the following equation:

where _Cout is the effluent 280 nm absorbance, _Cin is the inlet 280 nm absorbance, V is the volume of acetone solution added, and _V50 is the volume where _Cout / _Cin = 0.50. Larger Pe values were used to demonstrate the desirable properties of more closely approximating plug flow (lower axial mixing and better flow dispersion).

Table 4 below summarizes the Pe number results determined for hybrid and single-component nanofiber mats with various layer numbers and a commercially available regenerated cellulose felt/membrane layer. The results indicate that the nanofiber felts produced in this study have similar hydrodynamics.

Dynamic breakthrough analyses were completed to evaluate adsorption efficiency when operated under flow conditions. Higher capacity at low percent breakthrough indicates a more efficient adsorbent material. Dynamic breakthrough experiments were completed using a Pall Mustang coin holder according to the manufacturer's recommendations. Nine layers of either nanofiber felt or commercial membrane were used in the analyses. All experiments were operated with an AKTA Purifier (GE Healthcare, Piscataway, NJ) with online measurement of UV-280 nm absorbance, pH, and conductivity, and controlled by Unicorn software version 5.01. Fractions were automatically collected by the system in 0.60 mL increments (approximately 2 bed volumes). A minimum of 10 bed volumes was used for equilibration. Step elution into 100% Buffer B (equilibration buffer supplemented with 1.0 M NaCl) was used for each experiment. The flow rate was maintained at 1.0 mL/min for all dynamic breakthrough studies. Protein stock solution, prepared at 1.5 mg/mL in buffer, was loaded until 100% breakthrough was achieved. The felt was then washed with a minimum of 10 bed volumes of buffer before desorption. All eluents (load, wash, and elution flow-through) were collected and weighed to determine volume, and protein concentration was analyzed by UV-280 nm absorbance. Protein mass balance was then calculated based on the loaded volume and all fractions collected during the process.

The ultimate performance evaluation of any adsorption system is a dynamic breakthrough analysis, which combines equilibrium binding capacity, adsorption kinetics, and system dispersion. It also directly applies capacity to a flow-through mode of operation, where the bound molecules do not need to be selectively eluted from other impurities.

Table 5 below shows the dynamic binding capacity of proteins at 10% leakage for nanofiber felt and a commercially available regenerated cellulose adsorbent membrane. The hybrid nanofiber mat had a substantially higher dynamic capacity than any other adsorption media evaluated. Furthermore, the elution results indicated that, within experimental uncertainty, protein elution was complete for each adsorption system, with no observed loss in the overall mass balance.

The foregoing examples are set forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the composition embodiments, and are not intended to limit the scope of what the inventors regard as the invention. Modifications of the above-described modes of carrying out the invention (which would be obvious to those of ordinary skill in the art) are intended to be within the scope of the following claims. All publications, patents, and patent applications cited herein are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.

The present disclosure also discloses the following embodiments.
[Embodiment 1]
An electrospun hybrid nanofiber felt comprising composite nanofibers and single-component nanofibers, wherein the composite nanofibers comprise a mixture of derivatized cellulose and a first non-cellulosic polymer, and the single-component nanofibers comprise a second non-cellulosic polymer, and the first and second non-cellulosic polymers are differentially removable from the nanofiber felt.
[Embodiment 2]
2. The nanofiber felt of embodiment 1, wherein the derivatized cellulose is selected from the group consisting of organic esters of cellulose, inorganic esters of cellulose, and alkylcelluloses.
[Embodiment 3]
3. The nanofiber felt of embodiment 2, wherein the derivatized cellulose comprises an organic ester of cellulose selected from the group consisting of cellulose acetate, cellulose triacetate, and cellulose propionate.
[Embodiment 4]
3. The nanofiber felt of embodiment 2, wherein the derivatized cellulose comprises an inorganic ester of cellulose selected from the group consisting of cellulose nitrate and cellulose sulfate.
[Embodiment 5]
3. The nanofiber felt of embodiment 2, wherein the derivatized cellulose comprises an alkyl cellulose selected from the group consisting of hydroxyethyl cellulose and carboxymethyl cellulose.
[Embodiment 6]
2. The nanofiber felt of embodiment 1, wherein the first non-cellulosic polymer comprises a synthetic polymer.
[Embodiment 7]
7. The nanofiber felt according to embodiment 6, wherein the synthetic polymer is selected from the group consisting of vinyl polymers, polyethers, acrylic polymers, polyesters, polycarbonates and polyurethanes.
[Embodiment 8]
2. The nanofiber felt of embodiment 1, wherein the first non-cellulosic polymer comprises a natural polymer.
[Embodiment 9]
9. The nanofiber felt of embodiment 8, wherein the natural polymer is selected from the group consisting of polysaccharides, polyamides, and polylactides.
[Embodiment 10]
10. The nanofiber felt of embodiment 9, wherein the natural polymer is a polysaccharide selected from the group consisting of starch and chitin.
[Embodiment 11]
10. The nanofiber felt according to embodiment 9, wherein the natural polymer is a polyamide selected from the group consisting of proteins and gelatins.
[Embodiment 12]
2. The nanofiber felt of embodiment 1, wherein the second non-cellulosic polymer comprises a synthetic polymer.
[Embodiment 13]
13. The nanofiber felt according to embodiment 12, wherein the synthetic polymer is selected from the group consisting of vinyl polymers, polyamides, polyimides and polyesters.
[Embodiment 14]
7. The nanofiber felt of embodiment 6, wherein the first non-cellulosic polymer is a copolymer.
[Embodiment 15]
13. The nanofiber felt of embodiment 12, wherein the second non-cellulosic polymer is a copolymer.
[Embodiment 16]
Electrospinning method for making the nanofiber felt according to embodiment 1, comprising:
a) separately preparing a composite polymer spin dope and a single component polymer spin dope;
b) placing the spin dope into two different spinnerets;
c) applying a voltage to each spin dope using an electrode;
d) separately electrospinning the composite and single component nanofibers from the spinneret; and
e) collecting the solidified nanofibers as a randomly stacked or partially aligned nanofiber felt.
[Embodiment 17]
15. The method of embodiment 14, further comprising converting the derivatized cellulose to cellulose.
[Embodiment 18]
15. The method of embodiment 14, further comprising removing at least a portion of the non-cellulosic first polymer from the composite nanofibers in the nanofiber felt.
[Embodiment 19]
15. The method of embodiment 14, further comprising surface functionalizing one or more of the nanofibers in the nanofiber felt.
[Embodiment 20]
18. The method of embodiment 17, wherein the surface functionalization comprises binding of an affinity ligand.
[Embodiment 21]
1. A method for purifying a biomolecule from a fluid, comprising:
a) preparing a nanofiber felt according to embodiment 14;
b) flowing a fluid through the nanofiber felt; and
c) recovering the biomolecule from the nanofiber felt.

Claims (4)

1. A method for purifying a biomolecule from a fluid, comprising:
a) providing an electrospun hybrid nanofiber felt comprising composite nanofibers and single-component nanofibers, wherein the composite nanofibers comprise a mixture of cellulose and a first non-cellulosic polymer and the single-component nanofibers comprise a second non-cellulosic polymer, and wherein at least a portion of the first or second non-cellulosic polymer is removed from the nanofiber felt;
b) passing the fluid through the nanofiber felt; and
c) recovering said biomolecules from said nanofiber felt. The method of claim 1 , wherein the biomolecules include one or more of proteins, nucleic acids, carbohydrates, bacteria, viruses, and cells. The method of claim 1 , wherein one or more of the nanofibers comprises surface functional groups comprising affinity ligands, ion exchange groups, or hydrophobic groups. The method of claim 1 , wherein the cellulose comprises a derivatized cellulose.