Functionalized alumina adsorbent materials for removing contaminants from water

Abstract

A composition for removing a target substance from a fluid stream is provided. The composition can be used to remove a polyfluoroalkyl substance (PFAS) from water. The composition comprises a support material comprising alumina and a sorbent molecule, the sorbent molecule comprising a core polymer; wherein the core polymer is covalently attached to the support material; and wherein the sorbent molecule further comprises one or more covalently attached sorbent groups. Methods for removing a target substance, such as a PFAS, are also provided.

Description

TECHNICAL FIELD

[0001] The present invention relates to the use of polyamine chemically modified alumina-based substrate materials for the removal of target species from liquids, such as water, and methods for producing such materials. The invention also relates to materials for water treatment, sorbent media, amine surface functionalization, hydrophobic interactions, per- and polyfluoroalkyl substances (PFAS), water pollutants, and filtration equipment. BACKGROUND

[0002] There is a continuing need to remediate and recycle contaminated supplies of critical fluid solvents, such as water. There is a need to reuse and replenish existing resources rather than simply disposing of them. Global environmental protection regulations also require water supplies threatened by contamination from industrial activities to meet increasingly stringent purity standards. One such standard is NSF-53, established by the National Sanitation Foundation International, which sets standards for adsorption / filtration that occurs when a liquid, gas, or dissolved / suspended substance adheres to the surface or pores of a sorbent media. Carbon filters containing powdered carbon or activated carbon are examples of the type of media used in water filtration products that meet NSF-53.

[0003] Sources of environmentally damaging contaminants are diverse, including improperly treated wastewater from industrial plants and chemical processing facilities; surface runoff from agricultural areas containing fertilizers and pesticides; and fire-retardants used in cleaning detergents and fire-fighting foams. Many industrial chemical contaminants can persist in the environment for decades before degrading, causing great harm to plants, animals, and humans even at very low concentrations. The impact on ecosystems is also profound, with persistent contaminants often concentrating in the bodies of organisms at the upper levels of the food chain.

[0004] One particular class of persistent environmental contaminants includes halogenated organic compounds, such as per- and polyfluoroalkyl substances (PFAS). PFAS are organic fluorinated compounds that are considered chemically inert. They are persistent in the environment, and their use is controlled in many countries by the United Nations Framework Convention on Climate Change, the Kyoto Protocol, and the REACH regulation. Perfluorooctane sulfonate (PFOS) and its derivatives have been listed under the Stockholm Convention and are restricted under the Persistent Organic Pollutants (POP) regulation within the European Union (EU). PFOS and perfluorooctanoic acid (PFOA) are toxic PFAS compounds that have been widely used as surfactants and fire-retardants for fire-fighting foams and metal plating processes. Both PFOS and PFOA persist in the environment for a long period of time and are recognized as contaminants in most freshwater supplies worldwide.

[0005] Adsorption of PFAS compounds (e.g. PFOS and PFOA) onto granular or powdered activated carbon is the recommended solution for their removal from contaminated water at present. However, this method is both slow and inefficient. In particular, shorter chain PFAS contaminants rapidly “break-through” the activated carbon bed, meaning that large amounts of activated carbon are required, which must be frequently replaced once saturated with PFAS. Currently, adsorbed PFAS cannot be effectively washed off the activated carbon for “in situ” regeneration. Furthermore, a significant proportion of the activated carbon manufactured globally comes from fossil fuels, such as bituminous coal, which is activated by physical methods that release large amounts of carbon dioxide. Activated carbon is therefore an expensive, unsustainable and single-use solution to the problem of removing PFAS from contaminated water. Ion exchange methods are also commonly used, but this method also has a large carbon footprint and, again, reliability is an issue due to breakthrough. Operational lifetime and cost-effectiveness are also issues for ion exchange-based resins.

[0006] Previous efforts to remove PFAS contaminants from water include, for example, the use of high molecular weight polyamine (such as polyethyleneimine (PEI, typically 25 kDa)) functionalised cellulose support materials in combination with hydrophobic groups (see WO2017 / 203281). Such methods rely primarily on the adsorptive activity of the functional groups, rather than any special contribution from the support substrate material.

[0007] Alumina-based filtration membranes for the removal of perfluoroalkyl substances from water have also been considered, using alumina functionalised with linear fluorinated silanes containing 13 to 17 fluorine atoms (Johnson et al., ACS Omega, 2019, 4, 8001). While this membrane exhibits a high degree of specificity for the removal of PFAS from contaminated water, the amphiphilic silane molecules used as adsorbent molecules must be synthesised. This creates its own problem, as the total fluoro-waste produced in the synthesis of these adsorbent filtration membranes is greater than the actual mass of PFAS removed by the membrane in use. The improvement in performance is therefore offset by the environmental cost of the method and the subsequent commercial unviability.

[0008] There is a need to provide economic and reusable compositions and methods that are capable of removing low concentrations (<1 ppm) of target substances, particularly contaminating pollutants such as PFAS, from fluid streams (e.g. wastewater) or from the wider environment. The present invention aims to overcome the current challenges, including reducing the impact of industrial activities on aquatic environments, and achieve these objectives. SUMMARY

[0009] The present invention provides a further surprising development of the prior art, in particular with respect to new functionalization of alumina, to optimize and fine-tune various polyfluoroalkyl substances and perfluoroalkyl substances (PFAS) towards improved adsorption, indicating that the presently demonstrated surface functionalization can be widely applied to remediate PFAS contamination in water supplies.

[0010] A first aspect of the invention provides a composition for removing a target substance from a fluid stream, the composition comprising:

[0011] a support material comprising alumina; and

[0012] an adsorbent molecule comprising a core polymer;

[0013] wherein the core polymer is covalently linked to the support material; and

[0014] wherein the adsorbent molecule further comprises one or more covalently linked adsorbent groups.

[0015] Suitably, the core polymer is a linear polymer or a branched polymer.

[0016] A second aspect of the invention provides a composition for removing polyfluoroalkyl substances and perfluoroalkyl substances (PFAS) from an aqueous solution, the composition comprising:

[0017] a particulate support material comprising a bimodal alumina having mesopore and macropore content, wherein the bimodal alumina has a BET pore volume in the mesopore range of not less than about 0.20 cm 3 / g; and

[0018] an adsorbent molecule comprising a linear core polymer or a branched core polymer selected from one or more of: poly(allylamine), poly(methyl methacrylate), poly(vinyl alcohol), poly(vinylamine), poly(vinyl chloride), poly(vinyl imine), poly(2-vinylpyridine), poly(3-vinylpyridine), and poly(4-vinylpyridine);

[0019] wherein the core polymer is covalently linked to the support material; and

[0020] wherein the adsorbent molecule further comprises one or more covalently linked adsorbent groups.

[0021] A third aspect provides a method for removing a target substance from a fluid stream, suitably an aqueous feed stream, comprising contacting the feed stream with a composition as described herein. The target substance can comprise one or more PFAS. Typically, the aqueous feed stream is selected from: contaminated water, wastewater, groundwater, drinking water, seawater, and industrial or agricultural runoff.

[0022] In a fourth aspect, there is provided a method for removing a target substance from a fluid stream, the method comprising contacting the fluid stream with a composition comprising a support material consisting of particulate alumina, and an adsorbent molecule comprising a core polymer; wherein the core polymer is covalently attached to the support material; and wherein the adsorbent molecule further comprises one or more covalently attached adsorbent groups selected from one or more of the following groups: substituted or unsubstituted C1-C12 alkyl, substituted or unsubstituted C2-C12 alkenyl, substituted or unsubstituted C2-C12 alkynyl, substituted or unsubstituted C1-C12 alkoxy, substituted or unsubstituted C1-C12 acyl, substituted or unsubstituted arene, substituted or unsubstituted aryl, a heterocyclic group, and a hydrogen atom.

[0023] In a fifth aspect, the compositions described herein are configured within a filter, which can also consist of a bed or a packed column, through or across which the fluid stream passes. Suitably, the filter is used to adsorb one or more PFAS from a contaminated water source. The filter can be used in a point-of-use (POU) or point-of-entry (POE) water filtration system.

[0024] It is expressly intended within the scope of the present disclosure that the aspects, embodiments, examples, and alternatives listed in the foregoing paragraphs, claims, and / or the following description and drawings can be used alone or in any combination with one another. That is, all embodiments and / or features of any embodiment can be combined in any manner and / or combination, unless such features are incompatible. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 Batch test graphs showing removal of five different PFAS contaminants (adsorption %) from water samples are shown: Granular Activated Carbon (GAC), Resinex and Purolite represent comparators of existing known adsorbent materials. Untreated alumina (Alumina 1) was also tested as a comparator of an unfunctionalised support material. Two alumina samples were functionalised with core polymer 25KDa PEI at two levels of surface activation (10-08 and 10-10) and tested after further modification to incorporate C8 acyl adsorbent groups (10-12 and 10-13) respectively.

[0026] Figure 2 Batch test graphs showing removal of five different PFAS contaminants (adsorption %) from water samples treated with octanoyl (C8 acyl) PEI instead of a range of different alumina substrates are shown.

[0027] Figure 3Batch test plots showing removal of five different PFAS contaminants (adsorption %) from water samples treated with granular activated carbon (GAC), Resinex and Amberlite ion exchange chromatography, which represent comparative materials of existing known adsorbent materials. Two different alumina substrate materials (Alumina 1 and Alumina 4) functionalized with different molecular weight core polymers PEI were also tested.

[0028] Figure 4 Batch test plots showing removal of five different PFAS contaminants (adsorption %) from water samples treated with Alumina 1 functionalized with a 25 kDa PEI core polymer and then further modified by addition of adsorbent alkyl and quaternization of tertiary amines.

[0029] Figure 5 Batch test plots showing removal of five different PFAS contaminants (adsorption %) from water samples treated with granular activated carbon (GAC), Resinex and Lewatit, which represent comparative materials of existing known adsorbent materials. Alumina substrate materials (Alumina 1 in Figure 3 ) functionalized with core polymer PEI 25 kDa (Al-PEI) as comparative, and then further modified by addition of adsorbent alkyl groups of different length: C8 (acyl PEI-sub Al), C4 (butyl), C6 (hexyl) and C10 (decyl).

[0030] Figure 6 Batch test plots showing removal of five different PFAS contaminants (adsorption %) from water samples treated with granular activated carbon (GAC), Resinex and Lewatit, which represent comparative materials of existing known adsorbent materials. Two different alumina substrate materials (Alumina 1 and Alumina 4) functionalized with core polymer PEI 25 kDa and then further modified by addition of adsorbent alkyl groups of different length.

[0031] Figure 7 Batch test plots showing removal of five different PFAS contaminants (adsorption %) from water samples treated with granular activated carbon (GAC), Resinex and Amberlite, which represent comparative materials of existing known adsorbent materials. A material of class 1 according to embodiments of the present application, octanoyl PEI-substituted alumina 2 (10-73b), is compared to a material of class 3 according to another embodiment of the present application, octyl PEI-substituted alumina 2 (24-07b).

[0032] Figure 8Batch test plots showing removal of five different PFAS contaminants (adsorption %) from water samples treated with granular activated carbon (GAC), Resinex, and Amberlite, which represent comparators for existing known adsorbent materials. Four test compositions of embodiments of the invention were also tested using a range of alumina substrates.

[0033] Figure 9 Isotherm plots showing Figure 8 Isotherm plots for three best performing embodiments of small alumina adsorbent media (28-4a - 4c) against PAC (powdered activated carbon) for six PFAS contaminants.

[0034] Figure 10 Pressure drop vs. flow rate plots for 7 mm and 13 mm functionalized alumina deep membrane encapsulated adsorbent structure (MEAS) set-ups according to embodiments of the invention.

[0035] Figure 11 Breakthrough curve plots for PFOA in water in 7 mm MEAS set-ups.

[0036] Figure 12 Breakthrough curve plots for PFAS GenX in water in 7 mm MEAS set-ups.

[0037] Figure 13 shows breakthrough curves for (a) PFOA and (b) GenX for 13 mm MEAS set-ups.

[0038] Figure 14 Batch test plots showing removal of seven different PFAS contaminants (adsorption %) from water using hexyl-PEI functionalized alumina with either bromohexane or iodohexane as the alkylating agent.

[0039] Figure 15 (a) shows Hg porosity data plots for three different granular alumina materials before functionalization with adsorbent molecules; (b) shows batch test plots for removal of five different PFAS contaminants (adsorption %) from water using three of the same granular alumina functionalized with hexyl-PEI.

[0040] Figure 16 Hg porosity data plots for four different powdered alumina materials (alumina 8 and alumina A-C).

[0041] Figure 17 Batch test plots showing removal of four different PFAS contaminants (adsorption %) from water using four powdered aluminas that are the same as in Figure 16 but functionalized with hexyl-PEI, showing the effect of pore size.

[0042] Figure 18 Batch test plots showing removal of seven different PFAS contaminants from water using hexyl-PEI functionalized four spherical aluminas (S-1 to S-4) (absorption %).

[0043] Figure 19 Batch test plots showing removal of seven different PFAS contaminants from water using hexyl-PEI functionalized aluminas (using two different molecular weights of PEI) (absorption %).

[0044] Figure 20 Batch test plots showing removal of five different PFAS contaminants from water samples treated with granular activated carbon (GAC), Resinex, and Amberlite, which represent comparators of existing known adsorbent materials. Four test compositions of embodiments of the present invention using alkyl functionalized linear PEI were also tested in comparison to one test composition using alkyl functionalized branched PEI using alumina 8 substrate. DETAILED DESCRIPTION

[0045] The practice of the present application employs, unless otherwise indicated, conventional techniques of chemistry, material science, and process engineering, within the capabilities of one of ordinary skill in the art.

[0046] Before the present application is disclosed and described, it is to be understood that the

[0047] As used herein, the terms "comprising", "including", "containing", "have" and "has" mean including, but not limited to, any expressly identified elements. "Consisting essentially of when used herein, means including any expressly stated elements and those that do not materially affect the basic and novel characteristics of the listed elements, and can optionally include other elements. "Consisting of means excluding any element not specified. Embodiments defined by each of these terms are within the scope of the present application.

[0048] The term "target" or "target species" refers herein to a substance or compound that is desired to be removed or separated from a fluid. The target species can be dissolved (i.e., solute), suspended, emulsified, dispersed, or otherwise carried in the fluid, and thus can be soluble in the fluid, partially soluble in the fluid, or insoluble in the fluid. As described below, the target species can include contaminant species and / or valuable species that are desired to be removed from the target fluid and, in some cases, recovered.

[0049] The target substances contemplated herein can include "contaminants" or "contaminant substances". In the context of the present application, "contaminants" are intended to encompass substances that can be harmful to the health of humans or animals or to the environment. Accordingly, the derivative terms are defined accordingly, e.g., a contaminated fluid is a fluid that contains contaminant substances. Typically, the contaminants include one or more perfluoro and polyfluoroalkyl substances (PFAS), typically one or more perfluorinated carbons, optionally selected from perfluoro anionic surfactant compounds, including one or more selected from perfluoropentanoic acid (PFPeA), perfluorooctanoic acid (PFOA), perfluorobutane sulfonate (PFBS), perfluorohexane sulfonate (PFHS), perfluorohexanoic acid (PFHA), perfluorooctane sulfonate (PFOS), perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), 6:2 fluorotelomer sulfonic acid (6:2 FTSA), and hexafluoropropylene oxide dimer acid (HFPO-DA, also known as GenX chemical, chemical name 2,3,3,3-tetrafluoro-2-(heptafluoropropoxy)propanoic acid). In some embodiments, the contaminants include organic compounds, optionally including pharmaceutical or pesticide molecules selected from one or more of diclofenac, erythromycin, estrogen, oxadiazon, and thiamethoxam. In some embodiments, the contaminants can be metal ions or metalloid ions, optionally selected from copper, iron, lead, mercury, chromate, or arsenate.

[0050] The term "fluid stream" or "feed stream" refers to a flowable substance, suitably a liquid, including an aqueous solution, in which the target substances are dissolved, suspended, emulsified, dispersed, or otherwise carried. Typically, the liquid contains water, or is predominantly water-based, and can be in the form of contaminated water, wastewater, drinking water, seawater, and / or industrial or agricultural runoff.

[0051] The terms "sorption," "sorb," "sorbent," and derivatives thereof, as used herein, refer to the removal of a target species, such as a contaminant, from a fluid stream by combining the target species with a described modified carrier material. Sorption of a material can occur by any means, such as by adsorption to the surface of the material, which can occur by creating a chemical interaction between the target species and the carrier material, including electrostatic attraction, hydrophobic interaction, formation of covalent bonds, attachment, chelation, intermolecular interaction, hydrogen bonding, or other means. "Sorption" can also refer to the adsorption of a target species into a material. The target species can be physically entrapped within intermolecular spaces, pores, or other voids within the material. In particular, sorption can be adsorption that occurs by the formation of a chemical interaction between a target species molecule and a sorbent molecule as defined herein, where the sorbent material has been modified. This chemical interaction results in the sequestration of the target species within the sorbent material and outside of the fluid stream. Unless otherwise noted, the term "sorption" or derivatives thereof, as used herein, is not bound by any theory of limitation, but is intended to include sorption by other means as defined above.

[0052] The term "sorbent material" as defined herein refers to a material comprising a carrier or substrate material which also comprises sorbent molecules attached or bound thereto. The sorbent material is suitable for contacting a fluid stream comprising a target substance (e.g. a contaminant, which can be a PFAS) such that the target substance is adsorbed onto or otherwise absorbed from the fluid stream and sequestered by the sorbent material. Suitably, the sorbent material is configured within a filter / purifier and / or bed or packed column (e.g. comprising a plurality of stacked filters) and the fluid stream passes through or is passed through the filter, bed or packed column. The sorbent material can be configured within a mixed bed in combination with another sorbent material such as granular activated carbon or ion exchange resin. In one embodiment, the sorbent material is contained within a ready-made assembly (e.g. a filter cartridge) such that used sorbent material can be conveniently contained and similarly replaced or replenished with fresh or regenerated sorbent material as required. Alternatively, the sorbent material can be added as a dispersion to the fluid. The sorbent material can be particulate, that is to say, in the form of particles, flakes, beads, pellets or pastilles. The sorbent material can be a powder, which can advantageously provide a higher accessible surface area. The sorbent material can be incorporated into a membrane filter or a similar membrane filter. A membrane composed of an alumina substrate material can also be functionalised directly as described herein. In particular, a membrane or similar membrane product can be used to manufacture a filter. An advantage of such filters is that they can be made to a specific thickness and have a large surface area whilst still ensuring fluid passage when appropriately installed in a fluid flow path and reducing flow rate to a minimum. Furthermore, the filter can combine the functionality of the present invention with a particulate (size exclusion) removal capability. The filter bed or column can be occasionally back-flushed to clear accumulated blockages such as organics or scale which can reduce flow rate.

[0053] In one embodiment, the sorbent material is particulate, suitably in the form of particles, the average diameter size of the particles or granules (measured as the largest diameter of the particles) is greater than about 0.01 mm, suitably greater than about 0.1 mm, and typically less than about 5 mm, less than about 3 mm, and optionally less than about 1 mm, or even less than about 700 pm. In one embodiment of the present invention, the d50 average particle size of the particles is about 600 pm in diameter.

[0054] In another embodiment, the adsorbent material is a particulate, suitably in the form of a powder, the average diameter size of the particles (measured as the largest diameter of the particle) is greater than about 0.1 μιη, suitably greater than about 1 μιη, optionally greater than 50 μιη, and typically less than about 300 μιη. Typically, the powdered alumina suitable for use in embodiments of the present application has a d50 average particle size of about 100 μιη, optionally with a majority of the particles in the powder having a diameter greater than about 70 μιη, but less than about 300 μιη. The powdered alumina can include a powder having spherical particles.

[0055] In another embodiment, the adsorbent material is a particulate composition, suitably spherical or spherical, the average diameter size of the spheres (measured as the largest diameter of the particle) is greater than about 0.5 mm, suitably the size is greater than about 1 mm, and the size is up to about 5 mm.

[0056] In one embodiment of the present application, a composition for removing target species and / or contaminants from a fluid stream is provided. The composition includes an adsorbent material, the adsorbent material including a carrier material covalently linked to a target species adsorbent molecule. The carrier material has a high surface area to volume ratio, thus providing an efficient carrier for molecules capable of acting as adsorbents for the target species. The particulate adsorbent particles are designed to be set up as an adsorbent media for standard packed bed wastewater treatment. The particles have a certain porosity, but are hard, durable, and not easily degradable. The larger particulate adsorbent media is also suitable for large scale engineering applications (e.g., purifying water in industrial or municipal sites). However, the smaller particle size provides faster capture kinetics, allowing for the creation of broad spectrum target species removal products for point of use (POU) and / or point of entry (POE) markets.

[0057] According to embodiments of the present application, the sorbent material can be directly loaded into a POU device, such as a cartridge filter, or be compatible with third party technology so as to be packaged within a sandwich membrane (e.g., a functionalized alumina bed between porous metal foil membranes) that can then be configured into a POU device. The design of a typical POU device should be suitable for connection to a typical household tap / faucet or under-sink water supply pipe. The POU device and its components must withstand inlet pressures of up to 6 bar (87 psi) without leaking or structural damage. This pressure represents an upper limit in most home environments. Since taps / faucets have varying supply pressures, the POU device must also work at lower inlet pressure ranges, as low as 1.4 bar (20 psi). Typical device specified inlet pressure ranges are 1.4 bar to 5.5 bar (80 psi), with some devices rated up to 6.8 bar (99 psi). The minimum flow rate acceptable to most consumers is about 2 L / min, however some devices can treat at flow rates up to 3 L / min to 4 L / min. The service life of virtually all currently available products is 6 months (i.e., twice per year replacement). Given the unexpectedly high PFAS capacity of the compositions of the present application, a product life of 6 months to 12 months is expected to be readily achievable. To last 12 months, about 4400 L of water (about 1200 US gallons) should be treated, which requires that the device demonstrate, according to NSF-53 certification testing, the ability to adequately treat 5280 L of water (20% capacity increase) before PFAS breakthrough. In particular embodiments of the present application, the POU device can be connected between an existing conventional water purification cartridge (not intended to remove PFAS) and a tap / faucet. In this configuration, the POU device can provide an indication when replacement is needed - for example, according to time or volumetric flow through the device, or both. At this time, the user can replace the cartridge containing the composition of the present application without having to disassemble any plumbing connections. The spent cartridge can be regenerated according to the methods described herein.

[0058] For POE applications, where water enters a building, the system continuously treats the water of the entire water supply system of the building, the sorbent material can be contained within a fiberglass reinforced plastic (FRP) cylindrical tank or cylinder. The amount of sorbent material can be designed to be used in off-the-shelf FRP cylinder configurations that can typically hold about 10 L to 20 L of material, and the operational life of normal household use can reach one year or more before the sorbent material needs to be replaced or regenerated. A typical POE system can include multiple water treatment cylinders, and the sorbent material can be included as a subassembly within all of the cylinders, or as a dedicated assembly within one of the multiple cylinders.

[0059] As used herein, the term "alumina" refers to the various mineral forms of aluminum oxide / aluminum hydroxide. The term alumina includes, for example, corundum (AI2O3), boehmite (y- AI O(OH)), diaspore (a- AI O(OH)), and gibbsite (AI(OH)3) (e.g., bayerite, doyleite, and nordstrandite), and combinations thereof. The term alumina also includes the various phases and polymorphs of aluminum oxide / aluminum hydroxide - for example, gamma (y) alumina, alpha (a) alumina, theta (0) alumina.

[0060] The adsorbent compositions of the present application comprise a solid alumina support having a plurality of pores - i.e., cavities, channels, or interstices having a depth greater than a width. Pore is defined by pore width / size / diameter, generally taken to mean the distance between two opposing walls of the pore (e.g., the diameter of a cylindrical pore or the width of a slit-shaped pore). As with most porous materials, alumina can be classified as microporous, mesoporous, or macroporous. According to IUPAC, mesoporous materials are characterized by a predominance of pores having widths in the range of 2 nm to 50 nm (Roquerol et al. (1994) Pure & Appl. Chem. 66(8): 1739-1758).

[0061] In embodiments of the present application, the alumina support material consists of mesoporous alumina, wherein the majority of the pores have a diameter greater than 2 nm, suitably greater than 3 nm, typically greater than 4 nm, and optionally greater than 10 nm. By "majority" is meant that greater than 50% of the pore volume comprises pores within the defined mesoporous range, typically greater than 60%, 70%, 80%, and optionally greater than 90%.

[0062] In particular embodiments of the present application, the alumina support material consists of alumina having a bimodal porous structure, wherein a portion of the porosity falls within the mesoporous range and a portion falls within the macroporous range. Pores having a diameter in excess of about 50 nm are generally referred to as macropores. In embodiments of the present application, alumina having mesoporous characteristics as defined above and also having a higher proportion of macropores in the range of about 100 nm to about 20 μm, suitably about 0.5 μm to about 15 μm, and optionally about 1 μm to about 10 μm, is considered to exhibit advantageous adsorbent characteristics. Thus, in particular embodiments of the present application, a bimodal alumina support material is provided having a majority of the porosity in a first distribution of pores having diameters in the mesoporous range of about 2 nm to about 50 nm, and a second distribution of pores having diameters in the microporous range of about 1 μm to about 10 μm. Surprisingly, it has been found that such bimodal aluminas, when functionalized, provide a favorable balance between adsorptivity, structural flexibility, and optimal flow rates of liquids through beds composed of alumina particles. It will be appreciated that the present application is not limited to bimodal aluminas, but can also include multimodal aluminas having a porosity range falling within a combination of the mesoporous and macroporous ranges.

[0063] In particular embodiments of the application, the particle size of the alumina ranges from about 300 pm to 3000 pm for so-called larger particles, and from about 0.1 pm to 300 pm for so-called smaller particles, such as powders. As previously mentioned, the smaller particles show utility as components in the design and manufacture of POU filtration devices, while the larger particles can be used in packed beds for POE, industrial, or municipal scale purification plants. In particular embodiments of the application, the filtration device can comprise a combination of any one of the sorbent materials. Surface area and porosity can be determined by gas adsorption-desorption methods, such as the Brunauer-Emmett-Teller (BET) technique and mercury intrusion porosimetry, or generally a combination of these techniques. BET is a theory related to the adsorption of gases onto solid surfaces, and is essentially an extension of the Langmuir theory to include multilayer formation. Mercury (Hg) intrusion is a physical technique to assess pore size / volume by interpenetration of a material with liquid mercury.

[0064] The sorbent material is attached to a sorbent molecule consisting of a polymeric core, suitably a linear or branched polymer. The core polymer can be one or more selected from the group consisting of poly(allylamine), poly(methyl methacrylate), poly( vinyl alcohol), poly( vinyl amine), poly( vinyl chloride), poly( vinyl imine), poly(2-vinylpyridine), poly(3-vinylpyridine), and poly(4-vinylpyridine).

[0065] In particular embodiments of the application, the core polymer comprises a linear or branched polyamine as the structural core of the sorbent molecule. A polyamine is a compound comprising two or more amino groups. In embodiments of the application, the sorbent molecule comprises a polyamine core having a molecular weight greater than 500 Daltons (Da). In particular embodiments of the application, the polyamine core consists of a polymer having a weight average molecular weight greater than 1 K Da, 2 k Da, 3 kDa, 5 kDa, 8 kDa, 10 kDa, 12 kDa, 15 kDa, 20 kDa, 25 kDa, 30 kDa, 40 kDa, 50 kDa, 75 kDa, 100 kDa, 250 kDa, 500 kDa, 750 kDa, 1000 kDa, or greater. In particular embodiments, the weight average molecular weight of the polyamine core is about 25 kDa or greater. Highly branched polyamine polymers, sometimes referred to as “dendrimers,” comprise multiple primary amino groups per polymer molecule, and can be used in particular embodiments. The polyamines used in the sorbent molecules of the application comprise at least one terminal amine.

[0066] In one embodiment of the application, the polyamine core comprises polyethylenimine (PEI), also known as polyaziridine or poly(iminoethylene). Suitably, the PEI is linear PEI or branched PEI. Optionally, the core polymer comprises polyethylenimine having a weight average molecular weight of about 25 kDa.

[0067] Scheme 1 below shows the initial synthesis steps for surface functionalization of an alumina substrate for core polymer attachment according to one embodiment of the application. The first step is functionalization of the alumina surface. The initial functionalization of the alumina surface is not limited to the use of organosilanes. Those skilled in the art will appreciate that oxide surfaces readily react with compounds containing carbonyl groups (aldehydes, carboxylic acids) and phosphorous containing compounds (e.g., phosphonic acids, esters, etc.). Those skilled in the art will also appreciate that the density of hydroxyl groups on the surface of the substrate can be increased by treating the surface with compounds such as bases, acids, or mixtures of oxidizing agents (e.g., H2O2and HC1 or HNO3).

[0068] Scheme 1 : Synthesis of functionalized alumina substrate (any counterions are omitted for clarity).

[0069] Step 1 Step 2 Step 3

[0070] R1, R2, and R3may be the same or different, suitably R1, R2, and R3may be one or more selected from the group consisting of substituted or unsubstituted C1-C12alkyl, substituted or unsubstituted C2-C12alkenyl, substituted or unsubstituted C2-C12alkynyl, substituted or unsubstituted C1-C12alkoxy, substituted or unsubstituted C1-C12acyl, substituted or unsubstituted aromatic hydrocarbon group, substituted or unsubstituted aryl, heterocyclic group, and a hydrogen atom. In a particular embodiment, R1is C1-C12acyl, R2is alkyl or a hydrogen atom, and R3is absent. As the size of the polymer can vary, the above scheme shows approximate values defined in square brackets. It will also be appreciated that while the above scheme describes a branched polymer, it also applies to linear polymers, such as linear PEI.

[0071] Step 1 : Organosilanes are suitable reagents for functionalizing alumina surfaces because they are structurally diverse, readily available in large quantities, and known to react with a variety of oxide-containing surfaces. The functionalizing organosilane is typically chosen so that once it is attached to the inorganic oxide surface, the additional functional group serves as a point for further synthetic modification (as described in Step 2 below). The reactive functional group can be selected from, but is not limited to: chloride, bromide, iodide, protected alcohol group, ester, epoxide, acrylamide, olefin, and alkyne. It should be understood that alternative chemistries can also be employed to functionalize the alumina surface, and these are known to those skilled in the art.

[0072] In the present example, the reactive functional group is chloride, which can be subsequently displaced by a variety of nucleophiles. Nucleophiles such as amines or alcohols readily react with alkyl chlorides to form carbon-nitrogen or carbon-oxygen bonds via SN2-type nucleophilic substitution. It should be understood that organosilanes with other reactive functional groups can be chosen (see list above). In one embodiment, 2-propanol is chosen as the solvent for the reaction, but the reaction can also be carried out in other alcohol-containing solvents, or hexanes, toluene, or even water.

[0073] Step 2: The silane-functionalized alumina substrate is suitably reacted with a core polymer (as defined in square brackets in the scheme above, for example) that contains one or more functional groups that can be further derivatized by the addition of at least one adsorbent group. Such core polymers can be selected from the following groups: poly(allylamine), poly(methyl methacrylate), poly(vinyl alcohol), poly(vinylamine), poly(vinyl chloride), poly(vinyl imine), poly(2-vinylpyridine), poly(3-vinylpyridine), and poly(4-vinylpyridine). Core polymers containing nitrogen atoms are particularly suitable because these polymers can be further derivatized once attached to the silane-functionalized alumina material. Suitable weight average molecular weights for the polymers range from: PEI (0.8 kDa to 1000 kDa); PVP (about 60 kDa to 160 kDa); and PVC (about 48 kDa).

[0074] Step 3: The alumina intermediate containing the attached core polymer can be further derivatized to introduce adsorbent groups, thereby increasing the specificity of binding to particular target species present in a fluid feed stream. If the core polymer selected in Step 2 is a polyamine, the polyamine-alumina intermediate can be reacted with compounds such as acyl chlorides (including but not limited to alkylacyl chlorides) to install amide functional groups, or by reaction with organic halides, hydrocarbyl groups of various lengths or sizes (e.g. alkyl, alkenyl, alkynyl, cycloalkyl, or aryl) can be introduced. Amine nitrogen atoms are versatile because they can be derivatized by a variety of reagents. Thus, it is possible to derivatize the amine nitrogen center by reaction with other reagents, such as: reductive amination with an aldehyde to install an alkyl chain; condensation with a carboxylic acid to produce an amide; Michael addition with a compound containing a double bond; addition to an alkyne or alkene; amidation with an ester (or other carbonyl-containing compound); reaction with a strained cyclic system (e.g. an epoxide, aziridine); or reaction with an organic halide (e.g. alkyl, benzyl, aryl). Thus, different adsorbent groups can be added as needed for the particular target species to be removed from the fluid.

[0075] One considerable advantage is that the adsorbent compositions of the present invention have unique adsorption properties that can be tailored according to the specific requirements of the adsorbent material. Thus, one advantage of the present invention is that by changing the chemical nature of the adsorbent molecule, the adsorbent material can be easily optimized for a particular species and / or contaminant in a fluid stream. For example, by the combined selection of alumina properties (e.g. pore size), core polymer, and / or adsorbent group, the adsorption properties can be optimized for a particular target species. In this way, the physical properties of the support material work in concert with the chemical nature of the adsorbent molecule to provide unexpected properties in terms of the specific adsorptivity of interest. These are most evident in the embodiments of the present invention described in detail below, where certain compositions exhibit improved performance in the removal of a range of PFAS from water compared to industry standard activated carbon compositions. According to particular embodiments of the present invention, the primary target for treatment in water supplies is polyfluoro- or perfluoro-surfactants, such as PFOA, PFOS, PFPeA, PFHA, PFHS, PFBA, PFBS, 6:2 FTSA, and HFPO-DA.

[0076] It is also envisaged that runoff, wastewater, or other water supplies can be treated to remove other target species, contaminants, or valuable species (including precious or rare earth metals, such as present in wastewater from mining, purification, or manufacturing processes), or other fluids such as biological media, organic solvents, and oils, or to remove impurities in liquid product streams. Furthermore, the adsorbent materials according to the present invention can be used as adsorbents to remove target species from gas feed streams, such as carbon dioxide in direct air capture (DAC) devices.

[0077] Unlike other sorbents used in this way for organic contaminants, the sorbent material can be effectively regenerated in situ by a solvent washing step. The solvent washing can comprise an aqueous salt wash, an acid wash, a basic wash, or a combination, such as a salt and acid wash. The regeneration solution can also or alternatively comprise a non-aqueous polar solvent, such as acetone or an alcohol, for example ethanol, methanol or isopropanol can be used. Suitably, when an aqueous wash is used, the wash can comprise a liquid having a pH greater than 9 or a pH less than 5. Optionally, the wash solution comprises an aqueous solution of ammonium hydroxide, ammonium chloride, ammonium sulfate, potassium hydroxide, sodium bicarbonate or sodium hydroxide. In some embodiments, the pH of the wash liquid is greater than 8, suitably greater than 9 or greater than 10. When an acid wash is used, it is suitably selected from inorganic acids, including hydrochloric acid, sulfuric acid, nitric acid, phosphoric acid or organic acids suitably selected from acetic acid, hexanoic acid, oxalic acid or citric acid. The salt is suitably selected from sodium, potassium or magnesium salts having chloride, sulfate or phosphate counterions. The possibility of regeneration is particularly advantageous as it allows the removal of the target species for recycling, recovery or safe disposal, and allows the reuse of the sorbent material. In this way, the proposed method for removing target species further reduces costs and reduces the generation of waste in the form of spent sorbent material.

[0078] The regeneration process suitably involves removing the sorbent material from the fluid stream and contacting it with the wash solvent. In alternative embodiments, the regeneration process involves replacing the fluid stream with a solvent wash for a period of time to effect regeneration.

[0079] Without wishing to be bound by theory, the adsorption of the target species to the composition described herein appears to be primarily the result of non-covalent interactions (such as electrostatic interactions) of the polyamine core in combination with the hydrophobic-hydrophobic interactions of the covalently attached hydrophobic groups. In the regeneration solvent wash step, the interaction with the polar groups (such as anions) in the wash liquid replaces the electrostatic interactions with the anionic target species, thereby releasing the target species in the wash liquid. Increasing or decreasing the pH changes the protonation state of the polyamine core, which can further reduce the electrostatic binding interactions with the adsorbed target compounds. The presence of other ions (such as ammonium) can increase the solubility of the adsorbed target compounds, further increasing their removal in the regeneration solvent wash step.

[0080] Another significant advantage of the present system is its low cost and ease of production. The use of a relatively low cost core polymer and a very low cost support material (e.g. alumina) to produce the adsorbent material in the form of particles or other forms allows for the material to be produced on a large scale (approximately 1000 kg per batch) and efficiently, allowing for deployment in large volume wastewater applications (millions of litres / day flow). Furthermore, the reactions involved in linking the adsorbent substrate to the target substance adsorbent molecule can be carried out on a large scale and economically, typically at relatively low temperatures (e.g. below 100°C) and atmospheric pressure.

[0081] The present application is further illustrated by the following non-limiting examples.

[0082] Example

[0083] Physical properties of the alumina support material

[0084] The physical properties of the untreated alumina substrate used for functionalisation procured and selected are shown in Table 1 below.

[0085] Table 1 Properties of the support material

[0086]

[0087]

[0088]

[0089]

[0090] [1] BJH and BET analysis was performed using a Micromeritics Gemini VII surface area analyser at liquid nitrogen temperature using nitrogen as the adsorbate. Prior to analysis, all samples were degassed at 60 °C under vacuum until a constant weight was achieved.

[0091] [2] Powder XRD measurements were performed using a Bruker D8 ECO XRD machine using copper X-rays and a solid state Si detector at 40 kV 25 mA.

[0092] [3] Particle size distribution was measured using laser diffraction on a Malvern Mastersizer 3000.

[0093] The BET method is applicable to Type II (dispersed, nonporous or macroporous solids) and Type IV (mesoporous solids with pore diameters between 2 nm and 50 nm) adsorption isotherms. BET and full adsorption isotherms are typically used to measure surface area, total pore volume, mesopore volume, area and distribution, and micropore distribution. Pores with a width exceeding about 50 nm (0.05 pm) are referred to as macropores; pores with a width between 2 nm and 50 nm are referred to as mesopores; and pores with a width not exceeding about 2 nm are referred to as micropores.

[0094] For materials with larger pore sizes, mercury porosimetry is recommended, which covers a range of approximately 3 nm to 600 pm (mesoporous to macroporous materials), with larger pore sizes depending on the nature of the sample. The results provide porosity, pore size, pore area, and pore volume. Another method to determine the total pore volume of a sample is to use mercury pycnometry (bulk density measurement) in combination with helium pycnometry (absolute density measurement).

[0095] Materials with a high proportion of mesopores (2 nm to 50 nm) are desired. The pore size should be of sufficient size to accommodate the adsorbent molecules used in the functionalization derivatization step and remain large enough after functionalization to capture target contaminant molecules (e.g., PFAS molecules) at high capacity.

[0096] The categories of large alumina particles can be roughly divided into those with 155 m 2 / g to 330 m 2 / g surface area, 0.22 cm 3 / g to 0.41 cm 3 / g pore volume, and 3 nm to 5 nm average pore size. The smaller alumina particles have a surface area of 136 m 2 / g to 280 m 2 / g, a larger pore volume of 0.47 cm 3 / g to 0.76 cm 3 / g, and a larger average pore size of 7 nm to 20 nm.

[0097] Example 1 - Functionalization and derivatization of large particulate alumina particles

[0098] Three categories of alumina starting materials were exemplified using Steps 1 to 3 (see Scheme 1 above):

[0099] (1) Acyl-derivatized PEI-substituted alumina

[0100] (2) Quaternized acyl PEI-substituted alumina

[0101] (3) Alkyl-substituted PEI-substituted alumina

[0102]

[0103] Category 1 tests (large particulate alumina):

[0104] Figure 1 Batch test data for one category of materials are shown. The characteristic data for the alumina substrates are shown in Table 1 (above), and the characteristic data for the functionalized materials are shown in Table 2 (below). As Figure 1As shown, the untreated alumina material has moderate PFAS removal performance for PFOS, which can be related to the mesoporous structure of the untreated alumina. Figure 1 The batch test data below show that functionalization of the alumina substrate improves PFAS removal performance compared to the untreated, non-functionalized alumina substrate and intermediate material (PEI-alumina) in Example 1. Figure 1 The material in Example 1 was prepared using 3-chloropropyltrimethoxysilane activated alumina (Step 1, Scheme 1), 25 kDa PEI (Step 2, Scheme 1) and functionalized with octanoyl chloride (Step 3, Scheme 1).

[0105] In Material Series 1, different alumina substrates (Alumina 1, Alumina 4 and Alumina 6) were tested to demonstrate that the chemistry works for several different alumina substrates. The batch test data is shown in Figure 2 Table 2 (below). The characteristics of the non-functionalized alumina substrates are shown in Table 1 (above) and the characteristics of the products are shown in Table 2 (below).

[0106] It was observed that the PFAS removal performance of the alumina core followed the trend: 10-61b (Alumina 1) > 10-63b (Alumina 4) » 10-62b (Alumina 6). The PFAS removal performance of 10-62b (Alumina 6 core) was poorer relative to 10-61b (Alumina 1 core) and 10-63b (Alumina 4 core) Figure 2 ), which was believed to be related to the lower level of functionalization - determined from C and N elemental analysis of the material based on Alumina 6 (the portion that is able to capture PFAS) (see Table 2).

[0107] While the three large particle size alumina raw materials had similar specific surface areas (316 m 2 / g to 331 m 2 / g), the pore volume and average BJH pore diameter followed the trend: Alumina 1 (0.40 cm 3 / g, 5 nm) > Alumina 4 (0.38 cm 3 / g, 5 nm) » Alumina 6 (0.22 cm 3 / g, 3 nm). The optimized performance was a result of the complementary match of the polymer and functionalization to the pore size and pore volume of the alumina substrate.

[0108] The effect of the molecular weight of the PEI core polymer (Step 2, Scheme 1) was investigated on two main alumina raw materials (Alumina 1 and Alumina 4) in Example 1. The batch test data is shown in Figure 3The data show that PFAS removal performance on each of the alumina substrates tested improves with increasing PEI molecular weight (up to a weight average molecular weight of 25 kDa), with more pronounced effects for short chain PFAS targets. In addition, the performance of Alumina 1 is superior to that of Alumina 4.

[0109] Class 2 and Class 3 tests (large particulate aluminas):

[0110] The second and third classes of materials prepared utilize the first class of material and introduce a further level of functionalization through alkylation and quaternization reactions of the sorbent molecules. The motivation for this chemical reaction is twofold: (1) to introduce more alkyl chains, (2) to introduce quaternary nitrogen centers in the core polymer. Both of these additional steps result in a significant and unexpected improvement in PFAS removal compared to the control material functionalized with only octanoyl side chains. Batch test data is shown in Figure 4 The data show that PFAS removal performance on each of the alumina substrates tested improves with increasing PEI molecular weight (up to a weight average molecular weight of 25 kDa), with more pronounced effects for short chain PFAS targets. In addition, the performance of Alumina 1 is superior to that of Alumina 4.

[0111] The results show that further improvements in short chain PFAS removal can be seen by reacting 10-61b (octanoyl substituted PEI-Alumina 1) with methyl iodide or butyl iodide to produce 10-64b and 10-65b, respectively. To demonstrate the scalability of these results, 21-04b (same chemistry as 10-61b) was prepared at a scale of 1.2 kg and was also reacted with methyl iodide (21-05b) or butyl iodide (21-06b). Thus, alkylation of the octanoyl substituted PEI-alumina material (10-61b) results in improved short chain PFAS removal. In addition, reaction of the octanoyl substituted PEI-alumina material (10-61b) with butyl iodide (10-65b) results in improved short chain PFAS removal compared to reaction of the acyl substituted PEI-alumina material with methyl iodide (10-64b).

[0112] Based on the results from the Class 2 materials, a key hypothesis is whether acylation (through reaction of PEI-alumina with octanoyl chloride) can effectively remove PFAS. It has been demonstrated that simple alkylation using an excess of alkyl halide can generate materials with superior overall PFAS removal performance. The effect of alkylating agent chain length was first investigated in a C4-C10 series of butyl, hexyl, and decyl groups from studies on Alumina 1. Batch test data is shown in Figure 5 and characteristic data is shown in Table 2 (below). The effect of alkylating agent chain length was also investigated on Alumina 4. Batch test data is shown in Figure 6 and is compared to the data for Alumina 1 (as applicable). Characteristic data is shown in Table 2 below.

[0113] The results of these experiments show that the best performing material of the two classes (alumina 1 and alumina 4) based on a PEI core indicates that the hexyl chain length is optimal for the removal of the specific PFAS tested. The results also show the influence of chain length and the degree of functionalization that can be achieved on the porous substrate, and their influence on PFAS removal performance, is highly dependent on the porosity (pore volume and pore size) of the starting substrate. It is understood that the selection of the optimal adsorbent group can be different for other PFAS or target substances tested. Nonetheless, the results show the basic principle of the synergy of substrate material selection and adsorbent molecule functionalization in improving target substance removal.

[0114] Figure 7 The PFAS removal performance of octanoyl PEI-substituted alumina 4 (10-73b) was compared to octyl PEI-substituted alumina 4 (24-07b). Both chemistries have C8 side chains installed on the PEI core polymer backbone. The characteristic data is shown in Table 2 below. Figure 7 The batch test data in Table 2 shows that the 24-07b material has superior broad spectrum short and long chain PFAS removal performance compared to 10-73b.

[0115] Table 2 Characteristics of representative materials of classes 1-3

[0116]

[0117]

[0118] Example 2 - Functionalization and derivatization of small particle size alumina particles

[0119] Three commercial alumina feedstocks (d50 < 100 microns) were selected for functionalization (see Table 1 above and Table 3 below). A comparative example of a large particle size alumina particle (alumina 4) is presented.

[0120] Table 3 Smaller particle size alumina materials used for functionalization

[0121]

[0122] *Properties of the larger alumina 4 grade feedstock included for comparison

[0123] The 3-step process developed for large alumina (Scheme 1) for class 3 (see above) was used to surface functionalize the untreated alumina feedstocks, and is summarized as Scheme 2 below. The materials exemplified in this work (28-4a, 28-4b, 28-4c) used iodohexane in the final functionalization step (Scheme 2, step 3). However, it is understood that the scope of functionalization is not limited to iodohexane, but other haloalkanes (C4-C12) can be used, such as the examples of alumina 1 and alumina 4 (above).

[0124] Scheme 2: Synthesis of functionalized alumina substrates (any counterions are omitted for clarity).

[0125] Step 1

[0126] Step 2

[0127] Step 3

[0128] Table 4 below shows the analytical data for the three fully functionalized alumina materials (28-4a, 28-4b, 28-4c). For reference, the larger particle size alumina 24-06b, which performed best (by removing PFAS), is included, which has the same hexyl surface functionalization chemistry.

[0129] Table 4 Analytical data for functionalized small alumina particles

[0130]

[0131] * 24-06b, properties of the larger functionalized alumina 4 feedstock, are included for comparison.

[0132] It is noted that:

[0133] • The smaller alumina feedstocks (Table 3) have significantly larger average pore diameters (7 nm to 20 nm vs. 5 nm) and pore volumes (0.47 cm 3 / g to 0.76 cm 3 / g vs. 0.38 cm 3 / g) than the larger alumina 4 substrate.

[0134] • The degree of functionalization on the surface of the smaller functionalized alumina adsorbent media is significantly higher than the larger functionalized alumina adsorbent media as judged by carbon and nitrogen content (see Table 4) and confirmed by FTIR analysis (not shown).

[0135] • The carbon to nitrogen ratio of the smaller alumina adsorbent media is greater than the larger alumina adsorbent media

[0136] (approximately 4-4.5: 1 vs. 3: 1). This suggests that the polyethyleneimine (PEI) backbone nitrogen atoms in the smaller adsorbent media can have a higher degree of alkylation.

[0137] • The percent reduction in pore volume of the smaller alumina adsorbent media is greater than the larger particulate media (31% to 41% vs. 26%) after functionalization of the alumina substrate (Table 4).

[0138] Elemental analysis and porosity analysis confirmed that the smaller alumina adsorbent media had a greater degree of surface functionalization than the larger 24-06b material. The level of functionalization was positively correlated with the pore diameter, in fact inversely related to the BET determined surface area. This was unexpected as the surface area is generally considered to be the primary contributing factor, however, the current data shows that the pore diameter of the base material is an important feature that contributes to the performance of the adsorbent compositions of the invention. Mesopore dominated base materials, in particular those having an average pore diameter greater than about 4 nm, are particularly suitable for the removal of PFAS from water.

[0139] As previously described, the initial PFAS removal performance of the new smaller adsorbent materials was investigated in 24h batch tests. Figure 8 Batch test data for the new compositions (28-4a, 28-4b, 28-4c) are shown compared to the reference compositions of 24-06b as well as GAC and conventional ion exchange resins (Amberlite and Resinex).

[0140] Isotherm data

[0141] Isotherm batch tests were performed to understand the removal capacity of the new adsorbent media (28-4a, 28-4b, 28-4c). Figure 9 Representative plots showing the benchmark testing of powdered activated carbon (PAC) against six different PFAS species are shown.

[0142] Table 5 shows comparative performance of the new materials against small particle activated carbon materials in terms of PFAS removal capacity.

[0143] Table 5 small alumina versus PAC and GAC PFAS removal performance comparison

[0144]

[0145]

[0146] The best performing small alumina material in these tests was 28-4b, which can be related to the high level of functionalization of the particle surface.

[0147] Materials and methods

[0148] Silanization reaction (step 1, schemes 1 and 2). 50g alumina was added to 2-propanol (150mL) and 3-chloropropyltrimethoxysilane (1.4mL). The suspension was stirred at 65°C for 24h using an overhead stirrer at a speed of about 150rpm. The material was then washed with 2-propanol (3x 150mL) and then vacuum dried at 80°C for 3h. The material was then further vacuum dried at 40°C.

[0149] The PEI reaction (step 2, schemes 1 and 2) was performed on a 30 g scale. 30 g of the dried silylated product was added to a solution consisting of PEI (e.g. MW = 25 kDa, 9 g) and deionized water (150 mL). This was heated to 70 °C for 24 h while stirring at a speed of approximately 150 rpm. 10 M NaOH was added at 30 min, 1.5 h, 2.5 h, 3.5 h and 4.5 h (total 5 x 90 μL portions). The material was then washed using water (2 x 150 mL), the pH was adjusted to 4 using HCI, deionized water (150 mL), 0.1 M aqueous potassium carbonate (2 x 150 mL) and deionized water (2 x 150 mL) and then dried under vacuum at 40 °C.

[0150] The acylation reaction (step 3, scheme 1) was performed on a 40 g to 80 g scale. The aminated product (40 g) was carefully added to octanoyl chloride (20.4 mL = 3 mmol / g PEI substituted alumina) in dry acetonitrile (120 mL). The suspension was stirred at 25 °C for 5 h. The material was carefully washed with 2-propanol (4 x 50 mL) at 50 °C, then with deionized water (4 x 50 mL) at 50 °C and finally with 0.1 M aqueous potassium carbonate (250 mL) for 1 h. The final product was filtered, washed with water (100 mL) and dried under vacuum at 40 °C overnight.

[0151] The alkylation reaction (step 3, schemes 1 and 2) was performed on a 15 g scale. 15 g of the aminated product was added to a suspension consisting of potassium carbonate (14.9 g) and 2-propanol (45 mL). 1-iodohexane (13.28 mL) was added and the suspension was stirred at 80 °C overnight. The material was then washed using 2-propanol (4 x 50 mL), 0.1 M aqueous potassium carbonate (2 x 50 mL) and deionized water (2 x 50 mL) at 50 °C. The final product was dried under vacuum at 40 °C overnight.

[0152] To purify the functionalized material, the final product was divided into two 50 mL falcon tubes. 45 mL of deionized water was added to each tube and after shaking for 30 minutes, centrifuged. The water was then removed and the TOC, TN and aluminium content were analysed. A further 45 mL of deionized water was then added to each falcon tube and the process was repeated until a total of 8 washes had been performed.

[0153] Example 3 - Membrane-encapsulated filtration device for point-of-use (POU) applications

[0154] Following the successful testing of PFAS contaminated water in Examples 1 and 2, the membrane encapsulated adsorbent structure (MEAS) filled with functionalized alumina as described in this example was tested with additional PFAS species specified in the interim version of the updated NSF-53 industry standard protocol, which also includes GenX. The MEAS included the functionalized alumina composition between porous metal foil carrier membranes.

[0155] Table 6 lists the parameters required for a typical POU device. These parameters indicate the requirements for a typical under-sink cartridge, such as a home use.

[0156] Table 6 Typical design specifications for a POU device

[0157] Design specifications Parameter value or range Operating pressure range 1.4 bar to 6.8 bar Initial treatment flow rate 3 L / min Full contact time <10 seconds Provisional product life target Up to 12 months Provisional volume of water to be treated 5500L

[0158] Method

[0159] Operating parameters

[0160] Two MEAS test units were tested simultaneously, with alumina powder bed depths of 7 mm and 13 mm. The alumina powder loaded into these units followed the same functionalization steps, but came from different production batches. The 7 mm unit contained 20.1 g of powder, and the 13 mm unit contained 35.4 g of powder.

[0161] A peristaltic pump was used to feed water at a flow rate of 150 mL / min, and the flow rate was monitored periodically throughout the experiment and maintained between 140 mL / min and 160 mL / min. The system was run for approximately 8 h per day, and was allowed to stagnate overnight.

[0162] Table 7 Operating parameters for MEAS PFAS flow-through testing

[0163] Parameter Detection unit 1 Detection unit 2 Bed depth (mm) 7 13 Powder mass (g) 20.1 35.4 Membrane area (cm 2 )]]> 36.8 36.8 Flow rate (mL / min) 150 150 Flow (L / h m 2 )]]> 2440 2440 Contact time (s) 10 19

[0164] A POU device processing 2 L / min at an electrical current flux of 2440 L / hm 2 would require 500 cm 2 of membrane surface area. The required scaling factor is 13.3, keeping the constant operating parameters tested in the test equipment. This scaling factor can be used to convert the water volume processed by test units 1 and 2 to the equivalent volume of water processed by a full-scale device. Thus, processing 200 L of water through the test equipment is equivalent to processing 2670 L of water in a full-scale device.

[0165] Water matrix and sampling

[0166] The PFAS species used and their concentrations are listed in Table 8 below.

[0167] Table 8 Concentrations of PFAS species in the feed solution for flow-through testing

[0168] PFAS species GenX PFHpA PFOA PFNA PFHxS PFOS Concentration (ppt) 900 40 500 50 300 1000 Regulatory limit (ppt) Not specified (135) 20 20 6 30 20

[0169] The current NSF-53 protocol states that the test water should come from a public water supply and should always maintain the characteristics specified in Table 9.

[0170] Table 9. Test water specifications during NSF-53 certification

[0171] pH 7.5±0.5 Temperature (°C) 20±2.5 Total dissolved solids ‘TDS’ (mg / L) 200 to 500 Total organic carbon ‘TOC’ (mg / L) >1.0 Turbidity (NTU) <1

[0172] For preliminary testing purposes, spiked deionized water was used in place of London tap water. The ions added to the water are shown in Table 10 below. This water meets the NSF pH and TOC standards. The TOC was added as chlorinated tannin acid as specified by the NSF protocol. The other ions were chosen to represent median concentrations of many common water matrix components. It should be noted that the current NSF-53 does not specify an upper limit for GenX, however, for the purposes of the current testing, a recommended limit of 135 ppt has been assumed.

[0173] Table 10. Concentrations of water matrix components

[0174] Ion Concentration Concentration unit Na + ]] 103 ppm Ca 2+ ]]> 50 ppm Mg 2+ ]]> 10 ppm CO3 2- ]]> 180 ppm SO4 2- ]]> 70 ppm NO3- 1 ppm Cl - ]]> 118 ppm TOC 1.4 ppm Adjusted pH 7.5 -

[0175] Sampling regime

[0176] Samples were taken from a port on the effluent line every 2h during the 70h of the experiment. More frequent sampling was performed during the first 2h of the experiment.

[0177] Samples were sent for PFAS quantification by LC-MS every week. The detection limit for this technique is 0.5 ppt to 2 ppt depending on the PFAS species. The quantification limit is 2 ppt to 7 ppt.

[0178] Results

[0179] Flow control and pressure drop

[0180] A plot of pressure drop versus flow rate was generated by varying the pump speed before starting the PFAS flow tests. This data is shown in Figure Figure 10 The results are clearly linear, despite one test unit having almost twice the depth of alumina powder, the pressure drops were very similar.

[0181] The inlet pressures at the start of the PFAS tests were 0.2 bar and 0.325 bar for the 7mm and 13mm test units respectively.

[0182] The pressure steadily increased throughout the test. After 150 L of water had passed through the system, the pressure in both systems had risen to approximately 0.5 bar. After 300 L of water had passed through, the inlet pressure in both systems was 0.9 bar. At the end of the experiment, the inlet pressure in both systems was in excess of 1.2 bar.

[0183] Throughout the experiment, the inlet pressure increased significantly and the pressure in both test units was approximately equal. There was a noticeable brown deposit on the feed side of the membrane.

[0184] Although the increase in inlet pressure cannot be attributed entirely to the presence of tannins from this experiment, the colour of the deposit suggests that tannins can be a cause.

[0185] PFAS removal performance of the 7mm test unit

[0186] The PFOA concentration in the effluent of the 7mm unit only began to rise sharply from 5 ppt to 15 ppt after 396 L of water had been treated over 44 h - see Figure 11 When extrapolated to a full-scale POU device, this is equivalent to 5284 L of water. It is estimated that the average European uses approximately 20 L of water in the kitchen every day (Richter, C. P. and Stamminger, R. (2012). Water consumption in the kitchen - A case study in four European countries. Water Resource Management). Therefore, it can be expected that a scaled-up version of the 7mm unit could be run reliably as a POU device for more than six months, which is well in line with current industry standards.

[0187] The GenX concentration in the effluent of the 7mm unit was first detected at 6.6 ppt after 306 L of water had been treated (34 h, POU device treats 4080 L) - see Figure 12 Thereafter, the GenX concentration in the effluent of the 7mm unit increased gradually, reaching a peak of 62 ppt after 464 L of water had been treated (51.5 h, POU device treats 6184 L). The final GenX concentration recorded was 50 ppt. The GenX concentration never exceeded half of the assumed limit of 135 ppt.

[0188] The PFOS concentration in the effluent of the 7mm unit rose sharply to 14 ppt after 414 L of water had been treated (POU device, 46 h). Thereafter, the concentration remained between 12 ppt and 17 ppt for the remainder of the experiment, but never exceeded the limit of 20 ppt.

[0189] PFHxS was first measured above the limit of quantification after 396 L of water had been treated (44 h, 5284 L treated in the POU installation). After this limit was exceeded, the concentration of PFHS increased from 4 ppt to 11 ppt. This is below the limit of 30 ppt for PFHS.

[0190] PFHpA and PFNA were first detected after 396 L of water had been treated (POU installation, 44 h). PFHpA remained below its limit of quantification of 5 ppt, which is safely below its regulatory limit of 20 ppt. PFNA remained below its limit of quantification of 6 ppt, which is also its regulatory limit. Although the exact concentration of PFNA could not be calculated throughout the run, the largest peak area observed for PFNA throughout the experiment was approximately half the peak area of the 6 ppt standard.

[0191] PFAS removal performance of the 13 mm test unit

[0192] The concentration of PFOA in the effluent of the 13 mm test unit remained well below the regulatory limit of 20 ppt throughout the experiment. The amount of PFOA in the effluent never exceeded the limit of quantification (LOQ) of 4 ppt during the experiment. GenX was first detected in the effluent in the 13 mm unit above the limit of detection (LOD) but below the LOQ after 396 L of water had been treated (44 h, 5284 L treated in the POU installation). The GenX concentration rose from 8 ppt after 464 L had been treated (51.5 h) to 27 ppt at the end of the experiment (69 h).

[0193] PFHpA and PFNA were below the limit of detection throughout the experiment except for a few single points early on. PFHxS was detected around the limit of detection from 38 h (342 L treated). The amount of PFHxS never exceeded the LOQ of 2 ppt for the remainder of the experiment. PFOS was observed at the limit of detection from 61.5 h (554 L treated). When extrapolated to the full-scale POU sink installation, the volume of water treated is equivalent to more than a year of operation.

[0194] The breakthrough curves for PFOA and GenX are shown in Figures 13 (a-b).

[0195] PFAS capacity

[0196] The breakthrough results for the 7 mm unit can be used to estimate the capacity of the alumina adsorbent powder. By assessing the inlet and outlet PFAS concentrations for each species and multiplying the difference by the volume of water treated before breakthrough (554 L), the total mass of PFAS captured for each PFAS species is calculated. This is then divided by the mass of powder in the unit (20.1 g) to determine the capacity. The results are listed in Table 11 below.

[0197] Table 11 Capacity of functionalized alumina for PFAS species based on breakthrough of PFOA in 7 mm test cell

[0198]

[0199]

[0200] Conclusion

[0201] The PFAS adsorption test included additional PFAS contaminants (GenX, PFHpA, PFNA, and PFHxS) in accordance with the NSF-53 guidelines and was performed on Puraffinity media packaged in MEAS equipment. The test was performed in parallel using two cells with different bed depths of 7 mm and 13 mm. The contact times for these two cells were 10 s and 19 s, respectively, and the loading of the adsorbent powdered alumina media was 20.1 g and 35.4 g, respectively, with a flow rate of 150 mL / min for each cell.

[0202] In the 7 mm test cell, the effluent concentration of most PFAS changed from the non-detectable limit to the detectable limit after 44 h of sampling. The most significant concentration change was for PFOA and PFOS, which both increased from 5 ppt to 16 ppt, but did not exceed the regulatory limit of 20 ppt. No PFAS limits were exceeded in the second 13 mm bed depth cell. The PFOA concentration in the effluent of the 13 mm cell remained well below the regulatory limit of 20 ppt during the experiment.

[0203] These experiments demonstrate the feasibility of this technology on a laboratory scale. This utility also applies to a broad range of PFAS.

[0204] Example 4 - Functionalization of alumina 11 with hexyl-PEI using iodohexane and bromohexane

[0205] The alkylation reaction (step 3, schemes 1 and 2) was performed using bromohexane on a 15 g scale. The 15 g of aminated product was added to a suspension consisting of potassium carbonate (12.4 g) and 2-propanol (30 mL). 1 -Bromohexane (25.35 mL) was added and the suspension was stirred at 80 °C overnight. The material was then washed using 2-propanol (4 x 50 mL) and deionized water (10 x 50 mL) at 50 °C. The final product was dried under vacuum at 40 °C overnight. Table 12 provides the product elemental analysis results compared to the functionalization of alumina with iodohexane (see Example 2).

[0206] Table 12 Data from elemental analysis of hexyl-PEI-functionalized alumina 11 using iodohexane and bromohexane

[0207]

[0208]

[0209] As mentioned earlier, the PFAS removal performance of the alumina 11 adsorbent material was studied in a 24-hour batch test. Figure 14 Batch test data for the novel composition are shown.

[0210] Example 5 – Porosity Analysis of Particulate Alumina

[0211] Although the BET data for untreated alumina 1, 4, and 11 appear very similar – implying similar micropore and mesopore levels (see Table 1) – Hg porosity data reveal a key difference in macroporosity among the three substrates. Alumina 1 and 11 exhibit macroporosity (pore diameter approximately 3 μm), while alumina 4 does not. Surprisingly, this porosity, in addition to mesopores with diameters ranging from approximately 3 nm to 8 nm, can enhance the functional level of the alumina surface, thereby improving PFAS removal performance. Figure 15(a) shows the Hg porosity data for untreated alumina 1, 4, and 11.

[0212] While the presence of large pores is beneficial for improving performance, it can also impair the mechanical stability of the material. These two factors must be effectively balanced according to the desired end use. The high porosity level in alumina 4 makes the material physically too fragile for certain types of water treatment applications. However, alumina 11 has a lower porosity compared to alumina 4, resulting in a material that is sufficiently stable for the intended application and exhibits improved PFAS removal performance compared to purely mesoporous materials. Performance batch data for hexyl-PEI functionalized alumina 1, 4, and 11 are shown in Figure 15(b). Elemental analysis data for hexyl-PEI functionalized alumina 1, 4, and 11 are shown in Table 13.

[0213] Table 13 Elemental analysis of functionalized alumina 1, 4 and 11

[0214] Material Si (wt%) C (wt%) N (wt%) Hexyl-PEI-alumina 4 0.33 5.93 1.70 Hexyl-PEI-alumina 1 0.36 4.34 1.41 Hexyl-PEI-alumina 11 0.39 6.70 1.33

[0215] Example 6 - Porosity Analysis of Powdered Alumina

[0216] Similar analyses to those in Example 5 were repeated on powdered alumina materials based on alumina 8 or closely related alumina (referred to as alumina AC), revealing pore sizes (see...). Figure 16 ) and PFAS performance (see Figure 17 A similar trend is observed. However, compared to the 3nm to 8nm pore size range of particulate alumina, the pore size range has shifted to larger mesopores of 18nm to 40nm. Unlike particulate media, performance does not initially improve but decreases with increasing mesopore size (see [link to article]). Figure 17). Elemental analysis data also followed roughly the same trend, with a decrease in %C content observed as the pore size increased.

[0217] Table 14 Elemental analysis data for various functionalized powdered alumina materials

[0218]

[0219] Example 7 - Testing of Spherical Alumina Particles

[0220] Another common form of alumina (in addition to powders and particles) is spherical particles, commonly used as supports for catalysts, dryers for gases and liquids, and hydrocarbon adsorbents. This form of adsorbent can have advantages in large scale industrial applications, as its larger particle size (mm range) will result in lower pressure drops as fluids pass through packed beds composed of the particles. A drawback of these substrates is mechanical strength, which presents challenges for both the application and synthesis of functionalized spheres. Four different spherical alumina substrates, referred to as S-1 through S-4 (physical properties seen in Table 15 below), were tested from two different manufacturer sources.

[0221] Table 15 Physical properties of spherical alumina particles tested

[0222]

[0223]

[0224] S-1 and S-2 were functionalized using standard procedures, with hexyl iodide for alkylation (step 3, schemes 1 and 2), while S-3 and S-4 spheres were functionalized using the standard hexyl bromide procedure described previously.

[0225] Compared to the particles, the functionalization of the alumina spheres was found to be lower, without wishing to be bound by theory, this was thought to be due to attrition during the synthesis process. However, the S-4 spheres exhibited good performance metrics, which was thought to be due to the bimodal porosity distribution (with peaks for mesoporsity and macroporsity) and the overall higher porosity of these spheres.

[0226] Table 16 Elemental analysis data for hexyl-PEI functionalized alumina spheres

[0227] Material Si (wt%) C (wt%) N (wt%) S-1 0.26 3.25 0.82 S-2 0.12 1.27 0.41 S-3 0.11 1.66 0.23 S-4 0.29 8.29 1.30

[0228] Figure 18 PFAS performance batch data for hexyl-PEI functionalized alumina spheres are shown in Table 16.

[0229] Example 8 - Variation of Molecular Weight of PEI Polymers

[0230] Previously acylated products were prepared and tested using 0.8 kDa, 1.8 kDa and 25 kDa PEI functionalised alumina 1 and 4, see Example 1 and Figure 3 .

[0231] In this example, hexylated-PEI products were prepared using 750 kDa PEI (same mass as normally synthesised) with alumina 11 and compared to hexylated-PEI with 25 kDa PEI. As shown in Figure 19 PFAS performance batch data for the larger 750 kDa PEI was found to be within the error range of the standard product made with 25 kDa hexyl-PEI. Comparison of elemental analysis is shown in Table 17.

[0232] Table 17 Elemental analysis data for hexyl-PEI-functionalised alumina 11 made with 25 kDa and 750 kDa PEI

[0233] Material Si wt% C wt% N wt% 25 kDa PEI 0.29 6.20 1.17 750 kDa PEI 0.25 6.59 1.16

[0234] Example 9 - Use of linear PEI and change in chain length

[0235] Similar linear PEI functionalised materials were prepared using trimethoxysilyl propyl modified polyethyleneimine (1.5 kDa to 1.8 kDa) - synthesis method as follows. Alkylation reactions were performed using standard methods with different iodoalkanes of different carbon chain lengths. The materials are compared below to branched PEI analogues.

[0236] The data shows that several of these linear PEI functionalised materials perform as well as their branched counterparts. In terms of carbon chain length, while C>6 performs equally well, butyl performs significantly worse. PFAS performance batch data is shown in Figure 20 .

[0237] Synthesis method for linear PEI-alumina

[0238] Alumina 8 (50 g) was carefully weighed into a tared Duran bottle containing isopropanol (250 mL), then made into a slurry and poured into a 1 L three necked flask in a heating block with the stirring impeller above 10 cm rotated at 150 rpm for the addition. 1.65 kDa linear PEI modified propyltrimethoxysilane (66 mL, 50% in IPA) was added and the reaction mixture stirred at 65 °C overnight. After cooling, the powder was allowed to settle and the reaction solution was removed. The crude product was then washed with IPA (3 x 250 mL) with each wash stirred for approximately 45 min and then allowed to settle for approximately 15 min. The washed material was then filtered through a fritted funnel and dried in a vacuum oven at 80 °C (30 mbar) for approximately 4 h. Elemental analysis is shown in Table 18 below.

[0239] Table 18 Elemental analysis data for powdered adsorbents with different alkyl chain lengths

[0240]

[0241]

[0242] While particular embodiments of the present application have been disclosed in detail herein, this has been done by way of example for purposes of illustrative discussion. The embodiments described above are not intended to be limited to the scope of the appended claims. Various substitutions, alterations and modifications can be made to the application without departing from the spirit and scope of the application as defined by the claims.

Claims

1. A composition for removing polyfluoroalkyl substances and perfluoroalkyl substances (PFAS) from an aqueous solution, the composition comprising: a particulate support material comprising a bimodal alumina of pores in the mesopore range and pores in the macropore range; (i) wherein a majority of the pores in the mesopore range have a diameter greater than 2 nm and less than 50 nm; and (ii) wherein the bimodal alumina has a BET pore volume in the mesopore range of not less than 0.20 cm 3 / g; and adsorbent molecules comprising a linear or branched core polymer selected from one or more of: poly(ethyleneimine), poly(allylamine), poly(methyl methacrylate), poly( vinyl alcohol), poly( vinyl amine), poly(vinyl chloride), poly(2-vinylpyridine), poly(3-vinylpyridine), or poly(4-vinylpyridine); wherein the linear or branched core polymer is covalently attached to the particulate support material; and wherein the adsorbent molecules further contain one or more covalently attached adsorbent groups.

2. The composition of claim 1, wherein the alumina has an average pore diameter in the mesopore range of 2 nm to 20 nm.

3. The composition of claim 1, wherein a majority of the pores in the macropore range have an average pore diameter of 1 pm to 10 pm.

4. The composition of claim 1, wherein the core polymer comprises a linear or branched poly(ethyleneimine) having a weight average molecular weight of not less than 25 kDa.

5. The composition of claim 1, wherein the covalently attached adsorbent groups comprise one or more selected from the group consisting of: substituted or unsubstituted C1-C12 alkyl, substituted or unsubstituted C2-C12 alkenyl, substituted or unsubstituted C2-C12 alkynyl, substituted or unsubstituted C1-C12 alkoxy, substituted or unsubstituted C1-C12 acyl, substituted or unsubstituted arene, substituted or unsubstituted aryl, heterocyclic group, and hydrogen atom.

6. The composition of claim 1, wherein the core polymer contains at least one tertiary amino group, and wherein at least one tertiary amino group is converted to a quaternary nitrogen.

7. The composition of claim 1, wherein the core polymer comprises a C1-C10 alkyl substituted linear poly(ethyleneimine).

8. The composition of claim 1, wherein the core polymer comprises a C1-C6 alkyl substituted branched poly(ethyleneimine).

9. The composition of claim 1, wherein the support material is in a form selected from the group consisting of: particulate form, powder form, and spherical particulate form.

10. A method for removing a target substance from a fluid stream, the fluid comprising water, the method comprising contacting the fluid stream with the composition of claim 1, wherein the target substance comprises one or more of polyfluoroalkyl substances and perfluoroalkyl substances (PFAS).

11. The method of claim 10, wherein the PFAS is selected from perfluorinated anionic surfactant compounds including one or more selected from the group consisting of perfluorobutane sulfonate (PFBS), perfluorobutyric acid (PFBA), perfluoropentanoic acid (PFPeA), perfluorohexane sulfonate (PFHS), perfluorohexanoic acid (PFHA), perfluorooctanoic acid (PFOA), perfluorooctane sulfonate (PFOS), perfluorononanoic acid (PFNA), and perfluorodecanoic acid (PFDA), 6:2 fluorotelomer sulfonic acid (6:2 FTSA), and hexafluoropropylene oxide dimer acid (HFPO-DA).

12. The method of claim 10, wherein the linear PEI or branched PEI has a weight average molecular weight of no less than 25 kDa.

13. The method of claim 10, wherein the particulate alumina is in a form selected from the group consisting of particulate form, powder form, and spherical particulate form.

14. The method of claim 10, wherein the carrier material is contained within a packed bed.

15. The method of claim 14, wherein the carrier material is contained within a filtration unit.

16. A filter for adsorbing one or more PFAS from a contaminated water source, wherein the filter comprises the composition of claim 1.

17. The filter of claim 16, wherein the filter is configured as a point-of-use (POU) filter.

18. The filter of claim 17, wherein the filter is included within a replaceable filter cartridge.

19. The filter of claim 16, wherein the filter is configured as a point-of-entry (POE) system.

20. The filter of claim 16, wherein the composition is included within a packed bed configured to allow a minimum water flow rate of at least 2 liters / minute.

Images