Porous functionalized multiblock copolymer films, methods of making and using same, and products thereof

Abstract

Methods for forming films, films, and uses thereof. In various examples, a method comprises forming a film using a deposition solution comprising two or more multiblock copolymers, where the multiblock copolymers each comprise different corona-forming block(s) or different corona-forming block(s) and different core-forming block(s), where there is no substantial exchange of the polymer chains between the micelles formed from the multiblock copolymers. In various examples, a film comprises an isoporous top surface layer comprising two or more multiblock copolymers and an asymmetric layer, which may be an asymmetric substructure, where at least a portion the core blocks of the multiblock copolymers are isotropically or statistically distributed in, disposed in a plurality of distinct domains of, or disposed in a three-dimensional lattice in (or forming) at least a portion of the pore wall surface(s) of the isoporous surface layer. In various examples, a film is used in a filtration, separation, or catalytic application.

Description

Attorney Docket No.: 018617.01809 POROUS FUNCTIONALIZED MULTIBLOCK COPOLYMER FILMS, METHODS OF MAKING AND USING SAME, AND PRODUCTS THEREOF CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 520,308, filed August 17, 2023; the contents of the above-identified application are hereby fully incorporated herein by reference in their entirety. STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

[0002] This invention was made with government support under grant number 1707836 awarded by the National Science Foundation. The government has certain rights in the invention. BACKGROUND

[0003] Polymer membranes are of rapidly increasing societal interest, e.g., for industrial separations, virus filtration, or drinking water purification. Moving beyond size-selective filtration will require the development of fresh membrane formation strategies.

[0004] The design rules governing the self-assembly of multicomponent colloids or nanocrystals into complex architectures have been widely studied. By tuning the electrostatic interactions, shapes, sizes, and compositions of the colloidal components, a plethora of well- defined structures with distinct property profiles can be achieved. This progress is accompanied by synergies between experimental and computational work. Nanoscale visualization techniques and computational capabilities increasingly allow for in-depth characterization and predictive modelling of novel colloidal materials, respectively.

[0005] Block copolymer (BCP) self-assembly, regarded as a hallmark of soft condensed matter physics, generates a variety of ordered nanostructured morphologies (microphases), including spherical micelles (spheres), cylindrical micelles (cylinders), lamellae and co- continuous structures. Advances in synthetic polymer chemistry have expanded the accessible molecular architectures and microphases of these macromolecular building blocks. Despite the parallels between nanocolloids and BCP micelles, however, structural studies of chemically distinct multicomponent micellar systems have been sparse.

[0006] This is particularly true for non-equilibrium derived structures, whose complex formation pathways remain challenging to understand. Isoporous BCP membrane formation combining self-assembly (S) and industrially scalable non-solvent induced phase separation (NIPS), later termed SNIPS, involves such non-equilibrium processes. Despite the simplefabrication methodology, the final membrane structure is a confluence of multiple processes. After a BCP dissolved in a binary (or ternary) solvent system (referred to as the polymer dope) is doctor bladed into a thin film, selective solvent evaporation prompts BCP micelle self-assembly on the top surface and creates a concentration gradient along the film normal. This non-equilibrium structure is kinetically trapped when the film is immersed in a non- solvent (typically water) bath. Solvent-non-solvent exchange results in rapid precipitation of a BCP membrane structure with a thin (~ 100 nm) periodically ordered porous separation layer atop an asymmetric porous substructure. Such membranes display high pore density and narrow pore size distributions, paving the way for high permeability and selectivity separations, e.g. of biopharmaceuticals like antibodies and antibody-drug-conjugates, for clean drinking water, or for virus filtration, topics of rapidly increasing societal interest.

[0007] BCP SNIPS derived membranes address the permeability-selectivity trade-off, a contributing factor in membrane performance. The improved understanding, particularly of biological membranes, has unveiled an alternate design parameter to surpass the limits of size-selective filtration: fine-tuning of permeant-membrane pore wall interactions. This requires multifunctional membranes, however, with highly tuned combinations of pore chemistries, including van der Waals and electrostatic interactions, which, to date, may be approached via time-consuming post-synthesis membrane modifications.

[0008] The bottom-up self-assembly of nanoscale building blocks into macroscopic structures underlies biology, chemistry, and materials science. Promising emergent properties inspire transitions from single- to multicomponent materials. Colloidal assemblies are an attractive platform offering modular character and unlimited combination potential. Treating chemically distinct block copolymer (BCP) micelles as soft colloidal building blocks paves a path towards multicomponent assemblies with molecularly engineered property profiles. Poor atomic contrast between BCP constituents challenges structural characterization, however, thereby slowing progress.

[0009] As macroscopic properties of materials can be tailored by their structures on the nanoscale, bottom-up assembly approaches have garnered much attention. This has been pursued via myriad experimental and computational strategies involving single or multiple building blocks comprising real or ‘artificial’ atoms. Among these, high-entropy materials (metals, oxides, nitrides etc.) containing five or more atoms in near equimolar ratios have unlocked a vastly unexplored phase space and shown promise of emergent properties not exhibited by the individual components. Beyond atomic constituents, isotropic (spherical) and anisotropic single-component (macro)molecular building blocks have revealed intricatestructural assemblies (e.g., Frank-Kasper phases, hierarchical phase-in-phase morphologies). Towards multifunctional materials meeting the demands of an ever-evolving world, modular colloidal particles provide a rich playground for exploration. Self-assembly of binary (or higher order) colloidal mixtures into specific crystal structures – a confluence of their size ratio ( ^^), interaction potential (e.g., attractive / neutral / repulsive, hard / soft), and relative composition – offers a pathway towards ‘metamaterials’ with tailorable property profiles. The level of programmability in DNA-functionalized colloidal particles has even opened the door to crystal superlattices with no atomic equivalents. While uniform assemblies (e.g., colloidal crystals) have long been the desired structural outcome, heterogeneous distributions (e.g., controlled colloidal aggregation, phase separation, and stratification) find potential applications in biomimetics, photovoltaics, and functional coatings. Such structures can be accessed by departing from thermodynamic equilibrium via irreversible processes (e.g., evaporation) or the incorporation of active components.

[0010] Compared to the rich history of block copolymer (BCP) self-assembly, the treatment of chemically distinct BCP micelles as modular soft colloids is still in its infancy. Unlike hard colloids, BCP micelles comprise an outer corona and inner core – the latter essentially a nanoreactor capable of carrying cargo (e.g., for drug delivery or catalysis) – making them a platform for a generation of multicomponent functional materials with equilibrium and non-equilibrium structures. Thus far, studies on BCP micelle alloying have predominantly been theoretical. A roadblock to experimental translation is characterization of non-crystalline multicomponent nanostructures. Traditional micron sized colloidal assemblies are directly resolvable via optical microscopy, whereas metallic / semiconducting / magnetic nanoparticle superlattices containing heavy elements generate contrast under transmission electron microscopy (TEM). In contrast, BCP micelles generally have interpenetrating coronas that complicate individual resolution and consist of light elements lacking atomic contrast (e.g., C, N, O). SUMMARY OF THE DISCLOSURE

[0011] The present disclosure provides, inter alia, films (such as, for example, porous functionalized multiblock copolymer films and the like), compositions, and methods of making films (such as, for example, porous functionalized multblock copolymer films and the like). The present disclosure also provides uses of films (such as, for example, porous functionalized multiblock copolymer films and the like).

[0012] In an aspect, the present disclosure provides methods of making films. In various examples, a method of forming a film (e.g., a film comprising an isoporous region or layer and an asymmetric region or layer, where the isoporous region or layer is disposed on at least a portion of an exterior surface of the asymmetric region or layer) comprises: forming a first film (e.g., forming a first film on a substrate or the like) from a deposition solution comprising two or more multiblock copolymers, where each multiblock copolymer comprises one or more corona-forming block(s) and one or more core-forming block(s), and at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core- forming block(s)); and two or more solvents, removing at least a portion of the solvent system from the film (e.g., allowing the first film to stand for a period of time such that at least a portion of the two or more solvents is removed from the film; and contacting the second film with a phase separation solvent system, such that the film is formed (which may be disposed on a substrate). In various examples, a first multiblock copolymer is present in a plurality of first micelles and a second multiblock copolymer is present in a plurality of second micelles and there is no, no substantial, or no observable exchange of the polymer chains from the first micelles with the polymer chains of the second micelles.

[0013] In an aspect, the present disclosure provides compositions. In various examples, a composition is suitable for use in a deposition solution (e.g., a dope or the like), such as, for example, use as a deposition solution in a method for the present disclosure. In various examples, a composition comprises: two or more multiblock copolymers, where each multiblock copolymer comprises one or more corona-forming block(s) and one or more core- forming block(s), and at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s) or the like. In various examples, a composition further comprises one or more homopolymer(s), two or more solvents, one or more metal salt(s), one or more inorganic nanoparticle(s), or one or more additive(s), or the like, or any combination thereof.

[0014] In an aspect, the present disclosure provides films. In various examples, a film is formed by a method of the present disclosure. In various examples, a film comprises at least two multiblock copolymers, where the film comprises: an isoporous region or layer comprising two or more multiblock copolymers, where each multiblock copolymer comprisesone or more corona-forming block(s) and one or more core-forming block(s), and i) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core- forming block(s), a thickness of about 20 nm to about 500 nm, including all 0.1 nm values and ranges therebetween, an asymmetric region or layer comprising a thickness of about 5 microns to about 500 microns, including all 0.1 micron values and ranges therebetween, and a plurality of pores (e.g., pores independently, having a size (or an average size) of about 5 nm to about 100 microns, including all 0.1 nm values and ranges therebetween), where, the isoporous surface region is disposed on at least a portion, substantially all, or all of an exterior surface of the asymmetric pore region, at least a portion, substantially all, or all of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of the core blocks of a second multiblock copolymer core is / are isotropically or statistically distributed in at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer, or is / are disposed in a plurality of distinct domains forming at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer, or form(s) a three-dimensional pore lattice in at least a portion, substantially all, or all of the pore wall or pore wall surface(s) of the isoporous region or layer, and at least a portion, substantially all, or all of the functional groups of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of a functional group or functional groups of the core blocks of a second multiblock copolymer core are disposed on at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer.

[0015] In an aspect, the present disclosure provides uses of films of the present disclosure. In various examples, a device (such as, for example, a filtration device, a separation device, or the like comprises one or more film(s) of the present disclosure.

[0016] A film or films can be used in filtration applications, separation applications, catalytic separations, or the like, or any combination thereof. In various examples, a device is configured for one or more use(s) described herein.

[0017] In various examples, the instant methods allowed for film formation where a desired pore surface functionalization (e.g., spatial and areal control of pore surface functionalization) was realized. The controlled pore surface functionalization can expand applications, such as, for example, filtration and the like, beyond mere size exclusion and combine size exclusion with solute-film affinity. These films are useful in filtration, separation, and catalysis applications.BRIEF DESCRIPTION OF THE FIGURES

[0018] For a fuller understanding of the nature and objects of the disclosure, reference should be made to the following detailed description taken in conjunction with the accompanying figures.

[0019] FIG.1 shows a schematic of M2SNIPS membrane formation method and mixed membrane surface structures. a, Chemical structures of all components and schematic of M2SNIPS membrane formation process from two chemically distinct triblock terpolymers. b, Schematic of possible M2SNIPS membrane top surface structure consisting of 2D square packed and 2D hexagonally packed regions. c, Extension of M2SNIPS process to system consisting of three chemically distinct triblock terpolymers. Micelles not drawn to scale.

[0020] FIG.2 shows surface characterization of pure and blended membranes. a-d, Top surface SEM images of pure (ISV, SVPS), “patchy” (50 ISV:50 SVPS53 by wt.), and “disordered” (65 ISV:35 SVPS67 by wt.) membranes (Table 2). e-h, Segmentation of a-d into ISV pores (red dots), SVPS pores (green dots) and matrix (magenta background). Higher magnification insets (scalebar: 100 nm). i-l, STORM images depicting SVPS component (light areas) distribution for membranes a-d. m-p, ISV grain size comparison between “patchy” (m and n; same as f) and “disordered” (o and p; same as FIG.3a) membranes (blue lines: grain boundaries). q, FFT analysis of a and d. Tick marks indicate peak positions consistent with 2D square (ISV) and 2D hexagonal (SVPS) structures. Lattice spacings calculated via d = 2π / q* using first peak position. r, Quantitative comparison between SEM segmentation (e-h) and STORM (i-l) localization. Left axis: Area fraction of SVPS pores on membrane top surface, normalized to that of the pure SVPS membrane (d). s, Nearest neighbor distance comparison between ISV and SVPS populations in pure (FIG.11e,j) and “disordered” (FIG.3e) membranes at 100 s (s = second(s)) based on segmentation classification.

[0021] FIG.3 shows effect of evaporation time on M2SNIPS membrane surface composition and structure. a-e, Segmented SEM images of membrane top surfaces derived from a 67 ISV: 33 SVPS53(by wt.) casting dope and evaporated for 60 s, 70 s, 80 s, 90 s, and 100 s, respectively (Table 2). f-j, MVA corresponding to images depicted in a-e. Red, blue, and white Voronoi cells represent ISV, SVPS, and “defects”, respectively. k-o, MVA of Brownian dynamics (BD) simulation box top layer snapshots. Chosen timepoints match the experimental MVA based ISV compositions. p-s, Comparison between SVPS-SVPS pore spacing distributions in pure (p) and blended (b) membranes from 70 to 100 s evaporation time, respectively. Datapoints included neighboring SVPS pairs with sixfold symmetryaccording to MVA and local bond orientation, Φ6> 0.80. Corresponding Kernel density estimation (KDE) and standard deviation for each distribution. t, Comparison between ISV- ISV pore spacing distributions in pure (p) and blended (b) membranes at 90 s evaporation time and corresponding KDE. Datapoints included neighboring ISV pairs with fourfold symmetry according to MVA and local bond orientation, Φ4 > 0.97.

[0022] FIG.4 shows BD results and effects of evaporation time and dope composition on M2SNIPS membrane surface composition and structure. a,b, BD simulation snapshots of ISV (red) and SVPS (blue) spheres at the interface, suggesting interlayer coordination and heterogeneity within grains and at grain boundaries, respectively. Second layer spheres depicted using lighter shades in b. c, Top and side views of a single SVPS sphere surface transition at four timepoints (left to right). SVPS micelle beneath an ISV defect disrupts local surface order upon emergence. d,e, A single SVPS surface transition (d) seeds further transitions (e) when underlying SVPS reservoir is present. f-h, Segmented top surface SEM images of membranes (80 s evaporation) derived from compositions as indicated (Table 2). i- k, MVA corresponding to f-h. l,m, Number fractions of ISV (4-fold, ISVMVA), SVPS (6-fold, SVPSMVA) and defect (non-four-or-sixfold, DefectMVA) MVA cells as well as of ISV and SVPS (ISVseg, SVPSseg) pores (SEM segmentation) from top surface images as a function of only evaporation time (l) or SVPS weight fraction in casting dope (relative to total polymer concentration; m), respectively. Error bars represent standard deviations from three sample regions.

[0023] FIG.5 shows sub-surface M2SNIPS membrane structure and composition as revealed by O2 plasma etching. a,c, Schematics of pore chemistry distributions throughout “patchy” (a) and “disordered” (c) membrane separation layers as suggested by a combination of etching results and surface analysis. Red and green pore walls denote ISV and SVPS based pore chemistries, respectively. b, Plasma etching geometry: a membrane is partially sandwiched between two Si wafers, creating a gradient in etch depth. Steeper gradient shown for visualization purposes only (actual gradients: FIG.25a and FIG.26a). SEM image depicting the 3-4 micelle layers constituting the ~100 nm separation layer (top right). Segmentation results and original SEM micrographs of: d-g, regions with shallow etches in “patchy” (50 ISV: 50 SVPS53by wt.) (d and e; FIG.27a,h) and “disordered” (59 ISV: 41 SVPS53 by wt.) (f and g; FIG.27p,w) membranes (Table 2); h-k, regions with deeper etches in “patchy” (h and i; FIG.27d,k) and “disordered” (j and k; FIG.27s,z) membranes (Table 2), respectively. Red, green, and magenta regions represent ISV matrix, SVPS matrix, and void space, respectively.

[0024] FIG.6 shows comparison of pH dependent permeability between two-component and three-component M2SNIPS membranes. a, Normalized permeability of pure versus two- component membranes (ISV, SVPS67 versus ISV:SVPS67 by wt.; Table 2) at 1 psi over a range of pH values. b, Normalized permeability of ISV versus three-component membranes (ISV:SVPS67:ISA by wt.; Table 2) at 1 psi over a range pH values. Indicated errors are standard deviations from three replicate measurements. Insets depict corresponding absolute permeabilities (LMH / bar). c, SEM micrograph of a representative three-component (43 ISV: 42 SVPS67: 15 ISA by wt.; Table 2) M2SNIPS membrane top surface. d, Cross-sectional SEM micrograph depicting the separation layer and macroporous substructure of membrane in c.

[0025] FIG.7 shows the Effect of blend time on M2SNIPS membrane top surface structure. Characterization of M2SNIPS membranes casted from a dope stirred for: a,b, 15 minutes (77 ISV: 23 SVPS53 by wt.); c,d, one hour (73 ISV: 27 SVPS53 by wt.); and e,f, 33 days (77 ISV: 23 SVPS53 by wt.). All samples were evaporated for 80 s. g, 1D FFT profiles of samples depicted in a-f. FFT traces of pure ISV and SVPS53 membranes evaporated for 80 s are included for comparison. Top surface SEM images (a,c,e) were binarized (b,d,f), from which corresponding 1D FFT profiles were then obtained via the Radial Profile ImageJ plugin (g). Tick marks for ISV (black), 15 minutes (blue), and 1 hour (purple) traces correspond to (q / q*)2= 1, 2, 4, and 5, consistent with a 2D square lattice. Tick marks for 33 days (fuchsia) and SVPS53 (red) traces correspond to (q / q*)2= 1, 3, 4, and 7, consistent with a 2D hexagonal lattice. Domain spacings (d= 2π / q*) calculated from the position of the primary correlation peak (q*) are as indicated.

[0026] FIG.8 shows SEM micrographs of (M2)SNIPS membrane cross-sections at various magnifications. a,d,g,j,m,p Full membrane cross-sections, b,e,h,k,n,q, increased magnification of membrane near top separation layer, and c,f,i,l,o,r, increased magnification of mesoporous substructure. Samples in the top and bottom rows were evaporated for 80 s and 100 s, respectively. Membranes of parent polymers (left, right panel) that were used to fabricate M2SNIPS membranes (center panel). See Table 2 for sample details.

[0027] FIG.9 shows top surface SEM image processing protocol using ImageJ. a, Raw SEM image; b, bandpass filtered image, filtering out structures larger than 40 pixels or smaller than 3 pixels; c, pixel-based segmentation using TWS plugin of ImageJ; d, segmentation after removal of objects < 78.5 nm2, i.e., < 10 nm diameter assuming perfectly circular objects; and e, segmentation after pores containing more than one pixel class (red + green) were adjusted to only reflect the class with more pixels (only red or only green). Red,green, and magenta pixels denote ISV pores, SVPS pores, and the membrane matrix, respectively. Ratios in the upper right corners of c-e represent relative area fractions of ISV:SVPS pores for the example region shown. Changes resulting from the additional image processing steps did not affect the qualitative experimental trends observed, i.e., they were minor (< 3%) and systematically shifted towards higher ISV fractions regardless of evaporation time. f, Binarized version of e used to determine pore size distribution, pore density and porosity. g,h, Split of e into ISV and SVPS classes, respectively, to analyze same- component lattice spacing.

[0028] FIG.10 shows average gray value distributions for evaporation time and composition series depicted in FIG.3 and FIG.4. Distribution of average gray values for: top, evaporation time series and bottom, composition series. Values of 0 and 1 denote that all pixels in a given pore were classified as ‘ISV’ and ‘SVPS’, respectively. Intermediate values denote a mix of pixel classes. The results for three separate regions of the same sample (i.e., 60 s evaporation) were overlaid to assess sample uniformity, which is high.

[0029] FIG.11 shows SEM images of top surfaces of pure ISV and SVPS53 membranes as a function of evaporation time. a-e, ISV membranes derived from a 11 wt. % ISV solution in 7 DOX: 3 THF (wt.%) and evaporated between 60 and 100 s (as indicated; Table 2). f-j, SVPS53 membranes derived from a 15 wt. % SVPS53 solution in 8 DOX: 2 THF (wt.%) and evaporated between 60 and 100 s (as indicated; Table 2).

[0030] FIG.12 shows pore size distributions of pure membranes. a, Pore size distribution of membrane derived from a 11 wt. % ISV in 7 DOX: 3 THF (by wt.) solution (FIG.11c). b, Pore size distribution of membrane derived from a 15 wt. % SVPS53 in 8 DOX: 2 THF (by wt.) solution (FIG.11h). Evaporation time was 80 s for both membranes. Average pore sizes and standard deviations as indicated.

[0031] FIG.13 shows raw SEM images corresponding to FIG.3 and FIG.4. a-e, Effect of evaporation time (as indicated) on surface composition of blended (67 ISV: 33 SVPS53 by wt.) membranes (Table 2). f-h, Effect of casting dope composition (as indicated) on surface composition of blended membranes evaporated for 80 s. Labels depict relative wt.% of ISV:SVPS53 in casting dope (Table 2).

[0032] FIG.14 shows area and number fraction of ISV and SVPS pores on M2SNIPS membrane top surface according to SEM image segmentation. a, Evaporation time series (FIG.3a-e). b, Composition series (FIG.4f-h). Values displayed are calculated averages and standard deviations from three separate regions of a given sample.

[0033] FIG.15 shows methodology for revealing the inherent local square symmetry through Modified Voronoi Analysis (MVA). a,b, Colormap of the Voronoi cells constructed using Voronoi tessellation analysis (a) and the corresponding distribution of the number of neighbors, Ns(b), for membrane pores of pure ISV at 100 s evaporation time (FIG.11e). Inset in a shows smaller Voronoi edges (gray) caused by perturbations of the square lattice. δ is the angle subtended at the pore center from the vertices of the Voronoi cell edges. c, Distribution of δ for small (δ < 30°) and large (δ > 30°) Voronoi edges observed for pure ISV and SVPS membrane pores at 100 s evaporation time. d,e, MVA of the same region shown in a after removing small (δ < 30°) Voronoi edges (d) and resulting Nsdistribution (e). Inset in d shows same Voronoi cell as inset in a, now reflecting only the four nearest neighbors. d uses same color code as a.

[0034] FIG.16 shows distribution of number of neighbors (Ns) on M2SNIPS membrane top surface at different evaporation times. Distribution of Ns at different evaporation times (60 to 100 s) for membranes derived from a 67 ISV: 33 SVPS53(by wt.) solution (FIG.3f-j; Table 2). Ns was determined via MVA. Red, blue, and white bars correspond to populations classified as ISV, SVPS and defects, respectively. Each bar graph depicts cumulative data from three 2.5 μm x 2.5 μm SEM images.

[0035] FIG.17 shows comparison by means of n-bond orientational order parameters between membrane top surface pore geometry at 100 s evaporation time and the ordered structures of hard squares and hard disks simulated using isothermal-isobaric Monte Carlo simulations with 1600 particles. Structures are examined using the local n-bond orientational order parameter, ^^^^, where n= 4 or 6 for 4-fold or 6-fold symmetries, respectively (Methods). a-d, Distribution of the local ^^4(c) and ^^6(d) values for a square ordered structure (a) with packing density, η= 0.92 and the pore structure of a pure ISV (FIG.11e) membrane (b). e-h, Distribution of the local ^^4(g) and ^^6(h) values for a hexagonal ordered structure (e) with η= 0.74 and the pore structure of a pure SVPS53 (FIG.11j) membrane (f).

[0036] FIG.18 identifies highly ordered ISV and SVPS grains on the top surface of a 67 ISV: 33 SVPS53 (by wt.) solution blended system at longer evaporation times of 90 s and 100 s by means of n-bond orientational order parameters. ISV (red), SVPS (blue), and defect (white) populations are identified using MVA for: a, 90 s and b, 100 s evaporation times (FIG.3d,e). Bolded pore centers for ISV at 90 s (a) have ^^4> 0.97, while those of SVPS pores at 100 s (b) have ^^6> 0.8 (see FIG.17c,h). Circled regions show examples of ISV and SVPS clusters with high 4- and 6-fold order, respectively.

[0037] FIG.19 shows nearest neighbor distance comparison between pure and “disordered” (67 ISV : 33 SVPS53by wt.) membranes according to SEM image segmentation classification. Lattice spacing comparison between: top, pure ISV membranes (FIG.11a-d) and the ISV component in “disordered” membranes, and bottom, pure SVPS membranes (FIG.11g-j) and the SVPS component in “disordered” membranes for different evaporation times as shown in FIG.3 (Table 2). Histograms for blended components contain data from three separate 2.5μm x 2.5μm regions. For pure ISV membranes, objects smaller than 153.9 nm2, i.e. < 14 nm in diameter assuming perfectly circular objects, were removed to capture the primary lattice spacing. Standard protocol (removing <78.5 nm2objects) was used for all other samples.

[0038] FIG.20 shows local pair correlation of pure and “disordered” (67 ISV: 33 SVPS53 by wt.) membranes. a-d, Comparison of ISV-ISV lattice spacing in pure and “disordered” membranes from 60 to 90 s evaporation time, respectively (Table 2). Datapoints included neighboring ISV pairs with 4-fold symmetry according to MVA and local bond orientation, Φ4 > 0.97 (Methods). e-h, Comparison of SVPS-SVPS lattice spacing in pure and ‘disordered’ membranes from 70 to 100 s evaporation time, respectively. Datapoints included neighboring SVPS pairs with 6-fold symmetry according to MVA and local bond orientation, Φ6> 0.80. Corresponding Kernel Density Estimate (KDE), average, and standard deviation for pure (p) and blend (b) distributions.

[0039] FIG.21 shows a schematic of coarse-grained colloidal model. Self-assembled BCP micelles are modeled as soft spheres interacting via colloidal potentials.

[0040] FIG.22 shows a plot of soft, monotonic, purely repulsive Yoshida-Kamakura (YK) potential. YK potentials displaying simple cubic (SC) and body–centered cubic (BCC) structures in the bulk exhibit square and hexagonal organization at interfaces, respectively. Plot reflects modifications explained in text.

[0041] FIG.23 shows a schematic of BD simulation interfacial model. a, The functional forms of the chosen potential (see Methods). b, At the interface, there is an outer short-range highly repulsive region (black), which prevents particles from escaping, followed by an inner longer-range soft repulsive region (light red / light blue) where both ISV (red) and SVPS (blue) particles are repelled. At early times, the repulsion for both particle types is equal, and ISV populates the surface layer. c, The effect of solvent evaporation is approximated by modifying the strength of the repulsion parameter, K, strengthening it for ISV and softening it for SVPS. d, The final wall potential is shown, resulting in an abundance of SVPS at the interface.

[0042] FIG.24 shows component distribution along cross-section of BD simulation box. left, Cross-section snapshot of simulation box surface layers; ISV (red) and SVPS (blue). right, Histogram of particle distribution along cross-section (Z). SVPS component depletion at the interface is compensated for in the bulk.

[0043] FIG.25 shows SEM characterization of O2 plasma etched “patchy” M2SNIPS membrane. a, Regions of interest overview. b-i, SEM images taken at locations indicated in a. Membrane was derived from a 50 ISV: 50 SVPS53 dope evaporated for 80 s (Table 2), and subjected to O2plasma etching for 30 s.

[0044] FIG.26 shows SEM characterization of O2 plasma etched “disordered” M2SNIPS membrane. a, Regions of interest overview. b-i, SEM images taken at locations indicated in a. Membrane was derived from a 59 ISV: 41 SVPS53 dope evaporated for 80 s (Table 2), and subjected to O2 plasma etching for 30 s.

[0045] FIG.27 shows comparison of component distribution beyond the M2SNIPS membrane top surface. a-g, SEM micrographs of “patchy” (50 ISV: 50 SVPS53 by wt.; Table 2) membrane O2 plasma etched for 30 s. h-n, Segmented images corresponding to a-g. o-u, SEM micrographs of “disordered” (59 ISV: 41 SVPS53 by wt.; Table 2) membrane O2 plasma etched for 30 s. v-ab, Segmented images corresponding to o-u. Red (dark spots), green (light spots) and magenta (background) regions represent the ISV matrix, SVPS matrix, and void space, respectively. Numbering in lower left corner is consistent with that in FIG.25a and FIG.26a.

[0046] FIG.28 shows top surface SEM characterization of two-component (ISV121+ SVPS67) membranes. M2SNIPS membranes derived from: a, 58 ISV:42 SVPS67, b, 60 ISV:40 SVPS67, and c, 65 ISV:35 SVPS67(by wt.) solutions. All samples were evaporated for 100 s. Detailed sample characteristics are outlined in Table 2.

[0047] FIG.29 shows micellar block copolymer (BCP) building blocks and combinations thereof from which porous films were derived, as well as illustration of surface porosity for structural identification. (A) Chemical structures and micellar representations of BCP building blocks from which pure, binary, and ternary porous films are derived via evaporation induced non-equilibrium BCP self-assembly and phase inversion (SNIPS). Polyisoprene, polystyrene, poly(4-vinylpyridine), poly(2-vinylpyridine), poly(ethylene oxide), and poly(propylene sulfide) blocks abbreviated as I, S, V (red), V (blue), O, and PS, respectively (naming convention: corona｜core). Chemically distinct blocks assembled into the same micelle core / corona are depicted in the same color. (B) Schematic illustrating waysmicellar building blocks can be combined. χΑΑN = 0 and χABN >> 0, respectively, denote the enthalpically expected absence and presence of macrophase separation in the resulting films. Porosity introduced as a means by which micellar building blocks can be identified via differences in pore size and pore lattice geometry.

[0048] FIG.30 shows an overview of primary analytical methods applied to characterize film surface structures of pure (single-component) BCP micelle systems. (A to C) Top surface SEM micrographs and (D to F) derived segmentations of pure (A,D) ISV121, (B,E) ISVO165, and (C,F) SVPS53films evaporated for 80 s. Red, blue, green, and magenta pixels respectively represent ISV, ISVO, SVPS pores and the solid matrix. Insets depict enlarged 250 nm x 250 nm regions highlighting the individual lattice geometries. (G) Depiction of a coarse-grained ISV micelle, alongside Brownian Dynamics (BD) simulation box surface snapshots of square-packed (red) and hexagonally packed (green) single-component systems. (H) Pore size distributions (PSDs) of pure ISV121, ISVO165, and SVPS53 film surfaces derived from the segmentations in (D to F). (I) Fast Fourier transform (FFT) image analysis of (D to F) with lattice spacings derived from the first peak positions (q*) as labeled. Tick marks accompanying the ISV and ISVO traces correspond to (q / q*)2= 1, 2, 4, 5, 8, 9, and 10, consistent with a 2D square lattice. Tick mark for SVPS trace correspond to (q / q*)2= 1, 3, 4, and 7, consistent with a 2D hexagonal lattice.

[0049] FIG.31 shows characterization of surface structures derived from three binary mixtures of chemically distinct BCP micelles at two stoichiometries each. Top surface SEM images derived segmentations of binary (A,B) ISV121+ISVO165, (C,D) ISVO165+SVPS53, and (E,F) ISV121+SVPS53 films evaporated for 80 s. Red (dots), blue (dark dots), green, and magenta (background) pixels respectively represent ISV121, ISVO165, SVPS53pores, and the solid matrix. Compositions by wt. as labeled following the same color code. The larger non- porous spaces in panels (A) and (B) are a result of image processing (Methods). (G to L) BD simulation box surface snapshots of binary systems modeled at the labeled wt. compositions. Each pair of snapshots (e.g., G and H) reflects the same simulation time (Methods). FFT image analysis of 75:25 (M) ISV121: ISVO165, (N) ISVO165: SVPS53, and (O) ISV121: SVPS53 samples (by wt.) was performed after the populations were split (Methods). FFT traces from split populations are sandwiched between those of the corresponding pure samples (top and bottom in M to O). Vertical reference lines –corresponding to first four peak positions of the uppermost trace – serve as visual aids.

[0050] FIG.32 shows characterization of surface structures in binary macrophase separated ISV+SVPS systems from different degrees of size asymmetry. Micelles outlined inthe same color (top) are derived from the same BCP (e.g., ISV micelles depicted for ^^ =1.8 and ^^ = 2.3). SEM derived object-based segmentations of (A) ISV121+ SVPS67, (B) ISV121+ SVPS53, (C) ISV138 + SVPS53, and (D) ISV164 + SVPS53 film surfaces evaporated for 80 s, all made from 75:25 ISV:SVPS (by wt.) solutions blended for 15 minutes. Red (dots), green (light dots), and magenta (background) regions respectively represent ISV, SVPS, and the solid matrix. ^^ denotes BCP size ratio ( ^^^^, ^^ ^^ ^^ / ^^^^, ^^ ^^ ^^ ^^). (E) Quantitative comparison between outcomes of segmentation (s) and MVA (v). Averages and standard deviations from three separate regions reported. For MVA, the following correspondence between classification and number of cell edges (nearest neighbors, N) was used: ISV (N = 4), SVPS (N = 6), defect (def; N ≠ 4 or 6). (F to H) Monodisperse and (J to L) polydisperse BD simulation surface snapshots corresponding to size asymmetries of ^^ = 2.3, 2.6, and 3.1, respectively (Methods). (I) Approximation of BD simulation polydispersity via relative frequencies of five discrete populations (as labeled) derived from the experimental pure film PSD.

[0051] FIG.33 shows characterization of film surface structures from ternary BCP micelle system. (A to C) SEM segmentations of ternary ISV121:SVPS53:ISVO165film surfaces (relative compositions as labeled) evaporated for 80 s. Red, green, blue, and magenta regions represent ISV, SVPS, ISVO, and the solid matrix, respectively (Methods). (D) FFT image analysis of the ISV and ISVO regions (red and blue, collectively) in (A to C). Lattice spacing derived from the position of the first peak (q*) as labeled. Vertical reference lines reflect expected peak positions up to (q / q*)2= 1, 2, 4, and 5 for a 2D square lattice with 46.3 nm spacing, highlighting the compositionally dependent lattice expansion. (E) Relative solution compositions of (A to C) by wt. depicted on a ternary ISV-SVPS-ISVO phase diagram. (F to H) ^^4cluster analysis of (A to C), respectively, with ten largest detected clusters shown. Pores in the same cluster depicted in the same color (Methods). (I) Box-and-whisker plot summarizing ^^4cluster analysis statistics. Statistics for the 10 largest detected clusters from three separate 2500 nm x 2500 nm regions (i.e.30 clusters per sample composition) were reported. Data points outside 1.5x the interquartile range depicted as outliers (hollow circles).

[0052] FIG.34 shows non-equilibrium derived experimental ternary surface morphology diagram. Experimental non-equilibrium ISV121-SVPS53-ISVO165 ternary surface morphology diagram of film surface compositions ( ^^^^) following 80 s of evaporation. Each sample is denoted by a two- or three- colored circle, whose position and color reflect the relative solution (wt. %) and surface (number %) compositions, respectively. Surface compositionsderived from averages of three separate 2500 nm x 2500 nm experimental sample regions. Regions of surface morphology diagram shaded to reflect majority surface component. (A to D) 2D equilibrium Monte Carlo (MC) simulation snapshots whose number fractions correspond to the surface compositions of the experimental samples framed in the same color.

[0053] FIG.35 shows synthesis scheme and characterization of novel ISVO tetrablock quarterpolymer. (A) Schematic depiction of anionic polymerization protocol for the synthesis of poly(isoprene-b-styrene-b-2-vinylpyridine-b-ethylene oxide) (ISVO) tetrablock quarterpolymer. (B) THF GPC elugram of final tetrablock quarterpolymer. (C)1H NMR spectrum of ISVO. Relevant protons labeled.

[0054] FIG.36 shows pore size distributions (PSDs) of pure BCP evaporated for 80 s. For each sample, statistics were derived from the distributions of three 2500 nm by 2500 nm regions (region A, B, and C). Segmented surface SEM image of one such region shown for each sample. Mean (μ), standard deviation (σ), skewness, coefficient of variation (CV, σ / μ), and interquartile range (IQR) are as reported.

[0055] micelle coarse graining for simulations. (A) At the experimentallyrelevant concentrations (above the critical micelle concentration), BCPs are micellized. These micelles can be modelled as soft, colloidal spheres. (B) The interactions between different micelle species (denoted by different colors) can be approximated using appropriate colloidal interaction potentials (example of potential form shown).

[0056] FIG.38 shows Yoshida-Kamakura (YK) potential for simulations. The YK potential is a soft, monotonic, purely repulsive potential. It can exhibit both simple cubic / square (bulk / interface; red curve) and BCC / hexagonal (bulk / interface; green curve) packing geometries. Simple mixing rules can be used to find interactions between distinct micelle species (e.g., purple curve).

[0057] FIG.39 shows scaled and modulated interfacial wall potentials for BD simulations. (A) The hard and short repulsive potential (gray) stops the micelles from escaping the simulation box. The extent and strength of the chemically specific potential (blue, red, and green rectangles) depends on the size of the micelle species. interface is modelled by repulsive walls for each species in the system. (B) Some characteristic wall potentials, ^^^^^^( ^^), are shown in the plot for comparison. In black, the wall at the beginning of the red and green, the wall at the end for the different chemical species. Inset shows the time evolution of the wall strength ( ^^) during a BD simulation.

[0058] FIG.40 shows demonstration of negligible molecular exchange between micelles on experimental timescale in binary / blended samples. (A to D) Segmented SEM micrographs for single component / pure (A, B) and binary / blended (C, D) film surfaces (as labeled), (E) corresponding FFT analysis, (F) FFT comparison between pure ISV and blend ISV components, and (H) FFT comparison between pure SVPS and blend SVPS components. Examples of isolated / separated (G) blend ISV and (I) SVPS components, respectively. Blended membranes were derived from the same 60:40 ISV121:SVPS67 (by wt.) solution in 7:3 DOX:THF. After 15 minutes of stirring at 200 rpm, 100μL of solution was casted to obtain (C). The remainder was stirred for 3 days and casted to obtain (D). All samples were evaporated for 100 s. Dotted lines in (F) and (H) correspond to the expected peak positions for the indicated lattice geometries of the relevant single component / pure sample. See FIG. 45 for raw SEM of (C) and (D).

[0059] FIG.41 shows segmentation protocol for SVPS containing binary and ternary blends. (A) Bandpass filtered SEM micrograph, (B) TWS segmentation, (C) application of size threshold (> 10 nm) to (B), with removed domains outlined in black, and (D) transformation from pixel-based segmentation to object-based segmentation (see F) of 50:50 ISV121:SVPS53sample evaporated for 80 s. (E) Distribution of mean domain values, where red and green pixels represent values of 0 and 1, respectively (e.g., mean = 1 / 9 for the pixel- based segmentation in F). (F) To transform from a pixel-based to an object-based segmentation, pixel values in each domain are adjusted to reflect the majority (i.e., [0,0.5] ^ red). (G) Bandpass filtered SEM micrograph, (H) TWS pixel-baseddistribution of mean domain values, and (J) final object-based segmentation of 50:50 ISVO165:SVPS53sample evaporated for 80 s. (K) Bandpass filtered SEM micrograph, (L) object-based segmentation into non-SVPS and SVPS domains, (M) separation of non-SVPS domains into ISV and ISVO based on 24.5 nm cut-off, and (N) final segmented image of a 36:28:36 ISV121:SVPS53:ISVO165 sample evaporated for 80 s. In (M), examples of non- primary-lattice domains misidentified as SVPS are outlined in yellow. These were targeted for removal, with thereby removed domains indicated in yellow in (N). See Methods for more details.

[0060] FIG.42 shows characterization of binary ISV121+ ISVO165blended film sample surfaces. SEM micrographs and corresponding segmentations (red, blue, and magenta pixels represent ISV121, ISVO165, and the matrix, respectively) of (A,D) 27:73, (B,E) 50:50, and (C,F) 74:26 ISV121:ISVO165 (by wt.) samples evaporated for 80 s. (G) FFT analysis of aforementioned ISV121+ ISVO165samples alongside those of reference pure ISV121and pureISVO165samples. (H) Experimentally determined (calculated) ISVO165numerical surface fraction (fISVO, surface) plotted against weight fraction of ISVO in casting solution (fISVO, solution). For a detailed description of statistical compensation for ISV / ISVO thresholding and missing lattice domains see Supplementary Discussion 1.

[0061] FIG.43 shows binary ISV121 + ISVO165 blend segmentation protocol and associated rationale. (A) Pore size distributions of pure ISV121and ISVO165samples evaporated for 80 s (individually and then overlaid, as labeled). Pore areas extracted from the segmentation analysis (e.g., panel E) were converted to pore diameters by assuming perfect circularity (i.e., ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ = 2 ∗√^^ ^^ ^^ ^^^^ ). Gray shaded areas depict fraction of pores > 24.5 nm. In the overlaid graph, the the shaded area comprise 91.9% ISVO165. Reportedpercentages are averages three separate sample regions (for clarity, only one such region is shown). (B) Nearest neighbor distance (NND) distribution of ISV121 sample set shown in (A). Over 98% of the data exceeds NND = 35 nm (indicated by dashed line). (C) Bandpass filtered SEM micrograph, and (D) pore-matrix segmentation following removal of image artefacts (domains < 78.54 nm2) of a 50:50 ISV121:ISVO165sample evaporated for 80 s. The inset differentiates between domains that are (solid black circles) and are not (hollow red circles) part of the “primary lattice” (domain spacing > 40 nm). (E) To highlight the primary domain spacing for subsequent FFT analysis, domains not part of the primary lattice (outlined in red) were filtered out according to nearest neighbor distance and domain size criteria (see Methods). While this did not remove 100% of the secondary domains, results were sufficient for FFT. (F) Final segmentation of the same 50:50 ISV121:ISVO165 sample evaporated for 80 s. Red, blue, and magenta represent ISV, ISVO and matrix domains, respectively, with pores colored red or blue according to the described 24.5 nm diameter threshold.

[0062] FIG.44 shows dynamic light scattering (DLS) characterization of 3 wt.% BCP solutions (as labeled) in 7:3 DOX:THF (by wt.). (A) Autocorrelation functions and (B) distribution of hydrodynamic diameters for the unimer (smaller diameter) and micelle (larger diameter) populations.

[0063] FIG.45 shows raw SEM micrographs from which segmentations were derived. SEM micrographs of (A) 75:25 ISVO165:SVPS53, (B) 50:50 ISVO165:SVPS53, (C) 75:25 ISV121:SVPS53(D) 50:50 ISV121:SVPS53; 75:25 (E) ISV121:SVPS67,(F) ISV121:SVPS53, (G) ISV138:SVPS53, (H) ISV164:SVPS53, (I) 50:30:20 ISV121:SVPS53:ISVO165, (J) 35:30:35 ISV121:SVPS53:ISVO165, (K) 20:30:50 ISV121:SVPS53:ISVO165, (L) 60:40 ISV121:SVPS67(15minutes), (M) ISV138, (N) ISV164, (O) SVPS67, and (P) 60:40 ISV121:SVPS67(3 days) by wt. Corresponding figure panels indicated in bottom left corner of each image. Panels C and F are the same. Shown twice for ease of comparison within a series. For more sample details, see Table 7.

[0064] FIG.46 shows modification of Voronoi tessellation analysis to capture inherent square symmetry. (A) Segmented region of an 11 wt.% ISV121sample evaporated for 80 s. (B) Voronoi tessellation analysis (VTA) carried out using centroids of porous domains (black) in (A). Cells color coded based on number of neighbors (N). Inset shows the local geometry of a single Voronoi cell, where δ denotes the angle subtended by a cell edge. The inclusion of smaller edges (gray) arising from minor perturbations in the square lattice gives the impression of six-fold (green), rather than four-fold (red), symmetry. (C) The modified Voronoi analysis (MVA) capturing only edges with δ > 30°. δ = 30° cut-off was chosen based on (D) the δ distribution of pure ISV121 and SVPS67 samples, with the aim of eliminating the smaller ISV edges while minimizing the impact on the six-fold SVPS symmetry. (E) Bar graph showing the distribution of the number of nearest neighbors each ISV121 domain had before (black; VTA) and after (blue; MVA) the imposed angular cut-off. Sample region depicted in (A to C) stems from larger 2500 nm x 2500 nm region from which the statistics in (D and E) were obtained.

[0065] FIG.47 shows number of neighbors distribution in binary blends according to MVA. Results of MVA performed on top surfaces of films derived from various 75:25 ISV:SVPS (by wt.) solutions evaporated for 80 s (BCP combinations as labeled). Bar colors (bottom) consistent with Voronoi cell colors in A to D, illustrating that defects (N ≠ 4 or 6) predominantly have 5 nearest neighbors. Data for each BCP combination derived from averages and standard deviations of three separate 2500 nm x 2500 nm regions (one of which is shown in the top row of this figure).

[0066] FIG.48 shows comparing surface compositions of binary ISV+SVPS blends after normalizing or accounting for ISV:SVPS solution molar ratios. (A) Normalization of segmentation data presented in FIG.32E (solution = 75:25 wt. % ISV:SVPS) to account for the differences in molar ratio introduced by size asymmetry (e.g., 75:25 wt. % converts to 65:35 and 50:50 mol. % for ^^ = 1.8 and ^^ = 3.1, respectively). The surface number fractions (%) were divided by the respective molar fractions (%) to obtain the plotted dimensionless quantity, which follows the same general trend as seen in FIG.32E. (B to E) Segmented surface structures of ^^ = 1.8 (ISV121+ SVPS67) and ^^ = 3.1 (ISV164+ SVPS53) films derived from different molar ratios (as labeled; ISV and SVPS in red and green, respectively),demonstrating that the trend observed in FIG.32 is more strongly associated with size asymmetry than with molar ratio. Please note that while the sample in (D) has slightly less ISV in solution than that in (E), additional ISV would only be expected to bolster the proposed trend. (F and G) Raw SEM micrographs of (C) and (D), respectively. For those corresponding to (B) and (E), see FIG.45.

[0067] FIG.49 shows comparison of ISV surface fraction trends between monodisperse and polydisperse BD simulations before and after scaling. (A) System specific comparison between monodisperse and polydisperse surface composition evolution (size asymmetry ^^ as labeled). (B) Scaled monodisperse and polydisperse BD simulation trends to correct for the fact that the different scaling factors ( ^^^^ ^^ ^^ ^^; see last three rows of Table 8) correspond to the same SVPS53 BCP. Surface snapshots in FIG.32 correspond to the denoted timepoint, chosen to match the surface structure and composition of the ^^ = 2.3 system.

[0068] FIG.50 shows cluster analysis methodology for ternary blends. (A) Domain centroids of a 50:30:20 ISV121:SVPS53:ISVO165ternary blend sample surface evaporated for 80 s. Inset demonstrates how the local Φ4 bond orientational order vectors are calculated for a central domain k and its four nearest neighbors j (see Methods). (B) Global radial distribution function, g(r), derived from (A). (C) Estimated correlation between two neighboring domains k and j, d4(j,k), using the dot product of the local Φ4 bond orientational order vectors. (D) Ten largest clusters detected in structure (A) color-coded by size (number of domains per cluster). The cluster domains have at least two nearest neighbor connections satisfying d4(j,k)> 0.6. DETAILED DESCRIPTION OF THE DISCLOSURE

[0069] Although subject matter of the present disclosure is described in terms of certain embodiments and examples, other embodiments and examples, including embodiments and examples that do not provide all of the benefits and features set forth herein, are also within the scope of this disclosure. For example, various structural, logical, and process step changes may be made without departing from the scope of the disclosure.

[0070] As used herein, unless otherwise indicated, “about”, “substantially”, or “the like”, when used in connection with a measurable variable (such as, for example, a parameter, an amount, a temporal duration, or the like) or a list of alternatives, is meant to encompass variations of and from the specified value including, but not limited to, those within experimental error (which can be determined by, e.g., a given data set, an art accepted standard, etc. and / or with, e.g., a given confidence interval (e.g., 90%, 95%, or more confidence interval from the mean), such as, for example, variations of + / -10% or less, + / -5%or less, + / -1% or less, and + / -0.1% or less of and from the specified value), insofar such variations in a variable and / or variations in the alternatives are appropriate to perform in the instant disclosure. As used herein, the term “about” may mean that the amount or value in question is the exact value or a value that provides equivalent results or effects as recited in the claims or taught herein. That is, it is understood that amounts, sizes, compositions, parameters, and other quantities and characteristics are not and need not be exact, but may be approximate and / or larger or smaller, as desired, reflecting tolerances, conversion factors, rounding off, measurement error, or the like, or other factors known to those of skill in the art such that equivalent results or effects are obtained. In general, an amount, size, composition, parameter, or other quantity or characteristic, or alternative is “about” or “the like,” whether or not expressly stated to be such. It is understood that where “about,” is used before a quantitative value, the parameter also includes the specific quantitative value itself, unless specifically stated otherwise.

[0071] Ranges of values are disclosed herein. The ranges set out a lower limit value and an upper limit value. Unless otherwise stated, the ranges include the lower limit value, the upper limit value, and all values between the lower limit value and the upper limit value, including, but not limited to, all values to the magnitude of the smallest value (either the lower limit value or the upper limit value) of a range. It is to be understood that such a range format is used for convenience and brevity, and thus, should be interpreted in a flexible manner to include not only the numerical values explicitly recited as the limits of the range, but also to include all the individual numerical values or sub-ranges encompassed within that range as if each numerical value and sub-range is explicitly recited. To illustrate, a numerical range of “0.1% to 5%” should be interpreted to include not only the explicitly recited values of 0.1% to 5%, but also, unless otherwise stated, include individual values (e.g., 1%, 2%, 3%, and 4%) and the sub-ranges (e.g., 0.5% to 1.1%; 0.5% to 2.4%; 0.5% to 3.2%, and 0.5% to 4.4%, and other possible sub-ranges) within the indicated range. It is also understood (as presented above) that there are a number of values disclosed herein, and that each value is also herein disclosed as “about” that particular value in addition to the value itself. For example, if the value “10” is disclosed, then “about 10” is also disclosed. Ranges can be expressed herein as from “about” one particular value, and / or to “about” another particular value. Similarly, when values are expressed as approximations, by use of the antecedent “about,” it will be understood that the particular value forms a further disclosure. For example, if the value “about 10” is disclosed, then “10” is also disclosed.

[0072] As used herein, unless otherwise stated, the term “group” refers to a chemical entity that is monovalent (i.e., has one terminus that can be (is) covalently bonded to other chemical species), divalent, or polyvalent (i.e., has two or more termini that can be (are) covalently bonded to other chemical species). The term “group” also includes radicals (e.g., monovalent radicals and multivalent radicals, such as, for example, divalent radicals, trivalent radicals, and the like). Non-limiting, illustrative examples of groups include: refers to branchedor groups one or more degree(s) of unsaturation. Degrees of unsaturation can arise from, but are not limited to, carbon-carbon double bond group(s), carbon-carbon triple bond group(s), cyclic aliphatic group(s), or the like, or any combination thereof. For example, the aliphatic groups / moieties are a C1to C12aliphatic group, including all integer numbers of carbons and ranges of numbers of carbons therebetween (e.g., C1, C2, C3, C4, C5, C6, C7, C8, C9, C10, C11, or C12). Aliphatic groups include, but are not limited to, alkyl groups, alkene groups, and alkyne groups, and the like. In various examples, an aliphatic group is unsubstituted or substituted with one or more substituent(s). Examples of substituents include, but are not limited to, various substituents such as, for example, halogen groups (-F, -Cl, -Br, and -I), aliphatic groups (e.g., alkyl groups, alkenyl groups, alkynyl groups, and the like), halogenated aliphatic groups (e.g., trifluoromethyl group and the like), aryl groups, halogenated aryl groups, hydroxyl group, amine groups, nitro group, cyano groups, isocyano groups, azide group, silane groups (e.g., alkyl silane groups, aryl silane groups, alkyl / aryl silane groups, or the like), alkoxide groups, alcohol groups, ether groups, ketone groups, carboxylate groups, carboxylic acid groups, ester groups, amide groups, thioether groups, thioester groups, and the like, and any combination thereof.

[0074] As used herein, unless otherwise indicated, the term “alkyl group” refers to branched or unbranched hydrocarbon groups that include only single bonds between carbon atoms (not including substituent(s), if any). In various examples, an alkyl group is a C1 to C12 alkyl group (e.g., C1, C2, C3, C4, C5, C6, C7, C8, C9, C10, C11, or C12), including all integer numbers of carbons and ranges of numbers of carbons therebetween. In various examples, an alkyl group is a saturated group. In various examples, an alkyl group is a cyclic alkyl group, which may be a heterocyclic alkyl group, e.g., a monocyclic alkyl group or a polycyclic alkylgroup or the like. Examples of alkyl groups include, but are not limited to, methyl groups, ethyl groups, propyl groups, butyl groups, isopropyl groups, tert-butyl groups, cyclohexyl groups, adamantyl groups, benzyl groups and the like. In various examples, an alkyl group is unsubstituted or substituted with one or more substituent(s). Examples of substituents include, but are not limited to, various substituents such as, for example, halogen groups (-F, -Cl, -Br, and -I), aliphatic groups (e.g., alkyl groups, alkenyl groups, alkynyl groups, and the like), halogenated aliphatic groups (e.g., trifluoromethyl group and the like), aryl groups, halogenated aryl groups, hydroxyl group, amine groups, nitro group, cyano groups, isocyano groups, azide group, silane groups (e.g., alkyl silane groups, aryl silane groups, alkyl / aryl silane groups, or the like), alkoxide groups, alcohol groups, ether groups, ketone groups, carboxylate groups, carboxylic acid groups, ester groups, amide groups, thioether groups, thioester groups, and the like, and any combination thereof.

[0075] As used herein, unless otherwise stated, the term “structural analog” refers to any polymer, monomer (such as, for example, monomer(s) used to form a polymer or the like), or the like, or any portion thereof (such as, for example, one or more group(s) thereof or the like) or group if one atom or group of atoms, functional group or functional groups, or substructure or substructures is / are replaced with another atom or group of atoms, functional group or functional groups, substructure or substructures, or the like. In various examples, the term “structural analog” refers to any group that is derived from an original polymer, monomer (such as, for example, monomer(s) used to form a polymer or the like), or the like or portion thereof (such as, for example, one or more group(s) thereof or the like) or the like by a chemical reaction, where the polymer, monomer (such as, for example, monomer(s) used to form a polymer or the like), or the like or portion thereof (such as, for example, one or more group(s) thereof or the like) or the like is modified or partially substituted such that at least one structural feature of the polymer, monomer (such as, for example, monomer(s) used to form a polymer or the like), or the like or portion thereof (such as, for example, one or more group(s) thereof or the like) or the like is retained.

[0076] The present disclosure provides, inter alia, films (such as, for example, porous functionalized multiblock copolymer films and the like), compositions, and methods of making films (such as, for example, porous functionalized multblock copolymer films and the like). The present disclosure also provides uses of films (such as, for example, porous functionalized multiblock copolymer films and the like).

[0077] In an aspect, the present disclosure provides methods of making films. The films are porous films. In various examples, a film is a porous functionalized multiblock copolymerfilm, an isoporous graded film, an asymmetric film (e.g., comprising asymmetric pore structure along an axis substantially normal or normal to the film surface), or the like. In various examples, a method comprises use of a deposition solution (e.g., a composition of the present disclosure) to form a film. In various examples, a method comprises a block copolymer self-assembly and non-solvent induced phase separation (SNIPS) process or the like. In various examples, a method forms a film of the present disclosure. In various examples, a film produced by a method of the present disclosure comprises one or more feature(s) of film described herein. Non-limiting examples of methods of making films are provided herein.

[0078] In various examples, a method of forming a film (e.g., a film comprising an isoporous region or layer and an asymmetric region or layer, where the isoporous region or layer is disposed on at least a portion of an exterior surface of the asymmetric region or layer) (such as, for example, a porous functionalized multiblock copolymer film, an isoporous graded film, an asymmetric film (e.g., comprising asymmetric pore structure along an axis substantially normal or normal to the film surface), or the like) (e.g., a film of the present disclosure) comprises: forming a first film (e.g., forming a first film on a substrate or the like) from a deposition solution comprising two or more multiblock copolymers (e.g., where each multiblock copolymer comprises one or more corona-forming block(s) and one or more core- forming block(s), and at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s)); and two or more solvents, removing at least a portion of the solvent system from the film (e.g., allowing the first film to stand for a period of time such that at least a portion of the two or more solvents is removed from the film; and contacting the second film with a phase separation solvent system, such that the film (e.g., a film comprising an isoporous region or layer and an asymmetric region or layer, where the isoporous region or layer is disposed on at least a portion of an exterior surface of the asymmetric region or layer) (such as, for example, a porous functionalized multiblock copolymer film, an isoporous graded film, an asymmetric film (e.g., comprising asymmetric pore structure along an axis substantially normal or normal to the film surface), or the like) (e.g., a film of the present disclosure) is formed (which may be disposed on a substrate). In various examples, the deposition solution further comprises one or more homopolymer(s), one or more metal salt(s), one or more inorganic nanoparticle(s), or one or more additive(s), or the like, or any combination thereof. In various examples, a first multiblock copolymer ispresent in a plurality of first micelles and a second multiblock copolymer is present in a plurality of second micelles and there is no, no substantial, or no observable exchange of the polymer chains from the first micelles with the polymer chains of the second micelles.

[0079] A deposition solution may be formed from two or more distinct and / or independent multiblock copolymer solutions. In various examples, a deposition solution is formed from a first multiblock copolymer solution comprising one or more first multiblock copolymer(s) (e.g., which form a plurality of first micelles) and one or more first solvent(s) and a second multiblock copolymer solution comprising one or more second multiblock copolymer(s) (e.g., which form a plurality of second micelles) and one or more second solvent(s) (e.g., where a first multiblock copolymer solution and a second multiblock copolymer solution are combined (e.g., mixed or the like) to form a deposition solution). In various examples, the first multiblock copolymer solution and a second multiblock copolymer solution are combined (e.g. at a selected ratio and / or loading to achieve a desired amount of multiblock copolymer(s) in the deposition solution). In various examples, one or more or all of the first solutions are the same as one or more or all of the second solutions. In various examples, the first multiblock copolymer solution and a second multiblock copolymer solution are combined (e.g., mixed for time or the like) such that substantially all or all of the first micelles (formed from a first multiblock copolymer) and substantially all or all of the second micelles (formed from a second multiblock copolymer) are stable (e.g., the micelles do not exhibit any, any substantial, or any observable polymer chain exchange between first and second micelles in the deposition solution prior to film formation). Without intending to be bound by any particular theory, it is considered the lack of polymer chain exchange results in site-specific and / or areal control of film functionalization.

[0080] In various examples, a deposition solution comprises micelles. In various examples, each multiblock copolymer forms a plurality of micelles (e.g., distinct micelles) in the deposition solution. In various examples, micelles comprise a size (e.g., longest linear, cross-sectional dimension, such as, for example, a diameter or the like) of about 5 nm to about 50 nm, including all integer nm values and ranges therebetween (e.g., about 5 nm to about 30 nm). Without intending to be bound by any particular theory, it is considered molecular weight of a multiblock copolymer (or blocks thereof) is / are correlated to micelles size. In various examples, a deposition solution comprises a plurality of at least two (2–10 (2, 3, 4, 5, 6, 7, 8, 9, or 10)) different micelles. In various examples, each plurality of micelles is structurally different, compositionally different, or the like, or any combination thereof than the other pluarlit(ies) of micelles in the deposition solution. In various examples, eachplurality of micelles is different in terms of (i) lattice symmetry (different micelle populations may form identical lattices, distinct lattices, or no lattice at all); (ii) size (different micelle population sizes may be similar or vastly different, e.g., affecting their lattice spacings or the like); or (iii) corona chemistry (corona interactions between distinct micelle populations may be attractive or repulsive, i.e., χABN < 0 or χABN > 0), or any combination thereof.

[0081] In various examples, micelles of a deposition solution form one or more or a plurality of colloid layers (which form(s) an isoporous region or layer of the product film). In various examples, each colloid layer is formed from one or more or a plurality of different micelles (e.g., where the same or similar micelles form one or more discrete regions of an isoporous region or layer. In non-limiting, illustrative examples, an isoporous region or layer comprises a thickness of about 100 nm to about 150 nm and about 3 to about 5 colloid layers (e.g., that depend on the size of the micelles that form an individual colloid layer, which in turn depends on the molecular weight of the multiblock copolymer(s) (or block(s) thereof)).

[0082] In various examples, there is no substantial (e.g., 25% or less, 10% or less, 5% or less, 1% or less) or no chain exchange between micelles (e.g., micelles comprising the same and / or different multiblock copolymers) in the deposition solution (e.g., prior to film formation). In various examples, there is no or no observable chain exchange between micelles in the deposition solution. In various examples, the chain exchange (e.g., chain exchange on the timescale of film formation, such as, for example, on the minutes to an hour timescale) between the different micelles is negligible or not observed. In various examples, a plurality of at least two (2–10 (e.g., 2, 3, 4, 5, 6, 7, 8, 9, or 10)) different (e.g., compositionally and / or structurally different, such as, for example, chemically-distinct (e.g., at least one of the blocks of a multiblock copolymer in a chemically-distinct micelle comprises a different chemical composition and / or molecular weight than those in other chemically distinct micelle(s))) micelles does not exhibit any, any substantial, or any observable chain exchange (e.g., chain exchange on the timescale of film formation, such as, for example, on the minutes to an hour timescale) between the different micelles. Chain exchange between different micelles can be observed by methods known in the art. In various examples, chain exchange is observed by electron microscopy (such as, for example, by scanning electron microscopy (SEM), transmission electron microscopy (TEM), reflection electron microscopy (REM), or the like, or any combination thereof), x-ray structure (e.g., pore size or the like) determination, or Fast-Fourier Transform (FFT) analysis of electron microscopy (e.g., SEM, TEM, or the like) images. In various examples, isoporous region poresize or pore-size distribution correlates to chain exchange and is used to determine chain exchange, if any.

[0083] In various examples, the multiblock copolymers are selected to provide a desired Flory-Huggins interaction parameter, χ, (e.g., χABbetween monomers, such as, monomers A and B, that form the polymer chains of corona blocks of two distinct multiblock copolymer derived micelles) or product of the Flory-Huggins interaction parameter (e.g., between monomers, such as, monomers A and B, that form the polymer chains of corona blocks of two distinct multiblock copolymer derived micelles) and the degree of polymerization (such as, for example, ~χABN value (the product of the Flory-Huggins interaction parameter and the degree of polymerization for A, a corona region of a first multiblock copolymer and B, a corona region of a second multiblock copolymer)). In various examples, the degree of polymerization is the number of monomer units or is based on (e.g., calculated from) the molecular weight of the blocks of the multiblock copolymer. In various examples, the desired product of the Flory-Huggins interaction parameter and the degree of polymerization is about zero, is greater than about zero, or less than about zero. Without intending to be bound by any particular theory, it is considered that the product of the Flory-Huggins interaction parameter or the Flory-Huggins interaction parameter and the degree of polymerization results in a particular distribution and / or arrangement of the core blocks (e.g., as a particular pore surface functionalization) of the multiblock copolymers in the resulting film.

[0084] In various examples, at least a portion or all of the micelles exhibit attraction to other micelles in the deposition solution (such as, for example, ionic attraction, dipolar attraction, hydrogen bonding, or the like, or any combination thereof). In various examples, at least a portion or all of the micelles exhibit repulsion to other micelles in the deposition solutions (such as, for example, ionic repulsion, van der Walls interactions, or the like, or any combination thereof). In various examples, at least a portion, substantially all, or all of a plurality of first micelles and at least a portion, substantially all, or all of a plurality of second micelles exhibit attraction, exhibit repulsion, or do not exhibit attraction or repulsion (e.g., are neutral in their attraction). Without intending to be bound by any particular theory, it is considered that attraction or repulsion or lack thereof of the micelles results in a particular distribution and / or arrangement of the micelles and with them of the core blocks (e.g., a particular pore surface functionalization) of the multiblock copolymer micelles in the resulting film (e.g., it is considered deposition solutions comprising micelles exhibiting repulsion can form films comprising (or exhibiting) a plurality of domains (which may bedifferent domains (e.g., compositionally different, structurally different, or the like, or any combination thereof) or the like).

[0085] In various examples, the product of the Flory-Huggins interaction parameter or the Flory-Huggins interaction parameter and the degree of polymerization is about zero and / or where at least a portion, substantially all, or all of a plurality of first micelles and at least a portion, substantially all, or all of a plurality of second micelles do not exhibit attraction or repulsion (e.g., are neutral in their attraction) at least a portion, substantially all, or all of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of the core blocks of a second multiblock copolymer core is / are isotropically or statistically distributed in at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer and / or least a portion, substantially all, or all of the functional group(s) of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of the functional group(s) of the core blocks of a second multiblock copolymer core is / are isotropically or statistically disposed on at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer. In various examples, the product of the Flory-Huggins interaction parameter or the Flory-Huggins interaction parameter and the degree of polymerization is greater than about zero and / or where at least a portion, substantially all, or all of a plurality of first micelles and at least a portion, substantially all, or all of a plurality of second micelles exhibit repulsion to other micelles in the deposition solutions (such as, for example, ionic repulsion, van der Walls interactions, or the like, or any combination thereof) at least a portion, substantially all, or all of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of the core blocks of a second multiblock copolymer core is / are disposed in a plurality of distinct domains forming at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer and / or at least a portion, substantially all, or all of the functional group(s) of the core blocks of a second multiblock copolymer core is / are disposed in a plurality of distinct domains on at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer. In various examples, the product of the Flory- Huggins interaction parameter or the Flory-Huggins interaction parameter and the degree of polymerization is less than about zero and / or where at least a portion, substantially all, or all of a plurality of first micelles and at least a portion, substantially all, or all of a plurality of second micelles exhibit attraction to other micelles in the deposition solution (such as, for example, ionic attraction, dipolar attraction, hydrogen bonding, or the like, or any combination thereof) at least a portion, substantially all, or all of the functional group(s) ofthe core blocks of a second multiblock copolymer core is / are form(s) a three-dimensional pore lattice in at least a portion, substantially all, or all of the pore wall or pore wall surface(s) of the isoporous region or layer and / or at least a portion, substantially all, or all of the functional group(s) of the core blocks of a second multiblock copolymer core is / are disposed in a three-dimensional lattice on at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer.

[0086] In various examples, a corona-forming block forms at least a portion of or all of the corona of a micelle comprising (or formed from) a multiblock copolymer, where at least a portion of or all of the corona is disposed on at least a portion of or all of the core of the micelle. In various examples, a corona-forming block forms the matrix (e.g., non-pore structure) of the film. In various examples, a corona-forming block is a high Tg block. In various examples, in the case of a multiblock copolymer comprising three or more blocks, a corona-forming block is an interior block (or non-terminal block) of the multiblock copolymer.

[0087] In various examples, a core-forming block forms at least a portion of or all of the core of a micelle comprising (or formed from) a multiblock copolymer. In various examples, a core-forming block is disposed on at least a portion of or all of a pore surface of the film and / or a core-forming block is soluble in the phase separation solvent system. In various examples, in the case of a multiblock copolymer comprising two or more blocks or three or more blocks, a core-forming block may be a terminal block of the multiblock copolymer.

[0088] Various deposition solutions can be used. In various examples, a deposition solution comprises at least a first multiblock copolymer comprising a first core-forming block and a first corona-forming block and a second multiblock copolymer comprising a second core-forming block and a second corona-forming block, where the first corona-forming block and second corona-forming block are distinct (e.g., compositionally and / or structurally distinct). In various examples, a deposition solution comprises at least a first multiblock copolymer comprising a first core-forming block and a first corona-forming block and a second multiblock copolymer comprising a second core-forming block and a second corona- forming block, where the first corona-forming block and second corona-forming block are distinct (e.g., compositionally and / or structurally distinct) and the first core-forming block and second core-forming block are distinct (e.g., compositionally and / or structurally distinct).

[0089] In various examples, a core-forming block comprises one or more functional group(s), which may also be referred to herein as functionalizing group(s). A functional group may be a native functional group (a native functional group of the polymer repeat unitsof the core-forming block) or a functionalized group (a functional group formed by reaction of a native functional group of a monomer used to produce the polymer or the polymer).

[0090] In various examples, functionalized groups are distributed (e.g., selectively distributed) vertically and / or horizontally (e.g., relative to the largest exterior planar surface or the like) within the film. In various examples, the functionalized groups are isotropically or statistically distributed, locally segregated (such as, for example, in one or more domain(s), or independently or in the aggregate, form a 3-D lattice or lattices in at least a portion (e.g., one or more domain(s) or the like), substantially all, or all of an isoporous region or layer.

[0091] In various examples, a functional group is charged (e.g., positively charged, negatively charged, or has a net neutral charge). In various examples, the charge is a static charge, a coulombic charge, or the like. In various examples, the charge is pH dependent. Non-limiting examples of functionalized groups include peptides (natural or synthetic), cyclic peptides, ligands comprising a radio label (e.g.,124I,131I,225Ac,177Lu, or the like, or any combination thereof), antibodies, antibody fragments, DNA, RNA, simple sugars, oligosaccharides, small molecules, drug molecules (e.g. small molecule inhibitors, toxic drugs), and functionalized groups comprising a reactive group (e.g., a reactive group that can be conjugated to a molecule, such as, for example, a molecule comprising a functionalized group).

[0092] A functionalized group can be formed by methods / reactions known in the art. In various examples, monomer comprises one or more functionalized group(s) prior to formation of a multiblock copolymer comprising the polymerized monomer. In various examples, one or more functionalized group(s) is / are formed by reaction of a multiblock copolymer prior to film formation. In various examples, a functionalized group is (or all of the functionalized groups are) not formed after a film is formed (e.g., not formed by post-film formation functionalization).

[0093] A deposition solution can comprise various multiblock copolymers. In various examples, a deposition comprises two more multiblock copolymers. In various examples, a deposition solution comprises 2–10 (2, 3, 4, 5, 6, 7, 8, 9, or 10) multiblock copolymers. In various examples, two or more or all the multiblock copolymers are different (e.g., compositionally different (e.g., one or more different molecular weight block(s) or the like), structurally different (e.g., one or more structurally different block(s)), or both). In various examples, two or more or all of the multiblock copolymers are distinct (e.g., compositionally and / or structurally distinct, such as, for example, chemically-distinct (e.g., at least one of the blocks in a chemically distinct multiblock copolymer comprises a different chemicalcomposition and / or molecular weight than those in other chemically distinct block copolymer(s))). In various examples, the multiblock copolymers are independently chosen from diblock copolymers, triblock terpolymers, tetrablock copolymers, pentablock copolymers, and the like, and any combination thereof, and any combination thereof. Non- limiting examples of multiblock copolymers are described herein.

[0094] In various examples, a multiblock copolymer comprises (or has) a structure of form A-B or A-B-A or A-B-C, where A or B or C is polystyrene, poly-4-vinylpyridine, poly- 2-vinylpyridine, polybutadiene, polyisoprene, poly(ethylene-stat-butylene ), poly(ethylene- alt-propylene), polysiloxane, polyalkylenoxide, poly-ε-caprolactone, polylactide, polyalkylmethacrylate, polymethacrylic acid, polyalkylacrylate, polyacrylic acid, polyhydroxyethylmethacrylate, polyacrylamide, poly-N-alkylacrylamide, polyethylene oxide, polydimethylaminoethylmethacrylate, poly-tert-butylstyrene, or polyhydroxystyrene. In various examples, the multiblock copolymer is a triblock terpolymer comprising (or having) a structure of the form A – B – C, or A – C – B, or other variable arrangements or containing blocks of different chemical composition. In various examples, additional structures are higher order multi-block copolymer systems of the form A – B – C – B, or A – B – C – D, or A – B – C – B – A, or A – B – C – D – E, or other variable arrangements of these higher order systems.

[0095] Multiblock copolymers can be synthesized by methods known in the art. In various examples, a multiblock copolymer or copolymers is / are synthesized using anionic polymerization, atom transfer radical polymerization (ATRP), or other suitable polymerization techniques. Suitable multiblock copolymers can also be obtained commercially.

[0096] Non-limiting examples of suitable diblock copolymers include poly(styrene)-b- poly((4-vinyl)pyridine), poly(styrene)-b-poly((2-vinyl) pyridine), poly(styrene)-b- poly(ethylene oxide), poly(styrene)-b-poly(methyl methacrylate), poly(styrene)-b- poly(acrylic acid), poly(styrene)-b-poly(dimethylethyl amino ethyl methacrylate), poly(styrene)-b-poly(hydroxystyrene), poly(α-methyl styrene)-b-poly((4-vinyl)pyridine), poly(α-methyl styrene)-b-poly((2-vinyl) pyridine), poly(α-methyl styrene)-b-poly(ethylene oxide), poly(α-methyl styrene)-b-poly(methyl methacrylate), poly(α-methyl styrene)-b- poly(acrylic acid), poly(α-methyl styrene)-b-poly(dimethylethyl amino ethyl methacrylate), poly(α-methyl styrene)-b-poly(hydroxystyrene), poly(isoprene)-b-poly((4-vinyl)pyridine), poly(isoprene)-b-poly((2-vinyl) pyridine), poly(isoprene)-b-poly(ethylene oxide), poly(isoprene)-b-poly(methyl methacrylate), poly(isoprene)-b-poly(acrylic acid),poly(isoprene)-b-poly(dimethylethyl amino ethyl methacrylate), poly(isoprene)-b- poly(hydroxystyrene), poly(butadiene)-b-poly((4-vinyl)pyridine), poly(butadiene)-b-poly((2- vinyl) pyridine), poly(butadiene)-b-poly(ethylene oxide), poly(butadiene)-b-poly(methyl methacrylate), poly(butadiene)-b-poly(acrylic acid), poly(butadiene)-b-poly(dimethylethyl amino ethyl methacrylate), poly(butadiene)-b-poly(hydroxystyrene), poly(4-tert- butylstyrene)-b-poly((4-vinyl)pyridine), poly(4-tert-butylstyrene)-b-poly((2-vinyl) pyridine), poly(4-tert-butylstyrene)-b-poly(ethylene oxide), poly(4-tert-butylstyrene)-b-poly(methyl methacrylate), poly(4-tert-butylstyrene)-b-poly(acrylic acid), poly(4-tert-butylstyrene)-b- poly(dimethylethyl amino ethyl methacrylate), poly(hydroxystyrene)-b-poly((4- vinyl)pyridine), poly(hydroxystyrene)-b-poly((2-vinyl) pyridine), poly(hydroxystyrene)-b- poly(ethylene oxide), structural analogs thereof, and the like, and any combination thereof.

[0097] Non-limiting examples of suitable triblock copolymers include poly(isoprene-b- styrene-b-4-vinylpyridine), poly(isoprene)-b-poly(styrene)-b-poly((4-vinyl)pyridine), poly(isoprene)-b-poly(styrene)-b-poly((2-vinyl) pyridine), poly(isoprene)-b-poly(styrene)-b- poly(ethylene oxide), poly(isoprene)-b-poly(styrene)-b-poly(methyl methacrylate), poly(isoprene)-b-poly(styrene)-b-poly(acrylic acid), poly(isoprene)-b-poly(styrene)-b- poly(dimethylethyl amino ethyl methacrylate), poly(isoprene)-b-poly(styrene)-b- poly(hydroxystyrene), poly(isoprene)-b-poly(α-methyl styrene)-b-poly((4-vinyl)pyridine), poly(isoprene)-b-poly(α-methyl styrene)-b-poly((2-vinyl) pyridine), poly(isoprene)-b-poly(α- methyl styrene)-b-poly(ethylene oxide), poly(isoprene)-b-poly(α-methyl styrene)-b- poly(methyl methacrylate), poly(isoprene)-b-poly(α-methyl styrene)-b-poly(acrylic acid), poly(isoprene)-b-poly(α-methyl styrene)-b-poly(dimethylethyl amino ethyl methacrylate), poly(butadiene)-b-poly(styrene)-b-poly((4-vinyl)pyridine), poly(butadiene)-b-poly(styrene)- b-poly((2-vinyl) pyridine), poly(butadiene)-b-poly(styrene)-b-poly(ethylene oxide), poly(butadiene)-b-poly(styrene)-b-poly(methyl methacrylate), poly(butadiene)-b- poly(styrene)-b-poly(acrylic acid), poly(butadiene)-b-poly(styrene)-b-poly(dimethylethyl amino ethyl methacrylate), poly(butadiene)-b-poly(styrene)-b-poly(hydroxystyrene), poly(butadiene)-b-poly(α-methyl styrene)-b-poly((4-vinyl)pyridine), poly(butadiene)-b- poly(α-methyl styrene)-b-poly((2-vinyl) pyridine), poly(butadiene)-b-poly(α-methyl styrene)- b-poly(ethylene oxide), poly(butadiene)-b-poly(α-methyl styrene)-b-poly(methyl methacrylate), poly(butadiene)-b-poly(α-methyl styrene)-b-poly(acrylic acid), poly(butadiene)-b-poly(α-methyl styrene)-b-poly(dimethylethyl amino ethyl methacrylate), poly(butadiene)-b-poly(styrene)-b-poly(hydroxystyrene), poly(isoprene-b-4-tertbutylstyrene- b-4-vinylpyridine), poly(isoprene)-b-poly(4-tertbutylstyrene)-b-poly((4-vinyl)pyridine),poly(isoprene)-b-poly(4-tertbutylstyrene)-b-poly((2-vinyl) pyridine), poly(isoprene)-b- poly(4-tertbutylstyrene)-b-poly(ethylene oxide), poly(isoprene)-b-poly(4-tertbutylstyrene)-b- poly(methyl methacrylate), poly(isoprene)-b-poly(4-tertbutylstyrene)-b-poly(acrylic acid), poly(isoprene)-b-poly(4-tertbutylstyrene)-b-poly(dimethylethyl amino ethyl methacrylate), poly(isoprene)-b-poly(4-tertbutylstyrene)-b-poly(hydroxystyrene), structural analogs thereof, and the like, and any combination thereof.

[0098] In various examples, a multiblock copolymer comprises at least one hydrogen- bonding block. A hydrogen-bonding block can self-assemble with another structurally distinct polymer block of the multiblock copolymer (e.g., a hydrophobic block or the like). The hydrogen-bonding block comprises an acceptor group or donor group that can participate in intramolecular hydrogen bonding. The hydrogen-bonding block can be a hydrophilic block. Non-limiting examples of suitable hydrogen-bonding blocks include poly((4- vinyl)pyridine), poly((2-vinyl) pyridine), poly(ethylene oxide), poly(methacrylatexs), such as, for example, poly(methacrylate), poly(methyl methacrylate), poly(dimethylethyl amino ethyl methacrylate), and the like), poly(acrylic acid), poly(hydroxystyrene), and the like, and any combination thereof. In various examples, the hydrophilic block(s) is / are poly((4- vinyl)pyridine) or the like. In these examples, the multiblock copolymer has additional blocks that are hydrophobic blocks. In various examples, the hydrophobic blocks form the matrix of the film. For example, a multiblock copolymer has one or two hydrophobic block(s) in addition to the hydrogen-bonding block(s). Examples of suitable hydrophobic blocks include, but are not limited to, poly(styrenes) (such as, for example, poly(styrene), poly(alpha-methyl styrene), and the like), polyethylene, polypropylene, polyvinyl chloride, poly-tert- butylstyrene, polytetrafluoroethylene, and the like, and any combination thereof.

[0099] In various examples, at least one of the additional hydrophobic blocks is a low glass transition temperature (Tg) block. By low Tg block it is meant that the block has a Tg of about 25° C. or less. The multiblock copolymer can have multiple low Tg blocks. Examples of suitable low Tg blocks include, but are not limited to, poly(isoprene), poly(butadiene), poly(butylene), poly(isobutylene), and the like, and any combination thereof. In various examples, the multiblock copolymer comprises a low Tg polymer block, a poly(styrene) block, and a poly((4-vinyl) pyridine) block.

[0100] Individual polymer blocks of a multiblock copolymer can have a various molecular weight range. In various examples, individual blocks have a number averaged molecular weight (Mn) of about 1 x 103to 1 x about 106g / mol, including all values to the 10 g / mol and ranges therebetween. The total molar mass of the multi-block copolymer is suchthat the multiblock copolymer undergoes self-assembly (i.e., microphase separation). It is desirable that defect-free surfaces are formed upon meso- and macro-porous structure formation. In various examples, the total molar mass of the multiblock copolymer is from about 1×103to 1×106g / mol, including all values to the 10 g / mol and ranges therebetween (e.g., 5 x 103to about 5 x 105g / mol). Multiblock copolymers can have a range of polydispersities (Mw / Mn). In various examples, a multiblock copolymer has a polydispersity index (PDI) of about 1.0 to about 2.0, including all values to the 0.1 and ranges therebetween. It is desirable that a multiblock copolymer have a PDI of about 1 to about 1.4.

[0101] A deposition solution may comprise one or more homopolymer(s). In various examples, a deposition solution further comprises one or more homopolymer(s) and at least two of the two or more solvent(s) have different surface energy and / or surface tension and / or the lower surface energy and / or surface tension. In various examples, and at least two of the two or more solvent(s) preferentially solubilizes the one or more multiblock copolymer(s) relative to the homopolymer(s). In the case where the deposition solution comprises one or more homopolymer(s), removing at least a portion of the solvent system from the film (e.g., first film or the like) results in at least a portion, substantially all, or all of at least one of the multiblock copolymer(s) migrates to an exterior surface of the first film (e.g., a second film is formed). In various examples, where the deposition solution comprises one or more homopolymer(s) and the method comprises: providing a first polymer solution comprising one or more first polymer(s) (which may be multiblock copolymer(s) and / or homopolymer(s)) and one or more first solvent(s), where the first polymer solution optionally further comprises a metal salt; providing a second polymer solution comprising one or more second polymer(s) (which may be multiblock copolymer(s) and / or homopolymer(s)) and one or more second solvent(s), where the second polymer solution optionally further comprises a salt, where the solvent of the polymer solution comprising the multiblock copolymer has a lower surface energy and / or surface tension than the solvent of the polymer solution comprising the homopolymer(s) and / or the solvent of the polymer solution comprising the multiblock copolymer preferentially solubilizes the multiblock copolymer(s) relative to the homopolymer(s).

[0102] Various homopolymers may be used to make a film. In various examples, the homopolymer(s) is / are chosen from polysulfanes, polypropylenes, polylactic acids, polysulfones, poly(ether sulfone)s, poly(phenyl sulfone)s, cellulose acetates, poly(acrylonitrile)s, poly(amide)s, structural analogs thereof, and the like, and any combination thereof.

[0103] A deposition solution can comprise various amounts of one or more multiblock copolymer(s) and optionally, homopolymer(s). In various examples, the concentration of multiblock copolymer(s) and homopolymer(s), if present, in the deposition solution is about 5 to about 30 percent by weight (based on the total weight of the deposition solution or the total weight of the solvent(s), multiblock copolymer(s) and homopolymer(s), if present), including all 0.1 percent by weight values and ranges therebetween. In various examples, multiblock copolymer(s) comprise about 10 percent by weight or less of the deposition solution (based on the total weight of multiblock copolymer(s)). In various examples, homopolymer(s), if present, in the deposition solution) and / or the one or more homopolymer(s) comprise about 90 percent by weight or more of the deposition solution (based on the total weight of multiblock copolymer(s) and homopolymer(s), if present, in the deposition solution).

[0104] A deposition solution may comprise various solvent(s). A deposition solution may comprise two or more solvents (e.g., a solvent system). In various examples, the solvent(s) (e.g., the solvent(s) of a solvent system) is / are chosen from 1,4-dioxane, tetrahydrofuran, methanol, ethanol, toluene, chloroform, dimethylformamide, acetone, dimethylsulfoxide, N- methyl pyrrolidone, structural analogs thereof, and the like, and any combination thereof.

[0105] In various examples, two or more multiblock copolymers comprise (or are) polyisoprene-b-polystyrene-b-poly(4-vinylpyridine) (ISV) or one or more structural analog(s) thereof and / or polystyrene-b-poly(4-vinylpyridine)(P4VP)-b-propylene sulfide(PPS) (SVPS) or one or more structural analog(s) thereof and / or poly(isoprene-b-styrene-b-2-vinylpyridine- b-ethylene oxide) (ISVO) or one or more structural analog(s) thereof, and, optionally, the solvents comprise (or are) 1,4-dioxane (DOX) and tetrahydrofuran (THF).

[0106] A deposition solution may further comprise one or more additives. In various examples, a deposition solution further comprises one or more additive(s). Additive homopolymer(s) and / or small molecule may preferentially associate with one or more blocks of a multiblock copolymer and locate in the vicinity of that block. In non-limiting, illustrative examples, a poly(phenylene oxide) mixes with a poly(styrene) block of a multiblock copolymer or a poly(butadiene) mixes with a poly(isoprene) block of a multiblock copolymer. Non-limiting examples of additives(s) include additive homopolymer(s), small molecule(s), and the like, and combinations thereof. Non-limiting examples of suitable additive homopolymers and small molecules are provided herein.

[0107] In various examples, a deposition solution further comprises one or more small molecule(s) and the film further comprises the one or more small molecule(s). As used herein, unless otherwise stated, small molecule means a molecule having a molar mass ofabout1000 g / mol or less. In various examples, a small molecule is an organic molecule or the like. A film can comprise a variety of small molecules. In various examples, a small molecule can hydrogen bond or the like to at least one block of the multiblock copolymer. In various examples, a small molecule comprises hydrogen bond donor(s) and / or hydrogen bond acceptor(s). Nonlimiting examples of suitable small molecules include pentadecyl phenol, dodecyl phenol, 2- 4’-(hydroxybenzeneazo)benzoic acid (HABA), 1,8-naphthalene- dimethanol, 3-hydroxy-2- naphthoic acid, 6-hydroxy-2-naphthoic acid, and the like, and any combination thereof. A range of ratios of multiblock copolymer(s) to small molecule(s) can be used. In various examples, the molar ratio of multiblock copolymer(s) to small molecule can be from about 1 : 1 to about 1 : 1000, including all integer ratios therebetween.

[0108] In various examples, a deposition solution further comprises one or more inorganic nanoparticle(s) and a film further comprises the one or more inorganic nanoparticle(s). Non-limiting examples of inorganic nanoparticles include silica nanoparticles, silicon oxide nanoparticles, metal nanoparticles, and the like. In various examples, a deposition solution comprises a mixture of two or more different (e.g., compositionally different, structurally different, or the like, or any combination thereof) inorganic nanoparticles. In various examples, at least a portion or all of the inorganic nanoparticles are metal nanoparticle(s). In various examples, metal nanoparticles can complex (e.g., through weak intramolecular forces or the like) with the multiblock copolymer of the film surface. In various examples, the inorganic nanoparticle(s) (such as, for example, metal nanoparticles or the like) independently comprise a size (e.g., a longest linear cross- sectional dimension, such as, for example, a diameter or the like) or comprise an average size of about 1 to about 200 nm, including all values to the nanometer and ranges therebetween. Examples of suitable metals for the metal nanoparticles include, but are not limited to, gold, silver, platinum, palladium, cobalt, copper, nickel, iron, zinc, chromium, ruthenium, titanium, zirconium, molybdenum, aluminum, cadmium, and the like. The metal nanoparticles can be mixtures of different metal nanoparticles.

[0109] In various examples, an additive is a porogen(s) (e.g., polyvinylpyrrolidone, or the like, is added as a porogen). It may be desirable that the additive(s) (e.g., porogen(s) and the like) is / are water soluble. Any water-soluble polymers miscible with the homopolymer(s) (e.g., PSf (polysulfone)and the like) may be additives. Without intending to be bound by any particular theory, it is considered after the initial film formation, when the film is contacted with the phase separation solvent system (e.g., water or the like), the porogen (e.g., polyvinylpyrrolidone or the like), if present, dissolves in the phase separation solvent system(e.g., water), resulting in a more porous substructure formed by the remaining homopolymer(s) (e.g., PSf or the like) than would have resulted if the porogen(s) were not used. In various examples, the additive(s) is / are polyvinylpyrrolidone, PEG (poly(ethylene glycol), poly(meth)acrylic acid, or the like, or any combination thereof. In the case where a deposition comprises one or more porogen(s), the film may comprise at least a portion of the porogen(s) or may not comprise the porogen(s).

[0110] A deposition solution may comprise various amounts of additive(s). In various examples, the amount of additive(s) in the deposition solution is (independently or in the aggregate) about 1 percent by weight to about 40 percent by weight of the homopolymer(s) (e.g., PSf and the like), including all 0.1 percent by weight values and ranges therebetween.

[0111] A method may further comprise depositing one or more inorganic material(s) on at least a portion of a surface (e.g., an exterior surface or the like) of a film. Combinations of two or more different (e.g., compositionally different, structurally different, or both) inorganic materials may be deposited. Non-limiting examples of inorganic materials include metal nanoparticles, where the metal nanoparticles comprise one or more metal(s), and the like, and any combination thereof. In various examples, an inorganic material is in the form of nanoparticles. In various examples, the nanoparticles are about 1 to about 200 nm, including all values to the nanometer and ranges therebetween, in size or average size (e.g., a longest linear cross-sectional dimension, such as, for example, a diameter or the like). Examples of suitable inorganic materials include, but are not limited to, metals, metal oxides (e.g., silver oxide and copper oxide) and semiconductors (e.g., semiconducting nanoparticles such as CdS nanoparticles). The inorganic materials can be deposited on the film by methods known in the art. In various examples, an inorganic material is deposited by electroless deposition or the like.

[0112] After the partial solvent removal step, a film (such as, for example, a second film or the like) is contacted with a phase inversion solvent system. A phase inversion solvent system can be a single solvent or a mixture of solvents. In various examples, a phase inversion solvent system comprise (or is) a non-solvent for the block copolymer(s) (e.g., at least one of the blocks of one or more of the block copolymer precipitates in the solvent system). It is desirable that a solvent system (e.g., of a deposition solution) is miscible with the non-solvent for the NIPS process. Non-limiting examples of suitable solvents for use in the NIPS process include, but are not limited to, water, methanol, ethanol, acetone, and the like, and combinations thereof. Without intending to be bound by any particular theory, it isconsidered that contacting a film with a non-solvent causes the multiblock copolymer(s) and homopolymer(s), if present, to precipitate.

[0113] A film (and / or portions thereof) can have various thicknesses. In various examples, an isoporous region or layer comprises a thickness of about 20 nm to about 500 nm, including all 0.1 nm values and ranges therebetween, a plurality of pores having a size of about 5 nm to about 100 nm, including all 0.1 nm values and ranges therebetween and / or a asymmetric pore region or layer has a thickness of about 5 microns to about 500 microns, including all 0.1 micron values and ranges therebetween, and pores having a size of about 10 nm to about 50 microns, including all 0.1 nm values and ranges therebetween.

[0114] A method may form a free-standing film. In various examples, where a film (e.g., a first film or a first film and second film, or the like) is formed on a substrate, a method further comprising separating the film from the substrate to form a free-standing film.

[0115] In an aspect, the present disclosure provides compositions. In various examples, a composition is suitable for use in a deposition solution (e.g., a dope or the like), such as, for example, use as a deposition solution in a method for the present disclosure. In various examples, a composition is a deposition solution in or of method of the present disclosure. Non-limiting examples of compositions are provided herein.

[0116] In various examples, a composition (which may be (or be suitable for use as) a deposition solution (e.g., a dope or the like)) comprises: two or more multiblock copolymers (such as, for example, where each multiblock copolymer comprises one or more corona- forming block(s) and one or more core-forming block(s), and at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s) or the like). In various examples, a composition further comprises one or more homopolymer(s), two or more solvents, one or more metal salt(s), one or more inorganic nanoparticle(s), or one or more additive(s), or the like, or any combination thereof.

[0117] Non-limiting examples of multiblock copolymers, homopolymers, solvents, metal salts, and additives are provided herein. In various examples, a composition comprises two or more multiblock copolymers, optionally, homopolymer(s), optionally, solvents, optionally, metal salt(s), and optionally, additive(s) provided herein.

[0118] In various examples, a composition comprises one or more or all of the components of a deposition solution of the present disclosure. In various examples, the amount multiblock copolymer(s) homopolymer(s), if present, solvent(s), metal salts, ifpresent, and additive(s), if present, are as described herein (e.g., in the description of the deposition solution).

[0119] In an aspect, the present disclosure provides films. In various examples, a film is a porous functionalized multiblock copolymer film, an isoporous film, which may be an isoporous graded film, an asymmetric film (e.g., comprising asymmetric pore structure along an axis substantially normal or normal to the film surface). In various examples, a film is formed by a method of the present disclosure. Non-limiting examples of films are provided herein.

[0120] In various examples, a film comprises an isoporous surface region or layer and an asymmetric pore region or layer. In various examples, the isoporous region or layer is disposed on at least a portion of an exterior surface of the asymmetric pore region or layer. In various examples, the isoporous layer comprises two or more multiblock copolymers (such as, for example, distinct multiblock copolymers (e.g., chemically-distinct multiblock copolymers or the like)). In various examples, an isoporous surface region or layer comprises two or more multiblock copolymers, where each multiblock copolymer comprises one or more corona-forming block(s) and one or more core-forming block(s), and at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s). In various examples, at least a portion, substantially all, or all of the core blocks (such as, for example, the core blocks of a first multiblock copolymer and / or a second multiblock copolymer or the like) are, independently or in the aggregate, are isotropically or statistically distributed in (or form) at least a portion, substantially all, or all of the pore wall surfaces of, or are disposed in a plurality of distinct domains of (or form a plurality of domains of) at least a portion, substantially all, or all of the pore wall surface(s) of, or are disposed in (or form) a three-dimensional lattice, independently or in the aggregate, at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer. In various examples, at least a portion, substantially all, or all of the functional groups of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of a functional group of the core blocks of a second multiblock copolymer core are disposed on at least a portion, substantially all, or all of the pore wall surfaces of the isoporous region or layer.

[0121] In various examples, film is heterogeneous (e.g., a film comprises a plurality of domains, wherein at least two domains are different (e.g., structurally different,compositionally different, or the like, or any combination thereof)). In various examples, the isoporous surface region or layer) and asymmetric pore region or layer are not independently (or separately) formed and / or not laminated together to form the film.

[0122] Without intending to be bound by any particular theory, it is considered the amount of multiblock copolymer(s), homopolymer(s), if present, solvent(s), metal salts, if present, and additive(s), if present, in a film is correlated to (e.g., substantially the same as or the same as) the amount of the corresponding component in the deposition solution. In various examples, the amount of multiblock copolymer(s), homopolymer(s), if present, solvent(s), metal salts, if present, and additive(s), if present, in a film is correlated to (e.g., substantially the same as or the same as or the like) the amount of the corresponding component in the deposition solution.

[0123] In various examples, a film comprises a plurality of domains (which may form a region or layer of the film), two or more of which may be different. In various examples, at least two domains are different in terms of average pore size (e.g., structurally different, compositionally different, or the like, or any combination thereof). In various examples, a domain comprises about 2 to about 100 pores, including all integer number of pore and ranges therebetween (e.g., about 5 to about 50 pores or about 10 to about 30 pores). In various examples, different domains at least partially lie along a line normal to the largest exterior film surface and / or the different domains at least partially lie along a line horizontal to the largest film surface. In various examples, a film comprises a plurality of domains, where at least two domains are different in terms of: symmetry (e.g., 4-fold symmetry vs.6- fold symmetry or the like); pore surface chemistry; or the like; or any combination thereof. In various examples, a film comprises a plurality of domains, where the domains are ordered, disordered, or the like, or any combination thereof.

[0124] A film (or a portion thereof, such as, for example, an isoporous region, an asymmetric region, or the like) can have various dimensions. In various examples, an isoporous region or layer comprises a thickness of about 20 nm (nm = nanometer(s)) to about 500 nm, including all 0.1 nm values and ranges therebetween, a plurality of pores having a size of about 5 nm to about 100 nm, including all 0.1 nm values and ranges therebetween. In various examples, an asymmetric pore region or layer comprises a thickness of about 5 microns (micron(s) = micrometer(s)) to about 500 microns, including all 0.1 micron values and ranges therebetween, and pores having a size of about 10 nm to about 50 microns, including all 0.1 nm values and ranges therebetween.

[0125] A film can have various pore sizes, pore density, pore size distribution, or the like, or any combination thereof. In various examples, an isoporous region or layer comprises a plurality of the pores. In various examples, the pores of an isoporous region or layer, independently, comprise a size (e.g., a longest cross-sectional dimension, such, as for example, a diameter or the like) of about 5 nm to about 100 nm, including all 0.1 nm values and ranges therebetween, and / or a pore density of at least about 1 x 1014pores / m2. In various examples, an isoporous region or layer comprises a pore size distribution of about less than 3, where the pore size distribution is the ratio of the maximum pore diameter (dmax) to the minimum pore diameter (dmin). In various examples, an asymmetric region or layer comprises a plurality of pores. In various examples, the pores, independently, comprise a size (e.g., a longest cross-sectional dimension, such, as for example, a diameter or the like) of about 10 nm to about 100 microns in size (e.g., diameter), including all values to the nm and ranges therebetween.

[0126] A film can comprise various isoporous regions or layers. An isoporous region or layer may be referred to as a top surface region or layer or an isoporous surface (e.g., top surface) region or layer. Without intending to be bound by any particular theory, it is considered that the morphology of the isoporous region or layer is, in part, a result of the self- assembly of the multiblock copolymer(s) The morphology of this layer is dependent on the casting conditions (e.g., flow rate of environment around the film, water (humidity) / solvent concentration in environment around the film, evaporation time, casting speed, gate height) as well as the composition of the casting solution (e.g., polymer molar mass, chemistry, concentration, casting solvent or mixture of solvents).

[0127] In various examples, an isoporous region or layer comprises two or more cohorts of isoporous pores. In various examples, an isoporous region or layer comprises two or more cohorts of isoporous pores, where each cohort of isoporous pores is formed from a specific multiblock copolymer or multiblock copolymer(s). In non-limiting, illustrative examples, an isoporous region or layer comprises two or more cohorts of isoporous pores, where a first cohort of isoporous pores corresponds to (or is formed from) a first multiblock copolymer and a second cohort of isoporous pores corresponds to (or is formed from) a second multiblock copolymer, where the first multiblock copolymer is distinct (e.g., chemically- distinct or the like) from the second multiblock copolymer.

[0128] A pore of an isoporous region or layer or an array of pores of an isoporous region or layer, which may be disposed in a particular domain or domains) can have various morphology. In various examples, at least a portion, substantially all, or all the pores of anisoporous region or layer or a portion thereof (such as, for example, a domain thereof or the like) have substantially the same or the same morphology. In various examples, a pore or the pores of an array pores, which may be disposed in a particular domain or domains is / are, independently, a hexagonal pore, a cubic pore, or the like.

[0129] In various examples, a film (or a portion thereof, such as, for example, an isoporous region, an asymmetric region, or the like) comprises pores comprising pore walls. In various examples, at least a portion, substantially all or all the pore walls are functionalized with one or more pore functionalizing group(s) or the like. By “functionalized” it is meant that a film comprises a plurality of functionalizing groups (such as, for example, pore functionalizing groups or the like), functional groups or the like, which are introduced (e.g., areally and / or spatially introduced) to the film as a result of the film forming process as described herein. In various examples, functionalizing groups are functional groups of a core block of a multiblock copolymer. In non-limiting, illustrative examples, a functionalizing group is a 4-pyridine group (e.g., formed from a poly(4- vinylpyridine) block, a thiol group (e.g., formed from a propylene sulfide block or the like), or the like.

[0130] In various examples, individual pore functionalizing groups are covalently attached to the pore wall via at least one covalent bond to the pore wall. In various examples, functional groups, which may be the same functional groups or at least two or more of the functional groups are different (e.g., structurally different, compositionally different, or the like, or any combination thereof), are distributed (e.g., selectively distributed) vertically and / or horizontally (relative to the largest exterior planar surface) within the film. In various examples, a functionalized group is (or all of the functionalized groups are) not formed after a film is formed (e.g., not formed by a post-film formation process or the like).

[0131] In various examples, a functionalizing group is a hydrophobic group, a hydrophilic group or the like and / or a neutral or an ionic group (such as, for example, a positively-charged group, a negatively-charged group, a zwitterionic group, or the like), where the charge may be dependent on the film environment (e.g., pH or the like). In various examples, a functionalizing group is a clickable group (i.e., a group that can participate in a click reaction. Examples of click reactions and suitable clickable groups are known in the art. Non-limiting examples of pore functionalizing groups include halogen groups (-F, -Cl, -Br, and -I), aliphatic groups (e.g., alkyl groups, alkenyl groups, alkynyl groups, and the like), halogenated aliphatic groups (e.g., trifluoromethyl group and the like), aryl groups, halogenated aryl groups, hydroxyl group, amine groups, quaternary ammonium groups, nitrogroup, cyano groups, isocyano groups, azide group, silane groups (e.g., alkyl silane groups, aryl silane groups, alkyl / aryl silane groups, or the like), alkoxide groups, alcohol groups, ether groups (e.g., activated ester groups or the like), ketone groups, carboxylate groups, carboxylic acid groups, ester groups, amide groups, thiol groups, thioether groups, thioester groups, and the like, and any combination thereof.

[0132] In various examples, a functionalizing group is a clickable group (i.e., a group that can participate in a click reaction. Examples of click reactions and suitable clickable groups are known in the art. In the case where a film comprises clickable functionalizing group, the film may be functionalized (e.g., the pores of the film may be functionalized) after film formation. In various examples, a method of making a film comprising pores with reactive groups (e.g., functionalizing groups resulting from film formation) further comprises contacting the film with a functionalizing molecule (such as, for example, a clickable molecule or a functional group or moiety that can react with a reactive group in a pore), where the functionalizing molecule (or a portion thereof) is covalently bonded to the pore surface (e.g., through a reactive group (e.g., functionalizing groups resulting from film formation) or a portion thereof, or the like).

[0133] A film can have various asymmetric pore regions or layers. In various examples, an asymmetric region or layer is an asymmetric substructure or the like. In various examples, an isoporous region or layer is disposed on an asymmetric layer (which may be referred to as a support region or layer). In various examples, e.g., where the film is formed on a substrate, an asymmetric layer is disposed between the isoporous region or layer and the substrate. An asymmetric layer is a supporting sub-structure layer.

[0134] In various examples, an asymmetric layer (e.g., an asymmetric substructure or the like) comprises two or more multiblock copolymers (such as, for example, the same multiblock copolymers the isoporous region or layer comprises). In various examples, moving from the top of this region or layer (e.g., the part in contact with the isoporous region or layer) to the bottom of the layer (e.g., the free surface opposite the isoporous region or layer or surface in contact with a substrate), the pores increase in size. In various examples, an asymmetric layer has a “sponge-like” structure. In various examples, the pores, which may be graded, in a “sponge-like” structure comprise a diameter or equivalent circular support layer pore diameter of about 10 nm to about 5 microns, including all integer nm values and ranges therebetween, and / or have an aspect ratio of less than about 2. In various examples, an asymmetric layer has a “finger-like” structure. In various examples, the pores of a “finger- like” structure, which may be graded, comprise a diameter or equivalent circular supportlayer pore diameter of about 1 micron 1 micron to about 5 microns, including all integer nm values and ranges therebetween, and / or an aspect ratio of 2 or more. “Aspect ratio” as used herein means the ratio of length of the pore along the axis perpendicular to the surface-layer- support layer interface to the longest length of a cross-section of the pore parallel to the surface-layer-support layer interface. In various examples, at least a portion of the pores in the support layer have an aspect ratio of greater than about 3, greater than about 4, or greater than about 5. In various examples, a “finger-like” structure comprises a hierarchical structure (e.g., based on the bulk porosity and pore surface porosity).

[0135] In various examples, an asymmetric pore region comprises (or is) one or more homopolymer(s) or the like. In various examples, the asymmetric pore region or layer does not comprise (or have) about 10 percent by weight or more multiblock copolymer(s) (based on the total weight of multiblock copolymer(s) and homopolymer(s) in the asymmetric pore region or layer).

[0136] An asymmetric layer can have various pore configurations. In various examples, at least a portion, substantially all, or all the pores of an asymmetric pore region or layer increases in size along a dimension of the film moving away from a surface of the isoporous region or layer in contact with the asymmetric pore region or layer towards a surface of the asymmetric pore region or layer in contact with the substrate. In non-limiting illustrative examples, an asymmetric region or layer has pores having a size of about 10 nm at the top of the asymmetric region or layer (layer in contact with the isoporous surface layer) and the pores increase in size to 100 μm at the bottom of the support layer. An increase in pore size moving though the depth of the film (e.g., from the top of the support layer in contact with the surface layer to the surface of the film in contact with the substrate) provides an asymmetric structure. In various examples, an asymmetric layer is formed as a result of contacting (e.g., immersing) the film into a non-solvent bath (e.g., water in the SNIPS (such as, for example, a M2SNIPS process or the like). In various examples, a plurality of the pores in an asymmetric pore region or layer, which may be in a “finger-like” structure, lying on a plane about 20 microns from the surface layer and parallel to interface between the isoporous region or layer and asymmetric pore region or layer have a size in the plane of at least about 5 microns. In various examples, at least a portion of the pores in the asymmetric pore region or layer, which may be in a “finger-like” structure, comprise an aspect ratio of greater than about 2.

[0137] In various examples, a film further comprises inorganic nanoparticles disposed within and / or on the surface of the film (e.g., disposed on and / or in the isoporous region orlayer and / or the asymmetric region or layer). In various examples, inorganic nanoparticles are present in the polymer matrix of the film (e.g., throughout the isoporous region or layer and / or asymmetric region or layer) and / or on the surface of the film (e.g., including the pore surface(s)). Some portion of the inorganic nanoparticles may be present within the film (i.e., completely encapsulated by the polymer matrix of the film) and / or some portion of the nanoparticles may be present on the surface of the film. In various examples, inorganic nanoparticles are not present solely on a surface of the film. In various examples, a film comprises about 0.1 percent by weight to about 50 percent by weight, including all integer percent by weight values and ranges therebetween, inorganic nanoparticles. In various examples, a film comprises about 0.1 percent by weight to about 5 percent by weight or about 10 percent by weight to about 25 percent by weight inorganic nanoparticles.

[0138] In various examples, a film further comprises a plurality of metal nanoparticles. In various examples, metal nanoparticles are disposed on or in at least a portion of the film (e.g., disposed on and / or in the isoporous region or layer and / or the asymmetric region or layer). In various examples, metal nanoparticles can complex (e.g., through weak intramolecular forces or the like) with the multiblock copolymer of the film surface. In various examples, the nanoparticles comprise a size or average size (e.g., a longest linear cross-sectional dimension, such as, for example, a diameter or the like) of about 1 to about 200 nm, including all values to the nanometer and ranges therebetween. Examples of suitable metals for the metal nanoparticles include, but are not limited to, gold, silver, platinum, palladium, cobalt, copper, nickel, iron, zinc, chromium, ruthenium, titanium, zirconium, molybdenum, aluminum, cadmium, and the like. The metal nanoparticles can be mixtures of different metal nanoparticles. In various examples, metal nanoparticles are disposed on at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 99% of the surfaces of a film. In various examples, metal nanoparticles are disposed on 100% of the surface(s) of a film. Films with silver nanoparticles can exhibit antimicrobial behavior.

[0139] In various examples, a film further comprises one or more inorganic material(s) disposed on at least a portion of a surface (e.g., an exterior surface or the like) or surfaces (e.g, including pore surface(s)) of a film. A film may further comprise a combination of two or more different (e.g., compositionally different, structurally different, or both) inorganic materials. Non-limiting examples of inorganic materials include metal nanoparticles, where the metal nanoparticles comprise one or more metal(s), and the like, and any combination thereof. In various examples, an inorganic material is in the form of nanoparticles. In various examples, the nanoparticles are about 1 to about 200 nm, including all values to thenanometer and ranges therebetween, in size or average size (e.g., a longest linear cross- sectional dimension, such as, for example, a diameter or the like). Examples of suitable inorganic materials include, but are not limited to, metals, metal oxides (e.g., silver oxide and copper oxide) and semiconductors (e.g., semiconducting nanoparticles such as CdS nanoparticles). In various examples, an inorganic material and / or materials is / are disposed on at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 99% of the surface(s) of the film. In various examples, an inorganic material or materials is / are disposed on 100% of the surfaces of a film. The inorganic materials can be deposited on the film by methods known in the art. In various examples, an inorganic material is deposited by electroless deposition or the like.

[0140] The area of a film is not particularly limited. One of ordinary skill in the art will appreciate that based on relevant film making process knows in the art films with a broad range of area can produced using the instant methods. The films can be produced individually or in bulk (e.g., in a roll-to-roll process or the like). In various examples, a film comprises an area of square inches to square meters.

[0141] In various examples, a film is a free-standing film or the like. In various examples, a film is disposed on at least a portion, substantially all, or all of a surface or one or more or all the surfaces of a substrate, which may be a solid substrate or a porous substrate. In various examples, a substrate provides mechanical stability or the like. A wide range of substrate materials, sizes, and shapes can be used. Examples of suitable substrates include, but are not limited to, glass plates or rods, silicon, plastic (e.g., Teflon™) porous film supports (such as, for example, non-woven polyester, and the like), and the like, and any combination thereof. In various examples, a non-woven polyester on top of glass is used as a substrate.

[0142] In an aspect, the present disclosure provides uses of films of the present disclosure. In various examples, a device comprises one or more film(s) of the present disclosure. In various examples, where the device comprises two or more films, the films are the same (e.g., the films comprise substantially the same or the same structure and / or composition). In various examples, a device is configured for one or more uses described herein. Non-limiting examples of uses of films are provided herein.

[0143] A film or films can be used in filtration applications, separation applications, catalytic applications, or the like, or any combination thereof. In various examples, a device is configured for one or more use(s) described herein.

[0144] In various examples, one or more film(s) are used for a filtration application (such as, for example, a biomolecule filtration or the like). In various examples, a device is afiltration device or the like. In various examples, a filtration device is configured for (e.g., the film(s) is / are configured for) filtration (e.g., where the film(s) are filtration membranes or the like), or the like, or any combination thereof.

[0145] In various examples, one or more film(s) are used for a separation application (e.g., a biomolecule separation or the like). In various examples, a device is a separation device, which may be a filtration device or the like, or the like. In various examples, a separation device, which may be a filtration device or the like, is configured for (e.g., the film(s) is / are configured for) ion separation, or the like, or any combination thereof. In various examples, a separation device, which may be a filtration device or the like, is configured for (e.g., the film(s) separation (e.g., isolation or the like) of ion(s), such as, for example, potassium ions, calcium ions, or the like, or any combination thereof. In various examples, one or more or all of the ion(s) is / are biologically relevant ion(s).

[0146] The following Statements describe various examples of films (e.g., porous functionalized multiblock copolymer films and the like), methods of making films (e.g., porous functionalized multiblock copolymer films and the like), and uses of films (e.g., porous functionalized multiblock copolymer films and the like) of the present disclosure and are not intended to be in any way limiting: Statement 1. A method of forming a film (e.g., an isoporous graded film, such as, for example, a film comprising an isoporous surface region and an asymmetric pore region, where the isoporous surface region is disposed on at least a portion of an exterior surface of the asymmetric pore region) (e.g., a film of any of Statements 24–34), comprising: forming a first film from a deposition solution comprising two or more multiblock copolymers, where each multiblock copolymer comprises one or more corona-forming block(s) and one or more core-forming block(s), and i) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or ii) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s), two or more solvents, optionally, one or more homopolymer(s), optionally, one or more metal salt(s), optionally, one or more inorganic nanoparticle(s), and optionally, one or more additive(s), removing at least a portion of the solvent system from the film (e.g., allowing the first film to stand for a period of time such that at least a portion of the two or more solvents is removed from the film and at least a portion of the multiblock copolymer migrates to an exterior surface of the first film and a second film is formed); and contacting the second film with a phase separation solvent system, such that the film (e.g., the isoporous graded film, such as, for example, the filmcomprising an isoporous surface region and an asymmetric pore region, where the isoporous surface region is disposed on at least a portion of an exterior surface of the asymmetric pore region) is formed. Statement 2. A method according to Statement 1, where the one or more multiblock copolymer(s) comprise about 10 percent by weight or less of the deposition solution (based on the total weight of multiblock copolymer(s) and homopolymer(s), if present, in the deposition solution) and / or the one or more homopolymer(s) comprise about 90 percent by weight or more of the deposition solution (based on the total weight of multiblock copolymer(s) and homopolymer(s), if present, in the deposition solution). Statement 3. A method according to Statement 1 or 2, where the deposition solution comprises one or more homopolymer(s) and at least two of the two or more solvent(s) have different surface energy and / or surface tension and / or the lower surface energy and / or surface tension solvent preferentially solubilizes the one or more multiblock copolymer(s) relative to the homopolymer(s). Statement 4. A method according to any of the preceding Statements, where the deposition solution comprises one or more homopolymer(s) and the method comprises: providing a first polymer solution comprising one or more first polymer(s) (which may be multiblock copolymer(s) and / or homopolymer(s)) and one or more first solvent(s), where the first polymer solution optionally further comprises a metal salt; providing a second polymer solution comprising one or more second polymer(s) (which may be multiblock copolymer(s) and / or homopolymer(s)) and one or more second solvent(s), where the second polymer solution optionally further comprises a salt, where the solvent of the polymer solution comprising the multiblock copolymer has a lower surface energy and / or surface tension than the solvent of the polymer solution comprising the homopolymer(s) and / or the solvent of the polymer solution comprising the multiblock copolymer preferentially solubilizes the multiblock copolymer(s) relative to the homopolymer(s). Statement 5. A method according to any of the preceding Statements, where the pores of the isoporous region / layer have a pore size distribution of about less than 3, where the pore size distribution is the ratio of the maximum pore diameter (dmax) to the minimum pore diameter (dmin). Statement 6. A method according to any of the preceding Statements, where the isoporous region / layer has a thickness of about 20 nm (nm = nanometer(s)) to about 500 nm, including all 0.1 nm values and ranges therebetween, a plurality of pores having a size of about 5 nm to about 100 nm, including all 0.1 nm values and ranges therebetween, andthe asymmetric pore region / layer has a thickness of about 5 microns (micron(s) = micrometer(s)) to about 500 microns, including all 0.1 micron values and ranges therebetween, and pores having a size of about 10 nm to about 50 microns, including all 0.1 nm values and ranges therebetween. Statement 7. A method according to any of the preceding Statements, where the isoporous region / layer comprises a plurality of the pores about 5 nm to about 100 nm, including all 0.1 nm values and ranges therebetween, in size and / or a pore density of at least about 1 x 1014pores / m2. Statement 8. A method according to any of the preceding Statements, where at least a portion of the pores of the asymmetric pore region / layer increases in size along a dimension of the film moving away from a surface of the isoporous region / layer in contact with the asymmetric pore region / layer towards a surface of the asymmetric pore region / layer in contact with the substrate. Statement 9. A method according to any of the preceding Statements, where a plurality of the pores in the asymmetric pore region / layer lying on a plane about 20 microns from the surface layer and parallel to interface between the isoporous region / layer and asymmetric pore region / layer have a size in the plane of at least about 5 microns. Statement 10. A method according to any of the preceding Statements, where at least a portion of the pores in the asymmetric pore region / layer have an aspect ratio of greater than about 2. Statement 11. A method according to any of the preceding Statements, where at least a portion of the pore surface in the asymmetric pore region / layer is mesoporous. Statement 12. A method according to any of the preceding Statements, where the concentration of multiblock copolymer(s) and homopolymer(s), if present, in the deposition solution is about 5 to about 30 percent by weight (based on the total weight of the deposition solution), including all 0.1 percent by weight values and ranges therebetween. Statement 13. A method according to any of the preceding Statements, where at least one multiblock copolymer or all of the multiblock copolymers comprises one or more hydrogen- bonding block(s). Statement 14. A method according to any of the preceding Statements, where at least one multiblock copolymer or all of the multiblock copolymers further comprises one or more low Tgblock(s). Statement 15. A method according to any of the preceding Statements, where the homopolymer(s) is / are chosen from polysulfanes, polypropylenes, polylactic acids,polysulfones, poly(ether sulfone)s, poly(phenyl sulfone)s, cellulose acetates, poly(acrylonitrile)s, poly(amide)s, structural analogs thereof, and the like, and any combination thereof. Statement 16. A method according to any of the preceding Statements, where the deposition solution further comprises inorganic nanoparticles. Statement 17. A method according to any of the preceding Statements, where the solvent(s) is / are chosen from 1,4-dioxane, tetrahydrofuran, methanol, ethanol, toluene, chloroform, dimethylformamide, acetone, dimethylsulfoxide, N-methyl pyrrolidone, structural analogs thereof, and the like, and any combination thereof. Statement 18. A method according to any of the preceding Statements, where the deposition solution further comprises one or more small molecule(s) and the film further comprises the one or more small molecule(s). Statement 19. A method according to any of the preceding Statements, further comprising depositing one or more inorganic material(s) on at least a portion of a surface of the film. Statement 20. A method according to Statement 19, where the inorganic material is a plurality of metal nanoparticles, where the metal nanoparticles comprise one or more metal(s). Statement 21. A method according to any of the preceding Statements, where the first film is formed on a substrate, the method further comprising separating the film from the substrate to form a free-standing film. Statement 22. A method according to any of the preceding Statements, where the two or more multiblock copolymers comprise (or are) polyisoprene-b-polystyrene-b-poly(4-vinylpyridine) (ISV) and / or one or more structural analog(s) thereof and polystyrene-b-poly(4- vinylpyridine)(P4VP)-b-propylene sulfide(PPS) (SVPS) and / or one or more structural analog(s) thereof, and, optionally, the solvents comprise (or are) 1,4-dioxane (DOX) and tetrahydrofuran (THF). Statement 23. A composition (which may be (or be suitable for use as) a deposition solution (e.g., a dope or the like) comprising: two or more multiblock copolymers, where each multiblock copolymer comprises one or more corona-forming block(s) and one or more core- forming block(s), and i) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or ii) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s), optionally, one or morehomopolymer(s), optionally, two or more solvents, optionally, one or more metal salt(s); optionally, one or more inorganic nanoparticle(s), and optionally, one or more additive(s). Statement 24. A film made by a method of any of Statements 1–22 or 37–59. Statement 25. A film comprising an isoporous region and an asymmetric pore region, where the isoporous surface region is disposed on at least a portion of an exterior surface of the asymmetric pore region, where the isoporous layer comprises one or more multiblock copolymer(s) or the isoporous layer comprises one or more multiblock copolymer(s) and the asymmetric pore region comprises one or more homopolymer(s) and the asymmetric pore region / layer does not have about 10 percent by weight or more multiblock copolymer(s) (based on the total weight of multiblock copolymer(s) and homopolymer(s) in the asymmetric pore region / layer) as described herein; and / or the isoporous region / layer and asymmetric pore region / layer are not independently (or separately) formed and / or not laminated together to form the film, and the film is heterogeneous (e.g., a film comprises a plurality of domains, where at least two domains are different (e.g., structurally different, compositionally different, or the like, or any combination thereof). Statement 26. A film according to Statement 25, where the pores of the isoporous region / layer have a pore size distribution of about less than 3, where the pore size distribution is the ratio of the maximum pore diameter (dmax) to the minimum pore diameter (dmin). Statement 27. A film according to Statement 25 or 26, where the isoporous region / layer has a thickness of about 20 nm to about 500 nm, including all 0.1 nm values and ranges therebetween, a plurality of pores having a size of about 5 nm to about 100 nm, including all 0.1 nm values and ranges therebetween,, and / or the asymmetric pore region / layer has a thickness of about 5 microns to about 500 microns, including all 0.1 micron values and ranges therebetween, and pores having a size of about 10 nm to about 50 microns, including all 0.1 nm values and ranges therebetween. Statement 28. A film according to any one of Statements 25–27, where the isoporous region / layer comprises a plurality of the pores about 5 nm to about 100 nm, including all 0.1 nm values and ranges therebetween, in size and / or a pore density of at least about 1 x 1014pores / m2. Statement 29. A film according to any one of Statements 25–28, where at least a portion of the pores of the asymmetric pore region / layer increases in size along a dimension of the film moving away from a surface of the isoporous region / layer in contact with the asymmetricpore region / layer towards a surface of the asymmetric pore region / layer in contact with the substrate. Statement 30. A film according to any one of Statements 25–29, where a plurality of the pores in the asymmetric pore region / layer lying on a plane about 20 microns from the surface layer and parallel to interface between the isoporous region / layer and asymmetric pore region / layer have a size in the plane of at least about 5 microns. Statement 31. A film according to any one of Statements 25–30, where at least a portion of the pores in the asymmetric pore region / layer have an aspect ratio of greater than about 2. Statement 32. A film according to any one of Statements 25–32, where at least a portion of the pore surface in the asymmetric pore region / layer is mesoporous. Statement 33. A film according to any one of Statements 25–32, where the asymmetric pore region comprises (or is) one or more homopolymer(s). Statement 34. A film according to any one of Statements 25–33, where the film comprises pores having pore walls functionalized with one or more pore functionalizing group(s) and the individual pore functionalizing groups are covalently attached to the pore wall via at least one covalent bond to the pore wall. Statement 35. A device comprising one or more film(s) of the present disclosure (e.g., film(s) made by a method of the present disclosure, such as, for example, a method of any one of Statements 1–22, a film of Statement 23, a film made using a composition of Statement 24, a film of any one of Statements 25–34, or any combination thereof). Statement 36. A device according to Statement 35, where the device is a filtration device or the like. Statement 37. A method of forming a film (e.g., a porous functionalized multiblock copolymer film, an isoporous graded film, such as, for example, a film comprising an isoporous surface region and an asymmetric pore region, where the isoporous surface region is disposed on at least a portion of an exterior surface of the asymmetric pore region, an asymmetric film (e.g., comprising asymmetric pore structure along an axis substantially normal or normal to the film surface, or the like) (e.g., a film of any of Statements 60–72) comprising: forming a first film from a deposition solution comprising two or more multiblock copolymers, where each multiblock copolymer comprises one or more corona- forming block(s) and one or more core-forming block(s), and i) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or ii) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s),where, a first multiblock copolymer is present in a plurality of first micelles and a second multiblock copolymer is present in a plurality of second micelles and there is no substantial exchange of the polymer chains of the first of the multiblock copolymer from the first micelles with the polymer chains of the second multiblock copolymer of the second micelles, a solvent system comprising two or more solvents or the like, optionally, one or more homopolymer(s) or the like, optionally, one or more metal salt(s) or the like, optionally, one or more inorganic nanoparticle(s) or the like, and optionally, one or more additive(s) or the like; removing at least a portion of the solvent system from the first film to form a second film; and contacting the second film with a phase separation solvent system, such that the porous functionalized multiblock copolymer film is formed. Statement 38. A method according to Statement 37, where the deposition solution is formed from a first multiblock copolymer solution comprising one or more first multiblock copolymer(s) and one or more first solvent(s) and a second multiblock copolymer solution comprising one or more second multiblock copolymer(s) and one or more second solvent(s), such there is no substantial exchange of the polymer chains from the first micelles with the polymer chains of the second micelles. Statement 39. A method according to Statement 37 or 38, where ~χABN (the product of the Flory-Huggins interaction parameter and the degree of polymerization for A, a corona region of a first multiblock copolymer, and B, a corona region of a second multiblock copolymer) is about zero. Statement 40. A method according to any one of Statements 37–39, where ~χABN (the product of the Flory-Huggins interaction parameter and the degree of polymerization for A, a corona region of a first multiblock copolymer, and B, a corona region of a second multiblock copolymer) is greater than about zero. Statement 41. A method according to any one of Statements 37–40, where ~χABN (the product of the Flory-Huggins interaction parameter and the degree of polymerization for A, a corona region of a first multiblock copolymer, and B, a corona region of a second multiblock copolymer) is less than about zero. Statement 42. A method according to any one of Statements 37–41, where at least a portion, substantially all, or all of the plurality of first micelles and at least a portion, substantially all, or all of the plurality of second micelles exhibit attraction . Statement 43. A method according to any one of Statements 37–42, where at least a portion, substantially all, or all of the plurality of first micelles and at least a portion, substantially all, or all of the plurality of second micelles exhibit repulsion.Statement 44. A method according to any one of Statements 37–43, where at least a portion, substantially all, or all of the plurality of first micelles and at least a portion, substantially all, or all of the plurality of second micelles do not exhibit attraction or repulsion. Statement 45. A method according to any one of Statements 37–44, where at least one multiblock copolymer or all of the multiblock copolymers comprises one or more hydrogen- bonding block(s) or the like. Statement 46. A method according to any one of Statements 37–45, where at least one multiblock copolymer or all of the multiblock copolymers further comprises one or more low Tg block(s) or the like. Statement 47. A method according to any one of Statements 37–46, where the homopolymer(s) is / are chosen from polysulfanes, polypropylenes, polylactic acids, polysulfones, poly(ether sulfone)s, poly(phenyl sulfone)s, cellulose acetates, poly(acrylonitrile)s, poly(amide)s, structural analogs thereof, and the like, and any combination thereof. Statement 48. A method according to any one of Statements 37–47, where the one or more multiblock copolymer(s) comprise about 10 percent by weight or less of the deposition solution (based on the total weight of multiblock copolymer(s) and homopolymer(s), if present, in the deposition solution) and / or the one or more homopolymer(s) comprise about 90 percent by weight or more of the deposition solution (based on the total weight of multiblock copolymer(s) and homopolymer(s), if present, in the deposition solution). Statement 49. A method according to any one of Statements 37–48, where the deposition solution further comprises inorganic nanoparticles or the like. Statement 50. A method according to any one of Statements 37–49, where the concentration of the multiblock copolymer(s) and the homopolymer(s), if present, in the deposition solution (independently or in the aggregate) is about 5 to about 30 percent by weight (based on the total weight of the deposition solution or total weight of the solvents and the multiblock copolymer(s) and the homopolymer(s), if present), including all 0.1 percent by weight values and ranges therebetween. Statement 51. A method according to any one of Statements 37–50, where the solvent(s) is / are chosen from 1,4-dioxane, tetrahydrofuran, methanol, ethanol, toluene, chloroform, dimethylformamide, acetone, dimethylsulfoxide, N-methyl pyrrolidone, structural analogs thereof, and the like, and any combination thereof.Statement 52. A method according to any one of Statements 37–51, where the deposition solution further comprises one or more small molecule(s) or the like and the film further comprises the one or more small molecule(s) or the like. Statement 53. A method according to any one of Statements 37–52, further comprising depositing one or more inorganic material(s) on at least a portion, substantially all, or all of a surface of the film. Statement 54. A method according to Statement 53, where the inorganic material is a plurality of metal nanoparticles (e.g., where the metal nanoparticles comprise one or more metal(s)). Statement 55. A method according to any one of Statements 37–54, where the first film is formed on a substrate or the like. Statement 56. A method according to Statements 55, further comprising separating the film from the substrate to form a free-standing film. Statement 57. A method according to any one of Statements 37–56, where the two or more multiblock copolymers comprise (or are) polyisoprene-b-polystyrene-b-poly(4-vinylpyridine) (ISV) and / or one or more structural analog(s) thereof and polystyrene-b-poly(4- vinylpyridine)(P4VP)-b-propylene sulfide(PPS) (SVPS) and / or one or more structural analog(s) thereof, and, optionally, the solvents comprise (or are) 1,4-dioxane (DOX) and tetrahydrofuran (THF). Statement 58. A method according to any one of Statements 37–57, where the porous functionalized multiblock copolymer film comprises one or more or all of the following: (i) an isoporous region or layer thickness of about 20 nm (nm = nanometer(s)) to about 500 nm, including all 0.1 nm values and ranges therebetween, a plurality of pores having a size of about 5 nm to about 100 nm, including all 0.1 nm values and ranges therebetween, and / or an asymmetric pore region or layer has a thickness of about 5 microns (micron(s) = micrometer(s)) to about 500 microns, including all 0.1 micron values and ranges therebetween, and pores having a size of about 10 nm to about 50 microns, including all 0.1 nm values and ranges therebetween; (ii) an isoporous region or layer pore density of at least about 1 x 1014pores / m2; (iii) an isoporous region or layer pore size distribution of about less than about 3, where the pore size distribution is the ratio of the maximum pore diameter (dmax) to the minimum pore diameter (dmin); (iv) at least a portion of the pores of the asymmetric pore region or layer increases in size along a dimension of the film moving away from a surface of the isoporous region or layer in contact with the asymmetric pore region or layer towards a surface of the asymmetric pore region or layer in contact with the substrate;(v) a plurality of the pores in the asymmetric pore region or layer lying on a plane about 20 microns from the surface layer and parallel to interface between the isoporous region or layer and asymmetric pore region or layer have a size in the plane of at least about 5 microns; (vi) at least a portion of the pores in the asymmetric pore region or layer have an aspect ratio of greater than about 2; or (vii) at least a portion of the pore surface in the asymmetric pore region or layer is mesoporous. Statement 59. A composition suitable for use as a deposition solution comprising: two or more multiblock copolymers, where each multiblock copolymer independently comprises one or more corona-forming block(s) and one or more core-forming block(s), and i) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or ii) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core- forming block(s), optionally, one or more homopolymer(s) or the like, optionally, two or more solvents or the like, optionally, one or more metal salt(s) or the like; optionally, one or more inorganic nanoparticle(s) or the like, and optionally, one or more additive(s) or the like. Statement 60. A film (e.g., a porous functionalized multiblock copolymer film, an isoporous graded film, such as, for example, a film comprising an isoporous surface region and an asymmetric pore region, where the isoporous surface region is disposed on at least a portion of an exterior surface of the asymmetric pore region, an asymmetric film (e.g., comprising asymmetric pore structure along an axis substantially normal or normal to the film surface, or the like) comprising at least two multiblock copolymers, where the film comprises:i) an isoporous region or layer comprising a) two or more multiblock copolymers, where each multiblock copolymer comprises one or more corona-forming block(s) and one or more core- forming block(s), and i) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or ii) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s), b) a thickness of about 20 nm to about 500 nm, including all 0.1 nm values and ranges therebetween, ii) an asymmetric region or layer comprising a) a thickness of about 5 microns to about 500 microns, including all 0.1 micron values and ranges therebetween, and b) pores, independently, having a size (or an average size) (e.g., a longest cross-sectional dimension, such, as for example, a diameter or the like) of about 5 nm to about 100 microns, including all 0.1 nm values and ranges therebetween, where, the isoporous surface region is disposed on at least a portion, substantially all, or all of an exterior surface of the asymmetric pore region, at least a portion,substantially all, or all of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of the core blocks of a second multiblock copolymer core is / are isotropically or statistically distributed in at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer, or is / are disposed in a plurality of distinct domains forming at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer, or form(s) a three-dimensional pore lattice in at least a portion, substantially all, or all of the pore wall or pore wall surface(s) of the isoporous region or layer, and at least a portion, substantially all, or all of the functional groups of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of a functional group or functional groups of the core blocks of a second multiblock copolymer core are disposed on at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer. Statement 61. A film according to Statement 60, where the isoporous surface region or layer comprises a) a plurality of pores of about 5 nm to about 100 nm, including all 0.1 nm values and ranges therebetween, in size (independently or on average) (e.g., a longest cross-sectional dimension, such, as for example, a diameter or the like), and / or b) a pore density of at least about 1×1014pores / m2. Statement 62. A film according to Statement 60 or 61, where the pores of the isoporous region or layer have a pore size distribution of about less than about 3, where the pore size distribution is the ratio of the maximum pore diameter (dmax) to the minimum pore diameter (dmin). Statement 63. A film according to any one of Statements 60–62, where the distinct domains, independently, have a linear dimension of about 10 nm to about 3 microns, including all 0.1 nm values and ranges therebetween. Statement 64. A film according to any one of Statements 60–63, where the asymmetric region or layer comprises a) an asymmetic substructure (such as, for example, a “spongy” substructure, a “finger-like” asymmetric substructure, or the like), and b) a plurality of pores having a size (independently or on average) (e.g., a longest cross-sectional dimension, such, as for example, a diameter or the like) of about 5 nm to about 100 microns, including all 0.1 nm values and ranges therebetween. Statement 65. A film according to any one of Statements 60–64, where at least a portion of the pores of the asymmetric region or layer increases in size along a dimension of the film moving away from a surface of the isoporous region or layer in contact with the asymmetricsubstructure towards a surface of the asymmetric pore region or layer in contact with the substrate. Statement 66. A film according to any one of Statements 60–65, where a plurality of the pores in the asymmetric region or layer lying on a plane about 20 microns from the surface layer and parallel to interface between the isoporous region or layer and asymmetric substructure have a size in the plane of at least about 5 microns. Statement 67. A film according to any one of Statements 60–66, where at least a portion of the pores in the asymmetric region or layer comprise an aspect ratio of greater than about 2. Statement 68. A film according to any one of Statements 60–67, where at least a portion of the pore surface in the asymmetric substructure is mesoporous. Statement 69. A film according to any one of Statements 60–68, where the asymmetric region or layer comprises one or more homopolymer(s). Statement 70. A film according to Statement 69, where the asymmetric region or layer does not comprise about 10 percent by weight or more multiblock copolymer(s) (based on the total weight of multiblock copolymer(s) and homopolymer(s) in the film. Statement 71. A device comprising one or more film(s) of any one of Statements 60–70. Statement 72. The device of Statement 71, where the device is a filtration device, a separation device, or a catalysis device.

[0147] The steps of the methods described in the various embodiments and examples disclosed herein are sufficient to produce one or more films (e.g., porous functionalized multiblock copolymer film(s), or the like) or any combination thereof or carry out a method of the present disclosure. Thus, in an embodiment, a method consists essentially of a combination of the steps of the methods disclosed herein. In another embodiment, a method consists of such steps.

[0148] The following Examples are presented to illustrate the present disclosure. The Examples are not intended to be limiting in any manner. EXAMPLE 1

[0149] This example provides examples of functionalized multiblock copolymer films of the present disclosure, and methods of making and using same.

[0150] Provided herein are asymmetric membranes formed from mixtures of chemically distinct block copolymer micelles with negligible chain exchange in a non-equilibrium process. A combination of concepts from block copolymer self-assembly and colloidal science with advanced characterization, machine learning, and Brownian dynamicssimulation methodologies were used to determine strategies in which micelle cores determine the available pore surface chemistries while repulsive micelle coronas (together with various processing parameters) control their placement, particularly throughout the top separation layer. These results are provide general concepts for the formation of porous and bulk materials with advanced properties.

[0151] Provided herein are strategies to allow control over various pore chemistries and their distribution throughout the top separation layer, both parallel and normal to the membrane surface (i.e., laterally and vertically; FIG.1b). That control may be the result of spontaneous self-organization, perhaps dictated by the molecular engineering of multiple chemically distinct polymeric components in the dope submitted to a scalable fabrication methodology, rather than elaborate post-synthesis modifications of single polymer / BCP based membranes. The combination of concepts from BCP self-assembly and colloidal science with advanced characterization, machine learning (ML), and simulation methodologies contribute to identification of such strategies. To that end, it is feasible to control pore chemistries and their spatial distribution via BCP micelle core and corona chemistries, respectively, by subjecting a polymer dope consisting of multiple chemically distinct BCP micelle populations with negligible chain exchange to the SNIPS process.

[0152] With control over the positioning of multiple micelle core chemistries as the aim, BCP micelle mixtures with distinct corona blocks but identical core blocks, which translate to a single pore chemistry, or identical corona blocks but distinct core blocks, which limit control over the positioning of different pore chemistries, will not be considered here. Systems with distinct BCP micelle coronas and cores can be generally categorized based on: (i) lattice symmetry (distinct micelle populations may form identical lattices, distinct lattices, or no lattice at all); (ii) size (distinct micelle population sizes may be similar or different, e.g., affecting their lattice spacings); and (iii) corona chemistry (corona interactions between distinct micelle populations may be attractive or repulsive, i.e., χABN < 0 or χABN > 0). This provides a multitude of possible experimental realizations, even for binary systems of type A+B. While mixtures are not conceptually limited to binary BCP micelle populations (FIG. 1c), implementation requires finding experimentally viable parameter sets (e.g., a common solvent system) for which SNIPS can be successfully executed. Furthermore, while systems with same lattice symmetry and similar size (spacing) but different corona (and core) chemistries, or same lattice symmetry but different size (spacing) and corona (and core) chemistries, may be conceptually simple, they may pose challenges in component differentiability and identification. As demonstrated herein, conceptually more complexsystems with differences in lattice symmetry, size, and corona (and core) chemistries can overcome such challenges. Moreover, by employing advanced characterization methodologies such systems are demonstrated to offer high tunability of the spatial distribution of pore chemistries.

[0153] The structure formation of SNIPS membranes fabricated from chemically distinct BCP micelle mixtures of poly(isoprene-block-styrene-block-4-vinylpyridine) (PI-b-PS-b- P4VP, or ISV) and poly(styrene-block-4-vinylpyridine-block-propylene sulfide) (PS-b-P4VP- b-PPS, or SVPS) were studied. This blended triblock terpolymer micelle system has model character. Chain exchange between micelles is negligible and the distinct 4-fold and 6-fold symmetries of the respective ISV and SVPS top surface pore structures are visually distinguishable—enabling quantification of their surface distributions—and realized using the same solvent system. Furthermore, fluorescent probes can be selectively attached to the thiol- terminated PPS block covering the SVPS pore surface via thiol-ene click chemistry. This allows optical visualization of its location on the top surface, allowing correlation of structure with specific chemistry / function. Using a combination of SNIPS membrane formation, reactive ion etching, scanning electron microscopy (SEM), ML assisted image analysis, optical super-resolution microscopy (SRM), and computational modeling, it was quantitatively analyzed how macromolecular and experimental parameters including composition of the micellar polymer dope, BCP molar mass, and evaporation time determine top separation layer structure. Results from binary BCP micelle mixtures reveal conceptually simple strategies in which micelle cores determine the available pore surface chemistries while repulsive micelle coronas (together with various processing parameters) control their horizontal and vertical placement throughout the top separation layer. This extension to ternary BCP micelle mixtures suggests generalization to multicomponent membranes with finely tunable pore surface chemistries. While demonstrated here for porous membrane formation processes, it is expected that the application of colloidal science ideas to multicomponent BCP micelle self-assembly in the limit of negligible chain exchange between micelles could also contribute to classes of bulk polymeric materials. In such block copolymer micelle mixture (BCM2) based materials, core and corona chemistries of multiple micelle constituents may be chosen to induce advanced materials properties in the mixed systems. These may include periodic superstructures, both with ordered (e.g. from selectively attractive micelle coronas) and disordered local composition, as well as entirely non-periodic materials, thereby paving the way for soft matter analogues, e.g., of high-entropy alloys and ceramics.

[0154] Fabrication and structural characterization of multicomponent membranes.

[0155] Multicomponent BCP membranes were fabricated via a modified SNIPS process described elsewhere (FIG.1a). To distinguish this method involving chemically distinct micelles from earlier approaches, it will be referred to as micelle mixture-based SNIPS (M2SNIPS) hereafter.11 weight percent (wt.%) ISV solutions and 14-15 wt.% SVPS solutions in 1,4-dioxane (DOX) and tetrahydrofuran (THF) were individually prepared and stirred until homogeneous (polymer characteristics, Table 1; solution compositions for all membranes in this study, Table 2). Thereafter, the individual solutions were combined (by wt.) and stirred for 15 minutes. The multicomponent polymer dope was pipetted onto a glass substrate and cast into a thin film using an automated blade-casting machine. The film was evaporated for a set amount of time (60-100 s) prior to immersion in a non-solvent de-ionized (DI) water bath, which precipitated out the membrane and kinetically-trapped its non- equilibrium structure. FIG.1b schematically depicts an M2SNIPS membrane top surface, consisting of chemically distinct, macroscopically segregated porous domains. Macroscopic segregation is expected to occur due to the distinct ISV and SVPS micellar structures. In ISV and SV BCPs subjected to phase inversion, the pore forming P4VP blocks constitute the micelle cores, while the matrix forming PI and / or PS blocks constitute the micelle coronas. The short (< 3 kDa) PPS block in SVPS is assumed to reside in the core with the P4VP block. Unfavorable enthalpic corona interactions ~χN (the product of the Flory-Huggins interaction parameter and the degree of polymerization) between the ISV PI block and the SVPS PS block favor same species interactions, e.g., ISV-ISV, contributing to macroscopic segregation. A prerequisite for interspecies macroscopic segregation is negligible chain exchange between micelle species on the timescale of the M2SNIPS process. It is known that the molecular exchange dynamics of BCP micelles have a temperature dependence, a double exponential dependence on core block length (Ncore), and slow down for concentrated, ordered systems. The room temperature processing conditions and relatively large core blocks (Ncore, ISV = 163; Ncore, SVPS53 = 113; Ncore, SVPS67 = 161), combined with the fact that suitable casting dope concentrations >10 wt. % (Table 2) are just below the ordered solution onset concentration, suggest that chain exchange on the timescale of the experiment (~ minutes) is negligible. This is consistent with experimental results discussed below as well as further cross experiments (see FIG.7 and Supplementary Discussion 1).

[0156] Table 1. Polymer molar mass (Mn), weight fractions (f) and polydispersity index (PDI).Polymer Mn (kDa) fPI fPS fP4VP fPPS fPDMAEMA PDI ISV121121 0.20 0.66 0.14SVPS53 53 - 0.79 0.18 - SVPS67 67 - 0.76 0.20 0.04 - 1.08 ISA164164 0.20 0.68 - - 0.12 1.18

[0157] Table 2. Compositions and solution concentrations of BCP membrane samples. Solution DOX:THF Evaporation Time (s): FIG. BCP Mass Ratio (%) Concentration* , f60: FIGS.3a,f, 4l, 10, 13a, 14a, 16, 19, 20a, Table 4 d, 0, , b,52 ISV: 48 SVPS53 12.7 78:22 80: FIG.8d-f 66 ISV: 34 SVPS67 11.7 73:27 100: FIG.8m-o**Sample was cast at a relative humidity < 25%.***Sample was made from a separate solution batch and cast on a different day than the sample in the row above it and was thus listed separately.

[0158] FIG.2a-d shows scanning electron microscopy (SEM) images of single- (pure) and multicomponent (blended) SNIPS membrane top surfaces (for cross sectional SEM images see FIG.8). To better visualize the distribution of chemically distinct ISV and SVPS regions in M2SNIPS membranes, ML assisted segmentation (FIG.2e-h) was performed on the respective SEM images via the Trainable Weka Segmentation (TWS) ImageJ plugin, which relies on user-defined classes to train a classifier for subsequent segmentation (see Methods and FIG.9, FIG.10). Two types of component distributions were observed when the same ISV polymer was mixed with differently sized SVPS polymers (53 vs.67 kDa). Blends made with SVPS53 exhibited a “patchy” morphology (FIG.2f)—individual components exhibited long-range (~μm) order—whereas blends made with SVPS67exhibited a “disordered” morphology (FIG.2g)—individual components lacked long-range order—with ISV and SVPS adopting the same local pore packing symmetries (2D square vs.2D hexagonal, insets FIG.2a,d; FIG.2q) as in pure membranes.

[0159] To confirm that regions distinct by symmetry (4-fold vs.6-fold packed) were indeed chemically distinct (ISV vs. SVPS), rather than a manifestation of two coexisting morphologies stemming from a single BCP, optical super-resolution microscopy (SRM) was employed. A sulfo-Cy5(-) maleimide dye was attached to the sulfhydryl end group of the PPS block via thiol-ene click chemistry, allowing visualization of the SVPS component distribution in blended as well as pure SVPS-based membranes in FIG.2f-h. The dye conjugation protocol was also performed on an ISV membrane as a control. Stochastic optical reconstruction microscopy (STORM) images are shown in FIG.2i-l (Methods). To quantitatively compare the segmentation and SRM results for all four membranes (FIG.2r), the area fraction of SVPS pores (from segmentation; grey) and number of localizations (from SRM; green) were normalized to the values of the pure SVPS membrane. The “patchy” and “disordered” membranes had normalized SVPS area fractions of 18.0 ^ 3.1% and 46.4 ^ 3.7%, and normalized localization fractions of 21.9 ^ 0.3% and 49.7 ^ 2.8%, respectively.Close agreement between the two analyses suggests a strong correlation between nanoscale structure and chemistry / function, i.e., distinction by symmetry of packing also reflects chemical distinction.

[0160] Tuning relative humidity (RH) allowed realization of “patchy” and “disordered” distributions using the same blended polymer system (ISV + SVPS53). Larger ISV grains (FIG.2m,n)—predominantly observed at low RH (< 25%) conditions (for dope compositions close to 1:1 ISV:SVPS53 by wt.) after N2 purging the environment—led to “patchy” distributions (FIG.2f); whereas smaller ISV grains (FIG.2o,p)—predominantly observed under moderate RH (~ 40%) conditions (from 3:1 to 1:1 ISV:SVPS53 by wt., vide infra)— gave rise to “disordered” distributions (e.g., see FIG.3e). This suggested that grain size is affected by RH. In turn, smaller grain sizes increase the relative surface area associated with structural defects (e.g. grain boundaries), which, as will be discussed below, may facilitate the emergence of the second component (SVPS) on the top surface. Furthermore, it was assessed whether the co-assembly of the two BCP components on the membrane top surface affected individual component structure. To that end, the same-component lattice spacing (ISV-ISV and SVPS-SVPS pore-to-pore distances) distributions of the ISV and SVPS components in a “disordered” membrane (FIG.3e), ISVdisand SVPSdis, were compared to those of pure ISV and SVPS membranes, ISVpure and SVPSpure (FIG.11e,j). Early analysis (FIG.2s) suggested that, in a highly “disordered” system, lattice spacings of ISV components were reduced, while those of SVPS components were increased, relative to their single- component reference systems. This emphasizes differences between the present soft colloidal system and what is observed for hard sphere interactions precluding shrinkage. Based on these results, subsequent work focused on “disordered” membranes, whose smaller grains give rise to larger grain boundary areas from which structural changes originate (compare FIG.2m,p).

[0161] Controlling and understanding M2SNIPS membrane top surface structure and composition.

[0162] The ability to control top surface structure and composition of M2SNIPS membranes is expected to be crucial to tailoring properties. Top surface analysis will focus on multicomponent membranes fabricated from ISV and smaller SVPS53, since the differences in pore size (17.2 ^ 4.6 nm vs.13.6 ^ 2.7 nm, FIG.12) facilitated visual identification and ML based classifier training. The top surface evolution from 60 s to 100 s evaporation time in 10 s increments for membranes cast from a 67 ISV: 33 SVPS53 (wt. %)dope (Table 2) is revealed by segmented SEM in FIG.3a-e (raw SEM in FIG.13a-e). Subsequent quantification showed that the fraction of SVPS on the top surface (number and area of SVPS pores) increases linearly with evaporation time (FIG.14a). Surface segregation behavior in blended polymer systems is generally attributed to polymer surface tension: the polymer species with the lower surface tension is enriched on the top surface, decreasing the overall free energy. For the given BCP micelle structures, the corona components exposed to the polymer-air interface are PI and PS for ISV, and PS for SVPS, while the core components are shielded from the external environment. PI has a lower surface tension (32 mJ m-2) than PS (40 mJ m-2), thus ISV should be enriched on the membrane top surface. While true for shorter evaporation times (60 to 70 s), observed behavior at longer evaporation times (90 to 100 s) suggests additional factors at play.

[0163] Explicit variation of environmental RH was previously shown to affect surface order and uniformity for pure ISV membranes. It is considered that the indirect increase in local RH (at the polymer-air interface) with evaporation time, at constant environmental RH (40 - 45%), is partially responsible for the surface phenomena observed at longer evaporation times. Endothermic solvent evaporation is known to cause top surface temperature to decrease with increasing evaporation time, sometimes to the dew point. For a given water vapor content in air, a decrease in local temperature will result in an increase in local RH. As the polymer-air interface is increasingly replaced by a polymer-vapor interface, the original surface structure is destabilized, prompting reorganization. This type of surface reorganization has been shown to occur even when the polymer is in its glassy state. Given the water contact angles of PI (~102°) and PS (~87°), the polymer-vapor interface favors drawing SVPS to the surface to lower the interfacial tension with water vapor. The lower molar mass of SVPS53 relative to SVPS67 grants the micelles more mobility to facilitate this reorganization. Analysis of solvent surface tension effects, recently found to control surface segregation behavior in certain blended polymer films, led to inconclusive results (see Table 3 and Supplementary Discussion 2).

[0164] Table 3. Solubility parameters and surface energies of solvents and polymers. SolventSolubility parameters and molar volume* ^^solvent-polymer**SuWater andrfaceEn Contact Polymer ergy δDδPδΗVmAngle Species (MPa0.5) (MPa0.5) (MPa0.5)(mL / mol)(°)THF 16.8 5.7 8.0 81.7 - [0.33] [0.81] DOX 19.0 1.8 7.4 85.7 0.28 0.69 34 -[0.25] [0.80] PS 18.7 5.9 3.5 - - - 40 87.3 [18.5] [4.5] [2.9] PI 16.9 1.1 -0.4 - - - 32 102.4 [16.6] [1.4] [-0.8] **χsolvent-polymer parameters were calculated according to the following equation: ^^ ^^ ^^ ^^ = ( ^^ ^^^^ ^^)[ ^^ ^^, ^^ ^^ ^^ ^^ ^^ ^^ ^^ − 2 2( ^^ ^^, ^^ ^^ ^^ ^^ ^^ ^^ ^^) + 0.25( ^^ ^^, ^^ ^^ ^^ ^^ ^^ ^^ ^^ − ^^ ^^, ^^ ^^ ^^ ^^ ^^ ^^ ^^) +the error in Voronoi cell constructions arising from small lattice distortions present in the square-packed ISV lattice by placing a constraint on the angle edges subtend from the pore center (Methods, FIG.15). It was performed on the dataset shown in FIG.3a-e, to further elucidate the surface structural evolution as a function of evaporation time (FIG.3f-j). Polygons with 4- and 6-fold coordination, representing ideal ISV and SVPS pore structures, are denoted in red and blue, respectively. All other Voronoi polygons, predominantly fivefold coordinated (FIG.16), are considered defects and colored white. At 60 s evaporation time (FIG.3f), the top surface is clearly ISV-dominated, with defect polygons primarily constituting the grain boundaries. As evaporation time increases, ISV grains are further broken up by grain boundaries (defects), with SVPS cells emerging amidst them. This is consistent with the expectation that grain boundaries facilitate the emergence of SVPS micelles on the top surface as it is more energetically favorable over breaking up ordered ISV regions. MVA suggests that defects are precursors for SVPS, consistent with such thermodynamic arguments. The fractions of ISV, SVPS, and defect cells on the membrane top surface from MVA (FIG.4l) reflect the same overall trend as segmentation analysis, i.e., surface fraction of SVPS increases with evaporation time.

[0166] To analyze the impact of two-component co-assembly on same-component lattice spacing distributions, the local n-fold (n = 4 or 6) bond-orientational order parameter, Φn, was used (Methods) to identify ISV and SVPS clusters with high degrees of 4- and 6-fold symmetry, respectively (FIG.17). After imposing a threshold (Φ6 > 0.8) to focus analysis on non-grain-boundary regions (FIG.18b), the lattice spacing of the SVPS component (SVPSblend) in blend MVA panels shown in FIG.3g-j, was compared to that of pure SVPS membranes (SVPSpure) for 70 to 100 s evaporation (FIG.3p-s; corresponding SEM images of pure SVPS in FIG.11g-j). SVPSblend has consistently larger lattice spacings than SVPSpure.Lattice spacing analysis based on segmentation classification reveals the same trend (FIG. 19). This is consistent with lower pore densities in the blended membrane samples (Table 4) compared to those of pure SVPS membranes (7.0 – 9.2 ^ 1014 / m2versus 1.3 – 1.4 ^ 1015 / m2from 70-100 s). SVPS micelles emerging from the grain boundaries likely feel a strain field due to their incompatibility with the larger lattice of the surrounding ISV micelles, preventing them from adopting their native lattice structure (symmetry and spacing). This argument is consistent with the fivefold-coordinated defect cells bridging the 4- (ISV) and 6-fold- coordinated (SVPS) cells in the MVA. Additionally, up-concentration from continued solvent evaporation limits mobility. Together, these effects may contribute to the sizable defect fraction (> 50%) for longer evaporation times. FIG.3t compares the result of a similar lattice spacing analysis for ISVblend versus ISVpure at 90 s evaporation after imposing Φ4 > 0.97 (FIG.18a; corresponding SEM image of pure SVPS53in FIG.11d). The full dataset is shown in FIG.20. Compared to shorter evaporation times (60 to 80 s; FIG.20a-c), there is a pronounced shift of the ISVblenddistribution to smaller lattice spacings relative to ISVpureat 90 s. This shift can be rationalized by a similar lattice incompatibility argument as used earlier, where both lattices “compromise”. The pore density of the blend increases by 34% (Table 4) over a 30 s interval, as compared to only a 3% increase (6.4 – 8.6 ^ 1014 / m2versus 5.4 – 5.6 ^ 1014 / m2from 60-90 s) for pure ISV over the same interval. Soft ISV micelle- micelle interactions can accommodate additional smaller SVPS micelles on the top surface by increasing ISV micelle corona interpenetration, effectively decreasing ISV lattice spacing. Overall, variations in evaporation time demonstrate that this experimental parameter can be used to tailor surface compositions far from the equilibrium bulk composition.

[0167] Table 4. Top surface characterization results derived from segmented SEM images of the 67 ISV: 33 SVPS53blend derived M2SNIPS membranes depicted in FIG.3a-e. vaporation Time (s) Pore Diameter (nm)*Pore Density E*14 2 *Porosity (%) (10 / m ) 60 18.4 ± 0.417.5 ± 0.2 70 17.8 ± 0.1 7.0 ± 0.2 18.1 ± 0.3 80 17.0 ± 0.2 7.7 ± 0.1 18.0 ± 0.1 90 16.5 ± 0.3 8.6 ± 0.3 18.8 ± 0.1 100 16.4 ± 0.1 9.2 ± 0.2 20.1 ± 0.1*Standard deviations from analysis of three separate sample regions.

[0168] Coarse-grained (FIG.21) Brownian dynamics (BD) simulations were used to better understand the experimentally observed ISV / SVPS micelle rearrangements at the film- air interface during solvent evaporation (Methods). It should be noted that although these BD simulations do not capture all the macro- or mesoscale non-equilibrium effects occurring experimentally (e.g., thermal gradients, flow, phase inversion, etc.), they do describe the local structural evolution of micelles near the interface, and hence provide mechanistic insights difficult to achieve from isolated experimental time points. BCP micelles were modeled as soft spheres interacting via the Yoshida-Kamakura soft potential (FIG.22). Experimental Voronoi diagrams informed the dynamic variation of the quadratic interfacial potential (FIG. 23) as the only driving force to induce structural transitions in the simulation box interface layers (FIG.24), consistent with changes in local RH driving surface reorganization. The fact that MVA of top layer simulation snapshots captures features of the experimental data (compare FIG.3k-o with 3f-j) supports validity of this approach. BD simulation snapshots (FIG.4a,b) suggest interlayer coordination within large ISV (red) grains and interlayer heterogeneity—clusters of SVPS (blue) spheres beneath disordered ISV spheres—within grain boundaries, thereby precluding efficient sphere packing. As interfacial potential is tuned in favor of SVPS spheres, these sequentially deform defective ISV regions (e.g., grain boundaries) in the first layer as they surface (FIG.4c). Given the presence of an underlying SVPS reservoir, a single SVPS surface transition (FIG.4d) provides a pathway for more SVPS spheres to follow, leading to SVPS surface grain growth amidst ISV grain boundaries (FIG.4e), consistent with experimental MVA.

[0169] An alternative method of tailoring top surface composition, e.g. by varying blend composition in the casting dope, may be desired. To that end, a series of M2SNIPS membranes containing 25 to 50 wt. % SVPS53(relative to total polymer wt.) was fabricated at constant evaporation time of 80 s (Table 2). Segmentation (FIG.4f-h) and MVA (FIG.4i-k) were performed on top surface SEM images (FIG.13f-h) as detailed earlier. Quantitative analysis (FIG.4m and FIG.14b) reveals a positive correlation between the fraction of SVPS53polymer in the casting dope and on the top surface. The relative distribution of MVA polygons, i.e. SVPS emerging from the grain boundaries, and defect polygons bridging ISV and SVPS polygons, is consistent with prior discussion. Qualitatively, the effects of increasing evaporation time and increasing dope fraction of SVPS polymer are the same,resulting in increased (i) fraction of SVPS on the top surface, (ii) pore density, and (iii) porosity (Tables 4 and 5).

[0170] Table 5. Top surface characterization results derived from segmented SEM images of 80 s evaporation derived M2SNIPS membranes depicted in FIG.4f-h. Solution Composition Pore Diameter Pore Density Porosity (%)*(ISV121 : SVPS53 by (nm)*(1014 / m2)*weight) 75:25 18.7 ± 0.1 6.1 ± 0.2 17.2 ± 0.1 60:40 17.4 ± 0.1 7.8 ± 0.2 19.3 ± 0.3 50:50 16.8 ± 0.1 8.5 ± 0.2 19.7 ± 0.4*Standard deviations from analysis of three separate sample regions.

[0171] Sub-surface membrane structure and composition as revealed by etching.

[0172] Membrane separation behavior is determined by the ~100 nm thick top separation layer, typically comprising three to four micelle layers in SNIPS derived materials (FIG.5b, top right). For single-component single-pore-chemistry membranes, separation layer thickness (e.g., cylinder length in hexagonal SV based systems) is primarily tuned in the interest of increasing absolute permeability. As outlined earlier, for M2SNIPS membranes involving distinct pore chemistries, details of the separation layer component distribution both parallel and normal to the membrane surface (laterally and vertically) may provide strategies, e.g., to surpass the limits of size-selective filtration and achieve other advanced membrane properties. For example, vertically heterogeneous pores, which expose a solute to multiple chemical environments as it traverses the top separation layer, could be engineered to execute a desired series of reactions. In contrast, laterally heterogeneous (but vertically homogeneous) pores comprising, e.g., a charge-patterned membrane surface, could be used to create sets of singly charged channels that facilitate the concurrent transport of cations and anions. Pore structure / chemistry in the separation layer were examined beyond the top surface layer. To that end, a method in etching that overcomes the ambiguity associated with determining exact etch depth was used. M2SNIPS membranes were partially covered with a Si wafer and subjected to reactive ion etching (RIE) with O2plasma for 30 s (FIG.5b, left). At the wafer edge, the finite wafer thickness (500-550 μm) casts a ‘shadow’ on themembrane, creating an etch gradient that allows probing increasing etch depths along the edge normal (FIG.5b, bottom right). SEM images were taken at regular intervals along this gradient for “patchy” (FIG.25) and “disordered” (FIG.26) ISV / SVPS53 M2SNIPS membranes. Representative SEM images of regions with shallower (FIG.5e,f) and deeper etches (FIG.5i,j) show structural differences between “patchy” and “disordered” membranes, respectively. For crude visualization, TWS was used to segment images into three classes— ISV matrix (red), SVPS matrix (green), and pore space (magenta). Segmentations (FIG. 5d,g,h,k) are shown alongside their respective original SEM images. FIG.5d (shallow etch, “patchy”) shows extensive square packed ISV grains (red) with enlarged pores due to preferential etching of P4VP-lined pore walls, and a ‘deformed’ region (green) lining the ISV grain boundaries. The location, absence of visually identifiable hexagonal pore structures, and poorer mechanical properties of the small SVPS53 BCP suggest that this ‘deformed’ region represents the SVPS polymer. In comparison, the shallow etch of the “disordered” membrane (FIG.5g) has ‘deformed’ regions interspersed amidst smaller ISV grains, consistent with the previous observations about ISV grain size in “patchy” and “disordered” membranes, further suggesting that these (green) regions represent SVPS.

[0173] The same segmentation analysis was performed on regions with deeper etches (FIG.5h,k; full series of segmentations in FIG.27). Same-component correlation lengths were comparable to the corresponding regions of “patchy” and “disordered” membranes with more shallow etches. In “patchy” membranes, the large lateral grain sizes across multiple micelle layers suggest vertically homogeneous pores (FIG.5a), consistent with interlayer coordination and corroborated by BD simulations (vide supra). In “disordered” membranes, the absence of lateral long-range order makes vertical heterogeneity (FIG.5c) more probable. These results document feasibility of using the combination of BCP micelle self-assembly and approaches known from colloidal science with advanced characterization, ML and simulation methodologies to identify strategies that allow control over two pore chemistries and their lateral and vertical distribution throughout the membrane top separation layer.

[0174] Three BCP component M2SNIPS membranes and their pH dependent permeability.

[0175] In principle, it was expected that the M2SNIPS protocol (FIG.1a) is generalizable to systems containing more than two chemically distinct BCPs and thus more than two distinct pore surface chemistries. To test this, triblock terpolymer PI-b-PS-b-poly(N,N- dimethylamino)ethyl methacrylate) (PI-b-PS-b-PDMAEMA, ISA, see Table 1 for characterization) was introduced as a third component (FIG.1c). FIG.6c,d depicts SEMimages of top surface and cross-sectional views, respectively, of a three-component asymmetric M2SNIPS membrane consisting of 43:42:15 ISV:SVPS67:ISA by wt., and evaporated for 100 s (Table 2). While not assessable via SEM top surface image analysis, the presence of the ISA component in the separation layer pores can alternatively be confirmed by its ability to modulate the pH-dependent permeability in tests using a pressurized dead-end stirred cell (Methods). PDMAEMA has a pKa of 7.8, below which it is protonated, leading to chain stretching and pore closure, accompanied by a drastic permeability drop. ISA- containing membranes thus display a higher pore closure onset pH than P4VP-containing ISV and SVPS membranes (pKaP4VP = 4.6). As a control experiment, FIG.6a demonstrates that while, as expected, the pore size difference between ISV and SVPS affects the absolute permeability (inset), the relative amounts of ISV and SVPS in two-component membranes (top surface SEM images in FIG.28) do not substantially affect the pH-dependent behavior as revealed by data normalization. In stark contrast, normalized pH-dependent permeability curves (FIG.6b) of three-component membranes clearly establish that even minor (< 10 wt. %) differences in ISA content drastically modulate behavior. At the same time, addition of the third component had no substantial adverse effects on absolute permeability at higher pH prior to pore closure (compare FIG.6a, inset, with FIG.6b, inset).

[0176] In summary, it was demonstrated that the use of BCP micelle mixtures in the absence of chain exchange provides an advanced strategy to control the distribution of pore surface chemistries parallel and normal to the top surface in the separation layer of M2SNIPS derived polymer membranes. First experimental results further suggest that this approach can be extended to more than two distinct BCP chemistries, paving the way to a plethora of hitherto unexplored membrane compositions and properties. Modifications relative to conventional NIPS based membrane formation are limited to dope composition, while formation process steps are compatible with existing industrial membrane formation equipment, suggesting a scalable approach. Finally, combined BCP self-assembly and colloidal science concepts described here should be applicable not only to porous membrane formation but are expected to contribute to classes of bulk multifunctional BCM2based materials with advanced properties.

[0177] Methods.

[0178] Materials. All chemicals were used as received, unless otherwise stated.1,4- Dioxane (DOX, Sigma-Aldrich, anhydrous, 99.8%) and tetrahydrofuran (THF, Sigma- Aldrich, anhydrous, ^99.9%, inhibitor-free) were used to prepare the casting dope. Bond-breaker TCEP solution (Thermo Scientific), dimethyl sulfoxide (DMSO, MP Biomedicals), sulfo-Cy5(-) maleimide dye (Lumiprobe), β-mercaptoethanol, glucose, and glucose oxidase (Sigma Aldrich) were used to prepare samples for super-resolution imaging. Sodium acetate (Mallinckrodt Chemicals, anhydrous, ACS), acetic acid (Macron, ACS), hydrochloric acid (EMD Millipore, ACS), imidazole (Sigma-Aldrich, ACS, ^99%), sodium carbonate (Research Products International, anhydrous, ACS), and sodium bicarbonate (Ward’s Science, Lab Grade) were used to prepare buffers for permeability tests. Deionized (DI) water (18.2 MΩ-cm) was generated using a Barnstead NANOpure Diamond water purification system.

[0179] Polymer Synthesis and Characterization. Poly(isoprene-b-styrene-b-4- vinylpyridine) (PI-b-PS-b-P4VP, ISV), poly(styrene-b-4-vinylpyridine-b-propylene sulfide) (PS-b-P4VP-b-PPS, SVPS), and poly(isoprene-b-styrene-b-(N,N-dimethylamino)ethyl methacrylate) (PI-b-PS-b-PDMAEMA, ISA) triblock terpolymers were synthesized via a sequential anionic polymerization technique as reported elsewhere. The polydispersity index (PDI) of each block copolymer (BCP) was determined via gel permeation chromatography (GPC) on a Waters ambient-temperature GPC equipped with a Waters 410 differential refractive index (RI) detector using THF as the solvent. Total molar mass was determined using a combination of GPC and1H nuclear magnetic resonance (1H NMR). The molar mass of the first block was determined via GPC using PI (ISV and ISA) or PS (SVPS) standards. Molar ratios of the blocks were obtained via1H NMR analysis on a Bruker AV-500 NMR spectrometer using chloroform-d (D, 99.8%, Cambridge Isotope Laboratories) as the solvent. Polymer characterization is summarized in Table 1.

[0180] Dope Preparation. Casting dopes were prepared by individually dissolving BCPs in a solvent mixture of DOX and THF and stirring until homogeneous (~ 2 hours). BCP concentrations and DOX:THF weight ratios were chosen based on micelle formation, order onset concentration, and viscosity considerations. Casting dopes containing more than one BCP were prepared by mixing separate BCP solutions (by wt.) and stirring at 200 rpm for 15 minutes prior to casting. Dope concentrations and compositions of samples are summarized in Table 2.

[0181] Membrane Fabrication. Pure and blended membranes were fabricated via the SNIPS and M2SNIPS processes, respectively. The casting dope was pipetted onto a glass substrate, and cast into a thin film using an automated blade-casting machine with a pre- adjusted gate height between 0.203 mm and 0.229 mm. The thin film was allowed toevaporate for a set amount of time (60 to 100 s) to create a concentration gradient along the film normal and drive self-assembly on the film surface. Thereafter, the film was plunged into a non-solvent bath consisting of DI water, precipitating out the polymer and kinetically trapping the non-equilibrium state in the form of an asymmetrically porous membrane structure. The (M2)SNIPS process was conducted in a dry box at room temperature (19- 21°C). For most of this work, relative humidity was adjusted to 40-45% prior to casting by either purging with N2 gas or introducing water vapor. Low humidity (< 25 %) conditions were strictly achieved by purging with N2gas. Membranes for a particular series (i.e. evaporation time series, composition series) were all cast back-to-back on the same day to minimize the effects of day-to-day variations in environmental parameters.

[0182] SEM Analysis. Scanning electron microscopy (SEM) images were acquired via either a TESCAN MIRA3 FESEM equipped with an in-lens detector (accelerating voltage 5.0 kV) or a Zeiss Gemini 500 SEM (accelerating voltage 1.5-2.0 kV). Sample cross-sections were prepared under cryogenic conditions. All samples were coated with gold-palladium using a Denton Vacuum Desk II sputter coater prior to imaging.

[0183] SEM Image Processing.

[0184] General SEM Image Processing. SEM image processing was performed using the Fiji distribution of ImageJ. All images were brightness and contrast adjusted. The following image pre-processing protocol was used in preparation for segmentation using the Trainable Weka Segmentation (TWS) plugin—a machine learning tool for microscopy pixel classification. Images in the same series were obtained on the same instrument under the same imaging conditions. Following brightness / contrast adjustment, images were bandpass filtered (filtering out structures larger than 40 pixels or smaller than 3 pixels), autoscaled (adjusting lowest intensity to 0 and highest to 255) and saturated to produce better visual contrast (e.g., FIG.9b).

[0185] Pixel-Based Image Segmentation. The TWS graphical user interface was used to select image regions and assign those regions to segmentation classes. Images of pure materials were trained on two classes (i.e. ISV pore, matrix). Images of blended (ISV+SVPS) materials were trained on three classes (ISV pore, SVPS pore, matrix). Image features for classifier training were manually selected based on image contrast, visual differences in pore size, and the tendency for ISV and SVPS pores to pack into 2D square and 2D hexagonal geometries, respectively. Default classifier (FastRandomForest, 200 trees, 2 random features per node) and training features (Gaussian blur, Hessian, membrane projections, Sobel filter, and difference of gaussians) were used.

[0186] Post-TWS Adjustments. ‘Pores’ with area < 78.54 nm2(diameter < 10 nm, assuming perfectly circular pores) that arose due to minor contrast variations or image processing were manually removed to capture primary lattice spacing (compare FIG.9c,d). This value was chosen based on the pore size distributions of pure ISV and SVPS53membranes (FIG.12). The removal of these artifacts did not impact the observed experimental trends.

[0187] Given the pixel-based nature of TWS, minor gray value variations (i.e., near the pore edges) led to individual pores being classified as containing more than one class (red + green) (FIG.9c). The average gray values for three different regions (~2.5 μm x 2.5 μm) of each sample are portrayed in FIG.10 to capture sample uniformity and overall trends. Average gray values of 0 and 1 indicate that all pixels in a pore were classified as ISV and SVPS, respectively, while intermediate values suggest a mixture of classes as described previously. To correct for this, the class of each pore was adjusted to reflect the majority class, i.e., pores with average gray values of [0, 0.50] and [0.50, 1] were classified as ISV and SVPS, respectively. Adjustments were made using the ROI manager in ImageJ. As most pores fell within 5% of the extremes (i.e., [0, 0.05] or [0.95, 1]), this split did not modify observed experimental trends.

[0188] Quantification of Segmentation Results. The quantification of the ISV:SVPS ratio on the membrane top surface based on the segmentation results was performed on either a number (i.e., NISV_pores / Nall_pores) or area (i.e., AISV_pores / Aall_pores) basis. An example portraying the numerical relation between the two bases is shown in FIG.14. Quantification was performed on a minimum of three regions for each sample, and average values and corresponding standard deviations were reported. The pore screening and majority class adjustments detailed previously resulted in changes of < 3% (FIG.9c-e) and did not impact any observed experimental trends.

[0189] Fast Fourier Transform (FFT) Analysis. FFT analysis was performed on top surface SEM images using the ImageJ Radial Profile plugin. The following pre-processing protocol was used: (1) bandpass filtering (as detailed previously), (2) binarization by visually adjusting and setting the threshold, and (3) applying “find maxima” to reduce the pores to single points. Lattice spacing was calculated using d = 2π / q*, where q* is the first peak position.

[0190] Pore Size, Pore Density and Porosity Analysis. The ImageJ “analyze particles” function was called on binarized top surface SEM images to detect and characterize pores. Binarized images (FIG.9f) were generated using the probability maps (“get probability” onthe TWS graphic user interface) from the TWS segmented images. Pore area was converted to pore diameter, assuming perfectly circular pores. Porosity represents the ratio of total pore area to image area. Pore density was calculated by dividing the number of detected pores by the SEM image area. The tabulated averages and standard deviations (Tables 4 and 5) were obtained by analyzing three regions per sample.

[0191] Nearest Neighbor Distance Analysis on Segmented Images. Nearest neighbor distances were calculated using ImageJ’s Nearest Neighbor Distance (NND) plugin (Yuxiong Mao). To capture changes in the same-component lattice spacing (ex: ISV-ISV or SVPS- SVPS) in M2SNIPS membranes, segmented SEM images (the result of the Post-TWS Adjustments section, FIG.9e) were first split by class (ISV or SVPS, FIG.9g,h). The “analyze particles” function was used to generate a list of pore centroids, which served as the input for the NND plugin. This analysis was repeated for three separate membrane regions. Nearest neighbor distances were then plotted as histograms with bin size = 2 (FIG.19).

[0192] Bond Orientational Order and Voronoi Analysis. The square and hexagonal symmetries of the pure and blended ISV and SVPS membranes were characterized using the local packing arrangement obtained from Modified Voronoi Analysis (MVA) and the n-fold (where n= 4 or 6) bond orientational order parameter. FIG.15 illustrates the MVA methodology used to reveal the inherent local square symmetry for the pure ISV membrane by minimizing the error arising from the small lattice distortions typically present in the square lattice, and thus considering only the most representative neighboring pores. “Small” Voronoi cell edges obtained from the Voronoi tessellation analysis that subtend an angle δ < 30° with the pore center are excluded in the MVA and only edges with δ > 30° are considered to represent the nearest neighbors for the central pore. The δ value was optimized such that the perturbations due to the small Voronoi edges are screened without affecting the characteristics of the Voronoi cells having 4 and 6 nearest neighbors that correspond to ideal ISV and SVPS pore geometries, respectively. The MVA was particularly useful in identifying and characterizing the M2SNIPS membrane grain boundaries that are present between the square ISV and hexagonal SVPS domains.

[0193] The n-fold bond orientational order parameter was used to identify ISV and SVPS clusters that display a high order of 4-fold or 6-fold symmetry. The local n-fold bond- orientational order parameter ^^^^for the kthpore is given by: ^^ ^^where i =√−1, θjk is the angle between the vector connecting pore k with its neighbor j pores and a fixed reference unit vector, and n = 4 or 6. Nk is the number of nearest neighbors of the kthpore obtained from MVA. The nearest neighbors of the kthpore are defined as the j pores whose Voronoi cells share an edge with the cell of the kthpore. FIG.17 compares the ^^4and ^^6distributions for pure ISV and SVPS membranes with the distributions obtained for the square and hexagonal structures from Monte Carlo simulations (1600 particles) of hard squares and hard disks as reference. In FIG.20, the same-component lattice spacing distributions derived from the combined use of MVA and n-fold bond orientational order parameter restrictions were compared by plotting the kernel density estimate (KDE) (each a continuous probability density curve smoothed with a Gaussian kernel) using python’s Seaborn package.

[0194] Functionalized Dye Conjugation. SVPS-containing membranes were functionalized with sulfo-Cy5(-) maleimide dye via thiol-ene click chemistry to the terminal sulfhydryl groups of the PPS block (see green block in FIG.1a). Membranes (diameter = 1.8 cm) were soaked in 5 mM aqueous TCEP solution for 10 minutes to reduce disulfide bonds that may have formed during short-term sample storage. Sulfo-Cy5(-) maleimide dye was dissolved in a 0.1M pH3 sodium acetate-acetic acid buffer solution to create a 2.5 μM dye solution. Following disulfide reduction, each membrane was soaked in 5 mL dye solution for 7.5 hours. Membranes were subsequently washed with DMSO to remove non-covalently retained dye and stored in DI water for 3 days to allow residual non-covalently retained dye to diffuse out. DI water was changed daily.

[0195] Optical Super-Resolution Microscopy Imaging.

[0196] Total internal reflection fluorescence microscopy (TIRFM) measurements: Single- particle and single-molecule measurements were performed using an inverted Olympus IX81 microscope operated in TIRF geometry. BCP membranes were transferred to a container containing 1 mL of DI water combined with 5% (w / v) glucose, 50 μL glucose oxidase (10 mg / mL in PBS), and 20 μL catalase (2 mg / mL in PBS) as an enzymatic oxygen scavenging system. After several minutes, the membrane was dipped in β-merpcaptoethanol and sandwiched between two 1.5 cover glasses mounted in an Attofluor cell chamber (ThermoFisher Scientific) and placed on the microscope. Samples were placed on an 100x oil-immersion objective and simultaneously excited by a 561-nm laser and a 405-nm laser. Fluorescence signal was spectrally filtered using a 561-nm bandpass filter and 10,000 frames at a resolution of 50 ms / frame were recorded with an Andor iXon EMCCD camera.

[0197] Stochastic Optical Reconstruction Microscopy (STORM) Image Reconstruction. Localizations were fit with a Gaussian point spread function (PSF) and reconstructed super- resolution STORM images were obtained using the ThunderSTORM ImageJ plugin.

[0198] Reactive Ion Etching. Reactive ion etching (RIE) was carried out in an Oxford Plasmalab 80+ Reactive Ion Etcher system. An as-cast 1 cm x 1 cm membrane, separation layer facing upwards, was mounted onto a silicon substrate with carbon tape and allowed to dry. The membrane top surface was partially covered with a silicon wafer (thickness: 500- 550 μm) and etched in O2plasma for 30 s (150 W, 50 sccm, 60 mTorr). Shadowing from the silicon wafer created a gradient in etch depth by the membrane-wafer interface.30 s was the minimal time needed to etch through the ~100 nm membrane top separation layer (see FIG. 5b for etching geometry). The upper silicon wafer was removed prior to SEM characterization.

[0199] Characterization of Etched Membranes. SEM images of 2.5 μm x 2.5 μm regions were taken at regular intervals across the covered / uncovered membrane-wafer boundary to capture structural changes throughout the three to four ordered spherical micelle layers constituting the membrane top separation layer. TWS was trained on three classes (ISV matrix, SVPS matrix, and pore space), depicted in red, green, and magenta, respectively, to provide a crude visualization of component distributions. Classifier and training features were as described previously. Regions from both the “patchy” series (FIG.25) and “disordered” series (FIG.26) were used to generate the training dataset that was subsequently used to segment all images (FIG.27).

[0200] Membrane Performance Tests. pH-dependent permeability tests on pure and blended membranes were performed using a 10 mL dead-end stirred cell (Amicon 8010, Millipore, active area 4.1 cm2) pressurized to ~0.07 bar (1 psi) with N2 gas. To prevent damage from the stirred cell during testing, all membranes were placed on top of 0.2 μm Nylon supports (Sterlitech). These supports were previously shown to induce minimal changes in permeability across the relevant pH range (pH 3-8), and thus changes in permeability as a function of pH are primarily attributable to the membranes themselves. 0.1M buffer solutions of sodium acetate-acetic acid (below pH 6), imidazole-hydrochloric acid (pH 6-8), and sodium carbonate-sodium bicarbonate (pH 9-10) were prepared and tested with a pH meter prior to performance testing. Average permeability values and standard deviations were calculated from three replicate measurements performed at each pH under the same applied pressure. Absolute permeability values are expressed in units of LMH / bar (LMH: liters per square meter per hour). When relevant, pH-dependent permeabilities andstandard deviations were normalized to the highest permeability value (i.e., at pH 8) for each individual membrane.

[0201] Brownian Dynamics (BD) Simulation.

[0202] Colloidal coarse graining. Computational modeling of the ISV and SVPS micelle system is aimed at providing qualitative insight into the dynamic evolution and interlayer structural coordination during the M2SNIPS process inaccessible by experiments. Since simulating the self-assembly of BCPs into micelles and dynamically observing them on a scale comparable to experiments is computationally infeasible, the current work focused on coarse graining the system to a degree where the experimental timescale could be pragmatically albeit approximately probed (FIG.21).

[0203] Based on experiments suggesting negligible inter-micellar chain exchange on the timescale of the M2SNIPS process (FIG.7 and Supplementary Discussion 1), micelles were modelled as the basic dynamic units and the solvent described implicitly. Given the colloidal size scale of the micelles, their interactions were approximated using effective pair-wise potentials and their behavior modeled using simulation methodology typically employed to describe colloidal suspensions.

[0204] Particle interaction potential. Micelles were treated as colloidal particles, and their interaction potential chosen as follows. The target potential should: (i) be able to exhibit either square or hexagonal behavior at the interface under the same thermodynamic conditions, to thus observe structures consistent with those seen in experiments; (ii) exhibit a soft repulsive interaction tail, to allow the implicit polymer micelles to have some ability for penetration; (iii) be primarily repulsive in nature (the micelle coronas are dense and there is little propensity for inter-micellar chain exchange or aggregation in the presence of good solvents, implying that the effective interactions between micelles are repulsive), noting that different degrees of repulsiveness between colloidal species can still foster phase segregation (i.e., the least repulsive interaction translates into an effective relative attraction); (iv) be monotonic (BCPs are not expected to show any complex interactive patterns that would lead to non-monotonic potential of mean force interaction curves).

[0205] The Yoshida-Kamakura (YK) potential is a suitable choice for a potential that satisfies the above criteria. This potential (FIG.22) has the following form: ^^ ^^^^ ^^ −^^− −^ 2^^^ ^^

[0206] Here, ^^respectively, and ^^ is the softness parameter. These parameters control the structure of theself-assembled particles. Yoshida and Kamakura explored the role of a and the thermodynamic conditions on the resulting system structure. In the current work, ^^ = 3.0 led to square ordered particles (simple cubic, SC, lattice in the bulk) corresponding to ISV, and ^^ = 3.7 led to hexagonally ordered particles (body-centered cubic, BCC, lattice in the bulk), corresponding to SVPS.

[0207] To reflect the ISV and SVPS micelle size differences, a scaling size factor f was incorporated into the potential as follows: ^^ ^^( )= ^^ exp { ^^ (1 −^^^^ ^^) − 6 (1 −^^ 2^^ ^^ ^^ ^^) ln (^^^^ ^^)}(3)

[0208] based on the x-ray number (Nagg,max; i.e., assuming no solvent penetration into the micelle core) of each, leading to values of f = 1.0 (Nagg,max= 193) and f = 0.64 (Nagg,max= 117) for ISV and SVPS, respectively (data not shown).

[0209] To describe mixed ISV-SVPS interactions, the classical geometric mixing rule in both a and f was implemented as follows: ^^12=√^^1× ^^2^^12= √ ^^1× ^^2(4)

[0210] TheSVPS coronas were accounted for in the mixing value of ^^, called ^^^^ ^^ ^^. For the current work, ^^^^ ^^ ^^= 1.2 was found to be optimal, leading to the final potential form: ^^ ^^ 2^^(1 −^^ ^^)− 6(1 −^^ ^^)^^^^^^ ^^,12= ^^^^ ^^ ^^{12ln(^^ ^^)}

[0211] layer structural transition in the experimental system, an effective model of the interface and its interactions with the different micelles was employed.

[0212] The interface was modeled by introducing repulsive walls in the z direction that repel both ISV and SVPS, precluding their diffusion into the gas phase. These walls are soft quadratic potentials (effectively half-harmonic potentials, keeping only the repulsive half) that model the effects of an interfacial barrier on the micelles. In functional form, the potentials are: ^^^^^^^^^^ / ^^ ^^ ^^ ^^= ^^^^ ^^+ ^^^^ ^^ ^^ / ^^ ^^ ^^ ^^(6)where ^^ is the distance between the particle and the wall, ^^ sets the strength of the repulsion and ^^ is the Heaviside function, which effectively cuts the wall potential at a set distance ^^^^.

[0213] Each potential is composed of two walls: (i) a highly repulsive static potential, denoted by ^^ ^^, whose only role is to prevent the particles from escaping the simulation box, and (ii) a dynamic potential that reflects the changing interfacial energy affecting the different particle species (which in experiments arises from changes in the humidity and temperature of the gaseous layer above). This is achieved by dynamically varying the value of ^^^^ ^^ ^^ / ^^ ^^ ^^ ^^(= ^^^^ ^^ ^^ / ^^^^ ^^ ^^ ^^) during the simulation.

[0214] When the walls are introduced to the system, ^^^^^^^^ ^^= ^^^^^^^^ ^^ ^^(i: initial) was set and the system was allowed to evolve to a converged state. In this state, and for a wide range of initial ^^ values, ISV micelles completely populate the interface, and the surface is square ordered. Then, the value of ^^^^ ^^ ^^is gradually increased while the value of ^^^^ ^^ ^^ ^^is gradually decreased to approximate the effect of changing the interface selectivity for SVPS vs. ISV micelles at the film-air interface during solvent evaporation. During this dynamic change, SVPS particles slowly come up to the surface. Finally, the simulation continues for a third period where the ^^^^ ^^ ^^ / ^^ ^^ ^^ ^^value is left unchanged ( ^^^^^^^^^^and ^^^^^^^^ ^^ ^^are constant (f: final), reflecting that interfacial properties have reached a steady state). FIG.23 depicts this process. Values for the interfacial properties were chosen to best match the dynamic trends of the experimental surface composition. The parameter values used were: ^^^^^^^^= 0.5; ^^^^^^^^^^= ^^^^^^^^^^ ^^= 2.0 (the diameter of two ISV particles); ^^^^^^^^ ^^= ^^^^^^^^ ^^ ^^= 0.2; ^^^^^^^^^^= 0.9; ^^^^^^^^^^ ^^= 0.0.

[0215] Simulation process. BD solvent-implicit simulations were performed using the Large Scale Atomic / Molecular Massively Parallel Simulator (LAMMPS) software. A total of 300,000 particles was simulated, divided between ISV and SVPS according to experimental compositions. This led to interfacial layers (FIG.24) with 6,000 to 10,000 particles, comparable to the number of pores in a 2.5 x 2.5 μm membrane top surface SEM image.

[0216] All values in this section are in Lennard Jones units as implemented in LAMMPS55. The particles were first initialized at a low concentration. They were subjected to a Langevin thermostat with a temperature of 0.025 and pressurized using a Berendsen thermostat from a pressure of 1.0 to 5.0 over 5 ^105timesteps using the standard velocity- Verlet algorithm with a timestep of 0.05. These values were chosen to reach the thermodynamic conditions where the YK potential results in the desired self-organized bulk structure. This initial run was done with fully periodic boundaries, with the goal of randomizing the system.

[0217] The particle coordinates were passed to a second LAMMPS script which introduces walls in the z direction while the x and y directions remain periodic. The system is exposed to the initial wall properties and is annealed by first heating it to a temperature of 0.05 and then cooling to a temperature of 0.02. This run is 106steps long with a timestep of 0.01 to allow the system to reach a stable structure with the walls.

[0218] At this point, the simulation has reached a point comparable to the initial ordered M2SNIPS membrane top surface (60 s, FIG.3f), and further solvent evaporation is modeled by dynamically varying the values of ^^, as described in the previous section. This second run, at constant temperature, is 5 ^105steps long with a timestep of 0.05. Thereafter, a final run at constant ^^ for 106steps is implemented to equilibrate to a final structure, comparable to the M2SNIPS membrane top surface evaporated for 100 s (FIG.3j).

[0219] Simulation analysis. During simulation, particle positions are stored every 2,500 timesteps and the resulting trajectory file is read using the pyscal python module for analysis. The positions of the particles in the first two layers are retrieved by taking the histogram of the particles as a function of Z and finding the limits that define the first two layers (FIG.24, right).

[0220] The first layer structure was analyzed with MVA (FIG.3k-o). To better fit the minima observed in the simulations, the MVA angular constraint (δ) was modified from 30° (experimental) to 35°. The first two layers were analyzed concurrently to find instances representative of the mechanism for the sublayer-to-surface transition of SVPS micelles (FIG. 4a-e).

[0221] Supplementary Discussion 1: Effect of stir time on M2SNIPS membrane top surface structure.

[0222] After 15 minutes of stirring (the protocol followed for the rest of this work; FIG. 7a,b), the top surface packing of the blended membrane closely resembles that of a pure ISV membrane in terms of symmetry (2D square) and domain spacing (compare black and blue curves in FIG.7g), consistent with an ISV dominated blended membrane top surface. This suggests that 15 minutes of blending has not resulted in a noticeable micelle size change, i.e., chain exchange between micelle species is negligible. Membranes casted from solutions stirred for an hour (FIG.7c,d) show similar behavior. In contrast, after 33 days of stirring (FIG.7e,f), the top surface packing more closely resembles the 2D hexagonal packing of a pure SVPS membrane. However, the large disparity in domain spacing (25.1 nm vs.36.2 nm, FIG.7g) is not consistent with an SVPS53 dominated surface. This discrepancy can berationalized assuming that this represents a structure consisting of micelles that have undergone significant chain exchange. A linear relationship between domain size and blend composition is known when forming blended micelles from two SV polymers differing in molar mass and / or relative block fractions. For this system, linear interpolation using the pure membrane domain spacings and the blend composition yields d = 38.8 nm, close to the observed spacing, d = 36.2 nm. Remaining deviations could be due to the distinct BCP chemistries and incomplete chain exchange.

[0223] Supplementary Discussion 2: Effect of solvent surface tension.

[0224] It was recently found that the effect of solvent surface tension can be dominant, such that the polymer species adsorbing the lower surface tension solvent is enriched at the surface4. Thereby, complete surface coverage with a 2 wt. % minority component was achieved, despite it having similar polymer surface tension to the majority component. To that end, solvent-polymer interaction parameters (χsolvent-polymer) were calculated based on Hansen solubility parameters (Table 3). Solubility parameters for polymers obtained from two separate sources led to inconclusive results regarding the polymers’ solvent adsorption preferences. While χDOX-polymer < χTHF-polymer for both datasets, the relative solvent affinity for a given polymer was different (e.g., χDOX-PS~ χTHF-PSversus χDOX-PS< χTHF-PS). EXAMPLE 2

[0225] This example provides examples of functionalized multiblock copolymer films of the present disclosure, and methods of making and using same.

[0226] Provided herein, for film surface self-assemblies, is how binary and ternary BCP micelle mixtures, frozen by phase inversion, can be used to generate porosity that, in combination with machine-learning assisted image segmentation, can allow component classification via scanning electron microscopy. Results supported by mechanistic insights from Voronoi analysis, cluster analysis and Monte Carlo / Brownian Dynamics simulations reveal rich and controllable structural behavior. Such multicomponent materials are expected to allow emergent properties for a range of applications. Provided are experimental and analytical strategies that allow for the characterization of alloyed BCP micelles to unlock the full potential of such multicomponent materials.

[0227] The current work considers model systems comprising three chemically distinct BCPs – poly(isoprene-b-styrene-b-4-vinylypridine) (PI-b-PS-b-P4VP, ISV), a novel tetrablock quarterpolymer poly(isoprene-b-styrene-b-2-vinylypridine-b-ethylene oxide) (PI-b- PS-b-P2VP-b-PEO, ISVO; synthetic protocol and characterization in FIG.35), andpoly(styrene-b-4-vinylypridine-b-propylene sulfide) (PS-b-P4VP-b-PPS, SVPS) – that are first individually assembled into micellar building blocks and then blended together to generate multicomponent (binary and ternary) assemblies. The constituents’ similar elemental compositions (i.e., all BCPs predominantly comprise carbon and nitrogen, with only trace amounts of oxygen (ISVO) and sulfur (SVPS)) render identification of large sample numbers based on elemental contrast in TEM challenging. Provided herein is an alternative approach facilitating structural analysis. The non-equilibrium nature of BCP micelle assembly was used to generate porosity via a combination of solvent evaporation induced self-assembly and phase inversion (SNIPS). The size and lattice geometry of porous domains, as visualized under scanning electron microscopy (SEM), can then be employed to distinguish the individual building blocks. Machine-learning (ML) assisted image segmentation techniques can provide streamlined quantification and visualization of surface structures, from which assembly rules are construed with the assistance of Voronoi analysis, cluster analysis and coarse-grained Brownian Dynamics (BD) simulations. This approach is first applied to the characterization of single-component (ISV, ISVO, SVPS) based top surfaces, before advancing on to their binary and ternary combinations, probing the effects of micelle-micelle interactions (neutral and repulsive), solution stoichiometry, and size asymmetry on film surface structure. Finally, findings are summarized in a ternary morphology diagram.

[0228] BCP micelles as building blocks for multicomponent colloidal assemblies. Above the critical micelle concentration, BCPs aggregate into micellar structures consisting of an outer corona and an inner core to minimize unfavorable polymer-solvent interactions. Recent work on BCP micelles from PS-b-PVP (P2VP or P4VP) or PI-b-PS-b-P4VP has converged on PS (or PI-b-PS) constituting the micelle corona and PVP constituting the core in binary solvent systems of 1,4-Dioxane (DOX) plus tetrahydrofuran (THF) or N,N- dimethylformamide (DMF) plus THF. Here, the short PEO (~ 8 kDa) and PPS (< 3 kDa) blocks in ISVO and SVPS, respectively, are presumed to reside in the core with the PVP. FIG.29A introduces schematic depictions of the three BCP micelle building blocks (ISV, ISVO, SVPS), where parts with different colors reflect distinct core and corona chemistries. Combinations of building block chemistries are chosen based on two main considerations. First, BCPs need to form the desired micellar assemblies in the same DOX+THF solvent system to allow for subsequent blending of the micellar building blocks. This can be approached by choosing the same chemical base for all BCPs (> 80 wt.% PS + PVP; see Table 6). Second, inclusion of functional ‘accessory’ blocks (PI, PEO, and PPS) provide differentiators in structural film surface characteristics (vide infra). At the same time, theyillustrate how synthetic tweaks of the same base system (PS + PVP) open potential pathways for post-modification of these materials, e.g., via thiol-ene click chemistry (PI, PPS) or further conversions (PEO). While the current work focuses on unraveling the design rules for these multicomponent soft colloidal assemblies, it thereby sets the stage for future diversification of these systems via functionalization.

[0229] Table 6. Block copolymer characterization. Number average molar mass (Mnin kDa), weight fractions of polyisoprene (fPI), polystyrene (fPS), poly(vinylpyridine) (fPVP), poly(propylene sulfide) (fPPS), and poly(ethylene oxide) (fPEO), and dispersity (Đ). Subscripts in the fPVP column indicate whether BCP contains poly(4-vinylpyridine) (P4VP) or poly(2- vinylpyridine) (P2VP). PolymerMn fPI fPS fPVP fPPS fPEO Đ ISV1210.20 0.66 0.14P4VP- - 1.10 ISV1380.20 0.67 0.13P4VP- - 1.18 ISV164 0.25 0.65 0.10P4VP - - 1.19 SVPS53 - 0.79 0.18P4VP 0.03 - 1.09 SVPS67 - 0.76 0.20P4VP 0.04 - 1.08 ISVO1650.21 0.59 0.15P2VP- 0.05 1.18

[0230] Polymer blends have the tendency to macrophase separate, as their non-zero Flory-Huggins interaction parameters (χ) and appreciable degrees of polymerization (N) lead to a high segregation strength (χN). Covalently linking the incompatible polymer blocks (e.g., in a BCP) limits the system to microphase separation, giving rise to self-assembly on the length scale of the polymer coils. FIG.29B schematically demonstrates how microphase separation within a single BCP micelle can be combined with macrophase separation between chemically distinct micelle populations to control the nanoarchitecture of the final film surface. Binary combinations of the three building blocks comprise one reference case (ISV + ISVO) where no macrophase separation is expected given the chemically identical corona blocks (χIS-IS = 0, and therefore χN = 0), and two cases (ISVO + SVPS and ISV + SVPS) where macrophase separation is expected due to the coronal incompatibility between the isoprene block in ISV(O) and the styrene block in SVPS (χI-S > 0, χI-SN >> 0). It is considered, based on the corona-core block fractions (Table 6), that the core blocks areshielded and do not contribute to the micelle-micelle interactions. Using the unique BCP signatures (e.g., domain size and lattice geometry) – dependent on number average molar mass (Mn) and the breakdown of the block volume fractions – the building block distribution can be read from the final film surface (vide infra). The insights from the binary assemblies are then applied to understanding ternary ISV + ISVO + SVPS assemblies.

[0231] ISV121, ISVO165, and SVPS53BCPs (subscripts denote overall BCP Mnin kg / mol; see Table 6) were individually dissolved in mixtures of DOX plus THF for a final solution concentration of 11-15 wt. % (details in Table 7). Therefrom, films were generated via the non-equilibrium SNIPS process. Briefly, solutions were doctor bladed into thin films from which solvent was allowed to evaporate for 80 s (unless otherwise stated), up-concentrating the film surface past the threshold for self-assembly of micelles into ordered lattices. The films were then plunged into a non-solvent deionized (DI) water bath that precipitated out the polymer, freezing in the non-equilibrium structure. Selective evaporation of the corona- selective THF over the core-selective DOX leads to an asymmetric solvent distribution in the post-evaporated film (i.e., solvent-swollen cores amidst denser, solvent-poor coronas). When this structure is subjected to the non-solvent bath, the exchange of solvent for non-solvent causes the core chains to retract and collapse against the corona chains, generating pores. Surface SEM micrographs of pure ISV121 (FIG.30A), ISVO165 (FIG.30B), and SVPS53 (FIG. 30C) films convey high contrast between these pores (dark gray / black) and the solid matrix (light gray). Given the elemental makeup of the building blocks (vide supra), a quasi- equilibrium fully evaporated solid film would lack such pore derived contrast, highlighting a benefit of these non-equilibrium processing conditions.

[0232] Table 7. Detailed characterization of all film samples. Solvent system composition (by wt.), solution stoichiometry (expressed as weight fraction of each component in solution), final solution concentration ([ ]), and surface composition (numerical basis; average and standard deviations of three separate sample regions). BCP(s) Figure panel DOX:fISV fISVO fSVPS [ ]surfaceTHF composition ISV1212A,D,H,I,4I,36A,40A,7:3 100 - - 11 -8:2 - - 100 15 - ISV121+ISVO165 3A,M,6,42C,F,G,H 75:25 74 26 - 11.5 82:18 ± 2 ISV121+ISVO1653B,6,42B,E,G,H,43C-F 75:25 50 50 - 12 77:23 ± 2ISVO165+SVPS533C,N,6,45A 8:2 - 75 25 13.5 45:55 ± 3 ISVO165+SVPS53 3D,6,45B 8:2 - 51 49 14 22:78 ± 1ISV121+SVPS533E,O,4B,E,6,45C,F,47B, 48A 72:28 77 - 23 11.9 60:40 ± 2ISV121+SVPS53 72:28 53 - 47 12.9 33:67 ± 1 3 1 1 ± 2 ± 2 ± 1 3 2 1 3 1in subsequent binary and ternary BCP micelle mixtures, the SEM images were segmented into two classes (pore + matrix) with a trainable random forest ML algorithm using the Trainable Weka Segmentation (TWS) Fiji plugin (FIG.30D-F, Methods). Here, the pores were colored corresponding to the micelle core chemistries (ISV: red, ISVO: blue, SVPS: green) they originated from. For simplicity, the solid matrix was depicted in magenta, irrespective of corona chemistry. FIG.30H shows the pore size distributions (PSDs) of pure ISV121 (20.2 ± 3.2 nm), ISVO165 (24.6 ± 3.7 nm), and SVPS53 (15.0 ± 2.6 nm; full dataset in FIG.36). For further qualitative and quantitative structural characterization, the segmented images were subjected to Fast Fourier Transform (FFT) analysis (FIG.30I; Methods). ISV121 and ISVO165lattices are both consistent with a 2D square geometry – the higher MnISVO165BCP forming the larger lattice (55.6 nm vs.42.5 nm) – supporting the assumption that the presence of the PEO block likely does not affect the V-core micelle structure (vide supra). Conversely, SVPS53 forms a much smaller lattice (26.0 nm) consistent with 2D hexagonal geometry. Thus, ISV121and ISVO165are best identified by their pore and / or lattice sizes, whereas SVPS53 can be detected by lattice geometry. To complement and rationalize experimental results, micelles were coarse-grained in BD simulations as spherical colloidal particles exhibiting the same packing geometries as their experimental counterparts (FIG. 30G; Methods; FIGS.37, 38, 39), with relative sizes approximated by the averages of theexperimental PSDs (i.e., a monodisperse population of ‘average’ sized particles; experiments: FIG.36, simulation scaling factors: Table 8), unless otherwise stated.

[0234] Table 8. Scaling factors (f) for monodisperse BD simulations and all MC simulations. fISVfISVOfSVPSISV121 – ISVO165 0.95 1.12 -ISV121– SVPS531.0 - 0.736 ISV138– SVPS531.0 - 0.751 ISV164– SVPS531.0 - 0.661

[0235] Negligible chain exchange between chemically distinct BCP micelle populations. Binary films were generated by combining the solutions of pre-formed micelles (e.g., adding an SVPS67 solution to an ISV121 solution) in the desired weight ratios, ambiently stirring the resulting solutions at 200 rpm for 15 minutes, and then casting as before. The goal was to obtain a well-blended micelle solution while minimizing BCP chain exchange between the distinct micelle populations, the latter a critical prerequisite for the micelle-as-building-block treatment. The long micelle core blocks (ranging from Ncore = 113 for SVPS53 to Ncore = 423 for ISVO165), relatively concentrated solutions, and room temperature processing conditions all suggest that molecular exchange should be negligible on the timescale of the experiment (~ 1 hour total). For confirmation, SEM segmentation analysis was employed to compare the surface structures of SNIPS films derived from the same 60:40 ISV121:SVPS67(by wt.) solution blended for 15 minutes. (FIG.40C) vs.3 days (FIG.40D). Unlike the two-class (pore-matrix) segmentation of the pure samples, here, user defined traces were used to train the ML algorithm on three classes (ISV, SVPS, and matrix; FIG.41 and Methods). The three- class segmentation inherently assumes negligible molecular exchange, justified by the visual identifiability of ISV and SVPS based on lattice symmetry. Surface compositions of the two films were comparable (53:47 ± 3 and 55:45 ± 2 ISV121:SVPS67 for 15 minutes. and 3 days of stirring, respectively), suggesting no appreciable change in component makeup. As a more rigorous test, FFT analysis was performed on the split-component segmentations (e.g., FIG. 40G and I; Methods). Comparison of lattice spacing and geometry between the two films and against the corresponding pure samples (FIG.40F and H) revealed no detectible differences. Altogether, these results suggest negligible chain exchange between micelles for up to 3 days of ambient stirring (>> the 15 minutes of stirring applied in the present work), validate thetreatment of micelles as base building blocks – since SVPS BCPs are the smallest of the three systems studied (i.e., SVPS < ISV < ISVO), chain exchange in micelle solutions involving the largest ISVO BCP are even less likely to occur– and justify the use of three-class segmentation where applicable.

[0236] Tuning surface structure in binary systems: building block stoichiometry and interaction potential. Solution stoichiometry provides a rational and accessible means of tailoring multicomponent structures. This experimental axis was probed starting with the ISV121+ ISVO165system, whose building blocks were expected to display the same surface affinity (both have ‘pink’ IS coronas) while interacting neutrally (χIS-ISN = 0) with one another. Micelle solutions were combined in ISV121:ISVO165weight ratios of 75:25 and 50:50. Films evaporated for 80 s were surface characterized via SEM (FIG.42C and B). To streamline discussion, all solution stoichiometries will be rounded to the nearest multiple of 5 (for exact weight ratios, see Table 7). With pore size being the only viable classification mechanism (vide supra), SEM micrographs were two-class segmented (pore-matrix; Methods and FIG.43) and the pores nominally classified as ISV121 (red) and ISVO165 (blue) based on a 24.5 nm threshold (e.g., diameter ≥ 24.5 nm ^ ISVO165) rationalized by comparing pure ISV121 / ISVO165PSDs (FIG.43A).surface compositions of the films derived from two stoichiometrically distinct solutions (75:25 vs.50:50 ISV121:ISVO165; FIG.31A and B) exhibited no substantial differences (82:18 ± 2 vs.77:23 ± 2 ISV121:ISVO165), and even adjusting the solution to 25:75 ISV121:ISVO165 by wt. yielded an ISVO-poor surface (64:36 ± 1 ISV121:ISVO165,FIG.42D). This depletion effect persisted following statistical corrections for segmentation and thresholding choices (FIG.42H and Supplementary Discussion 3). As the deviation from solution stoichiometry increases with ISVO165solution weight fraction (compare distance between blue and red lines in FIG.42H), slower system dynamics in the increased presence of the larger ISVO165BCP could be a factor. To that end, dynamic light scattering (DLS) correlograms (FIG.44A) derived from lower concentration solutions (3 wt.%) indeed show that the dynamics of the larger ISVO165 BCP is approximately an order of magnitude slower than that of ISV121, suggesting that the latter would outcompete the former in assembling at the surface (for further in-depth discussion see Supplementary Discussion 4). The presence of the larger ISVO micelles in the film, however, is reflected by the FFT derived lattice spacings (FIG.42G), which scale with the weighted average of the two building blocks (Table 9). Therefore, solution stoichiometry presents an experimental dial that can be used to tune the lattice size of two chemically distinct micelle building blocks sitting on a single lattice. Please note that this FFT analysis was performed on the combinedISV121+ISVO165lattice. This is distinct from the split-component FFT analysis discussed below.

[0237] Table 9. FFT derived lattice spacings (d) of pure and blended ISV121 + ISVO165 samples. Predicted lattice spacings (d*) for blended membranes are weighted averages of pure values. Reported averages and standard deviations from N=3. experimental ISV121 : ISVO165 predicted d* (nm) d (nm) 0:100 - 55.7 ± 1.3 27:73 52.2 52.1 ± 0.2 50:50 49.3 48.7 ± 0.4 74:26 46.2 46.9 ± 0.5 100:0 - 42.8 ± 0.8

[0238] To ascertain whether the neutrally interacting ISV121+ISVO165assemblies display signs of macrophase separation, the split-component FFT curves derived from segmentation results in FIG.31A were compared against the respective pure component FFT curves (FIG. 31M). The presence of just a single lattice size amongst the two blend components (d = 46.9 ± 0.5 nm; see peak positions of the two sandwiched middle curves in FIG 31M) that is in between the lattice sizes of the pure films (ISV: d = 42.8 ± 0.8 nm; ISVO: d = 55.7 ± 1.3 nm) suggests that the size difference between the ISV and ISVO building blocks is not large enough to drive phase segregation, and that they thus randomly occupy sites of the same intermediate square lattice. The BD simulation snapshot of the 75:25 ISV121:ISVO165system surface (FIG.31G) shows the minority ISVO component predominantly inserting itself mid- ISV-grain rather than at the grain boundaries, echoing the experimental observation. Interestingly, the simulated 50:50 ISV121:ISVO165 system surface (FIG.31H) exhibits compositional variation from FIG.31G, which, while inconsistent with experiment – likely due to the degree of coarse-graining performed – is more in-line with expectations. Local size-dependent component clustering observed in the simulations (one to three particles wide) is not considered macrophase separation here.

[0239] Unlike the binary ISV121:ISVO165 film surfaces, those of the binary ISVO165:SVPS53system exhibit visible phase segregation (FIG.31C and D; see FIG.45 forraw SEM micrographs), as expected from the repulsive coronal interactions, χIS-SN >> 0. Simultaneous surface presence of two components with distinct surface tensions (for ISVO and SVPS, γPI = 32 mJ / m2and γPS = 40 mJ / m2) is a signature of non-equilibrium systems. Surface compositions suggest positive correlation with solution stoichiometry (44:55 ± 3 and 22:78 ± 1 ISVO165:SVPS53 for FIG.31C and D, respectively). These features are – to a lesser extent – echoed in the BD simulations (FIG.31I and J). Macrophase separation here refers to the coexistence of two distinct lattice sizes or geometries. FIG.31N experimentally quantifies the former, depicting the ~ 58 nm and 27 nm lattices respectively formed by the blue ISVO and green SVPS components. The disordered nature of the ISVO component, marked by the lack of higher order FFT peaks and visually evident in FIG.31C and D, precludes confirmation of the latter. Considering the system’s higher mobility as compared to that of ISV121+ISVO165 (compare relative positions of ISV121 and SVPS53 curves in FIG.44A), the structural disorder is puzzling. This can, however, be rationalized by thinking of the smaller, more mobile components as structure directing agents. ISV and ISVO individually assume the same 2D square lattice geometry and have no enthalpic drive to segregate. Therefore, assembly of the former prompts that of the latter. Conversely, SVPS and ISVO prefer distinct lattice geometries (hex. vs. square), and therefore efficient assembly of the former traps the latter in a disordered state due to ISVO’s limited mobility in the presence of two competing ‘forces’: (i) assuming its inherent square packing and (ii) assimilating to SVPS’ hex. packing.

[0240] To test this hypothesis, in the third binary system, ISV121 micelles were combined with those of SVPS53, essentially a more mobile execution of the ISVO165+SVPS53assembly. Like before, experiments (FIG.31E and F; see FIG.45 for raw SEM micrographs) and BD simulations (FIG.31K and L) both exhibit macrophase separation. Now, however, the components’ more similar mobilities (FIG.44A) allow both to assume their native lattice geometries (FIG.31O). Notably, the SVPS53lattice size (29.5 nm) is larger than it is in the prior ISVO165+SVPS53 system (27.1 nm) and for pure SVPS53 (26.0 nm). This could be due to the strain field set up by the larger and highly ordered 43.8 nm ISV lattice, prompting the soft SVPS coronas to expand and assimilate, an effect that is likely more subdued in the presence of the disordered ISVO lattice.

[0241] Examining effects of micellar building block size asymmetry and dispersity in binary systems. Film surface structures for systems with repulsive interactions (χΝ >> 0) have thus far been a convolution of building block chemistry (ISVO+SVPS vs. ISV+SVPS) and size asymmetry ( ^^ = Mn,ISV(O) / Mn,SVPS). The more ordered ISV+SVPS system was therefore chosen to isolate effects of the latter. To that end, different Mn ISV and SVPS BCPswere combined in weight ratios of 75:25 ISV:SVPS, size ratios of ^^ = 1.8 – 3.1, evaporated for 80 s, and film surface structures characterized via SEM (see FIG.45 for raw SEM images). FIG.32A to D shows that SVPS surface prevalence scales with ^^. The simultaneous surface presence of two different surface energy (i.e., corona chemistry) entities again points towards non-equilibrium structures frozen under evolving environmental conditions (e.g., where initial conditions favor X, final conditions favor Y, but the system is deliberately frozen prematurely). As has been discussed elsewhere, solvent evaporation from a polymer film surface under ambient conditions comprises a polymer-air interface that is increasingly replaced by a polymer-water vapor interface as evaporative cooling leads to a local increase in relative humidity – sometimes to the point of condensation. As such, while initial interfacial conditions favor ISV (γPI < γPS; vide supra), final conditions favor SVPS (θwater-PI > θwater-PS), prompting environment-induced structural re-equilibration. In the context of the semi-concentrated SNIPS casting solutions, the surface prevalence of SVPS then becomes a measure of surface accessibility. As size asymmetry increases, the smaller entity (SVPS) can more readily insert itself into the interstices of the lattice formed by the larger entity (ISV), giving rise to a positive correlation between ^^ and SVPS surface composition (FIG.32E). Modified Voronoi analysis (MVA; Methods and FIG.46) of the segmentation derived pore centroids showed diminishing four-fold (ISV) correlation length with larger ^^ (FIG.47), consistent with expectation and segmentation results. The trend with size asymmetry persists even when the SVPS surface composition is normalized to its mol.% (as opposed to wt.%) in solution (FIG.48).

[0242] To substantiate the surface accessibility argument put forth, BD simulations reflecting the ^^ = 2.3, 2.6, and 3.1 experimental systems were conducted (FIG.32F to H). The ^^ = 1.8 system was excluded to keep one constituent (SVPS53) constant in this intersystem comparison. As evident, BD simulations from monodisperse micelles reflecting just the average pore size do not capture the experimental trend in FIG.32B to D, due to the marginally smaller pore size of ISV138(19.9 nm) vs. that of ISV121(20.2 nm; FIG.36). Independent of the experiments, these simulations do highlight that greater size asymmetry improves surface accessibility (i.e., the mixtures with the smallest / largest average ISV pore size have the least / most SVPS on their top surfaces). Another factor that could affect SVPS surface accessibility is the ordered lattice formation kinetics of ISV (e.g., fast lattice formation would make the surface harder to penetrate). Increasing polydispersity completely suppresses ordered lattices in hard-sphere colloids. While packing restrictions of polydispersesoft colloids are expected to be less stringent, minor differences in lattice formation kinetics could still have major implications for SVPS surface accessibility. Approximating the building block polydispersity by the second moment of the PSD (i.e., the standard deviation, ^^; FIG.36A to C), it is evident that for the three ISV building blocks studied, the polydispersity increases with Mn. This trend remains even when the ‘polydispersity’ is normalized to the mean (via the coefficient of variation,^^^^). To probe the effects of polydispersity in a computationally feasible manner, the experimental PSDs were approximated as the weighted combination of five discrete populations (FIG.32I; see Table 10 for simulation scaling factors). Simulation box snapshots of the corresponding polydisperse assemblies are shown in FIG.32J to L (quantitative data in FIG.49) – now conforming to the experimental trends – suggesting that the experimental outcomes are likely a convolution of size asymmetry and building block polydispersity. While hard spheres with negatively skewed distributions (skewness = third moment of size distribution) were previously found to form ordered lattices up to an order of magnitude slower than those with symmetric distributions (compare FIG.36A to C), combined experimental and simulation results (vide supra) suggest that this effect is outweighed by that of polydispersity.

[0243] From binary to ternary micelle mixtures: ternary surface morphology diagram. Having carefully studied surface assemblies of the individual binary micelle mixtures, it was investigated whether the trends observed there would inform the behavior of the ternary ISV121+SVPS53+ISVO165 system. Here, one of the most interesting questions was whether the stoichiometric variance of ISV121:ISVO165(i.e., tuning the lattice size of these two micelle building blocks sitting on a single cubic lattice; vide supra) could be used to modulate the effective size asymmetry, and thus surface structure, of the ternary system (i.e., generating a ternary analogue of FIG.32A to D). The series of component weight ratios in solution (FIG. 33E) was chosen to ensure sufficient representation of each population. Resulting film surface SEM segmentations are presented in FIG.33A to C (raw SEM images in FIG.45). Four-class segmentation (3 building blocks + solid matrix) was accomplished by combining the prior two- and three-class segmentation methods: components were first classified as SVPS, non-SVPS or matrix (three-class segmentation), after which non-SVPS components were split into ISV and ISVO based on the 24.5 nm size threshold (Methods and FIG.41). The general surface depletion of ISVO in the presence of ISV and minimal variation with solution composition (Table 7) echo the outcomes from the binary ISV-ISVO system. FFT analysis of the non-SVPS components (FIG.33D) suggests that while the average sizes of the2D square-packed lattices scale with ISVO solution presence, they are smaller than would be expected from a binary ISV+ISVO system with the same ISV:ISVO ratio (e.g., compare weighted averages, 46.6 nm, 49.3 nm, and 52.0 nm, for ISV:ISVO compositions in A, B, and C, respectively, with the values listed in FIG.33D). This result highlights the soft colloidal nature of these building blocks (e.g., the expansion and compression of the micelle coronas in response to the composition of the colloidal ‘soup’). To describe how square lattice size impacts the propensity for SVPS emergence (i.e., how vulnerable the square lattice is to disruption), cluster analysis was performed on the entire pore coordinate array to detect clusters with high local Φ4 (four-fold) orientational order (Methods and FIG.50). FIG.33F to H respectively depicts the ten largest four-fold (presumably ISV / ISVO) clusters detected in FIG.33A to C, while FIG.33I statistically summarizes the size distribution of the ten largest clusters across three sample regions (30 clusters per sample). As effective size asymmetry increases (i.e., from FIG.33A to C), the mean and median cluster sizes decrease, pointing towards an increased tendency for SVPS to insert itself interstitially, in line with what was observed for the binary ISV-SVPS systems. The substantial ISV / ISVO surface presence despite the increased effective size asymmetry (compare FIG.32C to D to FIG.33A to C) and system polydispersity (due to the square lattice comprising both ISV121and ISVO165) could be a result of the significant ISVO165 solution presence. ISVO165, with its drastically slower dynamics and larger hydrodynamic size (FIG.44), is likely preventing the ternary system from efficiently responding to the changing interfacial environment.

[0244] Binary and ternary results were summarized in a ternary surface morphology diagram (FIG.34), where the multicolored circles reflect the relative surface compositions of the three building blocks for each film sample (as illustrated by the inset), and regions of the diagram are color-coded based on the majority surface component following 80 s of evaporation. As depicted (by component wt. fraction), the smallest (SVPS53) and largest (ISVO165) building blocks respectively dominate the largest and smallest compositional regions. To investigate how far this experimental snapshot deviates from equilibrium, 2D equilibrium Monte Carlo (MC) simulations (Methods) were conducted as a point of comparison. FIG.34A to D depicts MC simulations whose numerical component ratios reflect those of the experimental surface compositions. Structurally, systems with neutral interactions closely resemble equilibrium structures (compare FIG.34A and FIG.31B), whereas those with repulsive interactions (FIG.34B to D) do not (e.g., compare FIG.34B and FIG.31F). In the latter case, the equilibrium structures display a larger extent of macrophase separation than do their experimental counterparts, which is consistent with thefilms being phase inverted (structurally frozen) following a short period of evaporation (80 s) – insufficient time for equilibration. The repulsive interactions serve as an example of the structural diversification accessible via non-equilibrium processing of ‘active’ colloidal assemblies. Nevertheless, there are two degrees of separation between the component weight ratios reported in this work and the number ratios that govern traditional colloidal crystal structures. While the weight-to-molar ratio conversion is straightforward (always at the expense of the larger Mn component), the molar-to-number conversion involves accurate knowledge of BCP aggregation numbers, which would require further in-depth experimentation. Following the general rule of thumb that aggregation number scales with the number of monomer units constituting the micelle core, this once again disadvantages the ISVO165 BCP the most (given its Mn and 20 wt.% core fraction; Table 6). Therefore, while the compositional space explored in this work likely leans heavily towards the SVPS-rich corner of a number-based ternary surface morphology diagram, exploring the ISVO-rich corner would likely result in dynamically hindered self-assembled structures, as is already evident in the 75:25 wt. % ISVO:SVPS (FIG.31C) and ISVO:ISV (FIG.42A) samples.

[0245] In summary, it was demonstrated how binary and ternary mixtures of chemically distinct BCP micelles could be – upon solvent evaporation induced assembly – treated as soft colloids whose surface structures are tailorable via traditional colloidal parameters including solution stoichiometry, neutral / repulsive colloidal interactions, size asymmetry and polydispersity. Film porosity derived from a non-equilibrium synthetic technique was used in conjunction with ML-assisted image segmentation analysis to overcome the lack of atomic contrast traditionally rendering component classification in such systems challenging. BD simulations were employed to test mechanistic surface structure formation hypotheses and untangle associated complex experimental phenomena. An 80 s evaporation time was chosen to generate multicomponent surface structures and showcase the classification abilities of the techniques described in this work (e.g., ML-assisted image segmentation, Voronoi analysis and cluster analysis). Design rules established here, and in future studies on such multicomponent BCP micelle assemblies, promise advanced non-equilibrium and equilibrium structures with emergent properties for applications ranging from coatings and membrane- based separations all the way to self-healing materials.

[0246] Materials.1,4-dioxane (DOX, Sigma-Aldrich, anhydrous, 99.8%), tetrahydrofuran (THF, Sigma-Aldrich, anhydrous, > 99.9%), chloroform (Macron Fine Chemicals, ≥ 99.8%), sec-butyllithium (sec-BuLi, Sigma-Aldrich, 1.4 M solution in cyclohexane) and potassium chloride (Alfa Aesar, now Thermo Scientific Chemicals, ultra dry, 99.95%) were used asreceived. For anionic polymerization, the following chemicals were used after purification: Benzene (Sigma-Aldrich, ≥ 99.9%), tetrahydrofuran (Fisher Chemical, OptimaTMfor HPLC), isoprene (Sigma-Aldrich, 99%) and ethylene oxide (Sigma-Aldrich, 99.5%) were distilled over n-butyllithium (n-BuLi, Acros Organics, now Thermo Scientific chemicals, 1.6 M solution in hexanes) at room temperature; 1,1-diphenylethylene (DPE, Thermo Scientific Chemicals, 98%) was distilled over n-BuLi at 110°C; Styrene (Sigma-Aldrich, ≥ 99%) was distilled over di-n-butylmagnesium (Sigma-Aldrich, 1M in ether and hexanes) and 4- vinylpyridine (4VP, Acros Organics, now Thermo Scientific Chemicals, 95%) over ethylaluminum dichloride (Sigma-Aldrich, 1.0 M in hexanes); 2-vinylpyridine (2VP, Sigma- Aldrich, 97%) and propylene sulfide (Merck, 96+%) were dried over calcium hydride (CaH2, Alfa Aesar, now Thermo Scientific Chemicals, 90-95%) overnight prior to distillation. Methanolic HCl (Supelco, 3 M HCl in methanol) was degassed prior to use. Deionized (DI) water (18.2 MΩ-cm) was freshly generated with a Barnstead water system directly before use as a non-solvent bath.

[0247] Methods: Experimental Work and Image Processing.

[0248] Polymer Synthesis and Characterization. Triblock terpolymers poly(isoprene-b- styrene-b-4-vinylpyridine) (PI-b-PS-b-P4VP, ISV) and poly(styrene-b-4-vinylpyridine-b- propylene sulfide) (PS-b-P4VP-b-PPS, SVPS) were synthesized using sequential anionic polymerization techniques described elsewhere. The novel tetrablock quarterpolymer poly(isoprene-b-styrene-b-2-vinylpyridine-b-ethylene oxide) (PI-b-PS-b-P2VP-b-PEO, ISVO) was synthesized for the first time using the following adapted anionic polymerization technique. The polymerization of isoprene was initiated by sec-BuLi in benzene at room temperature. After 6 hours, an aliquot was taken and terminated for GPC analysis. Distilled styrene was then added to the reactor and allowed to react overnight. Thereafter, living anionic chains were capped with DPE by adding a 10x molar excess (relative to sec-BuLi) before benzene was exchanged for THF through successive distillations. The reactor was cooled to -78°C and 2VP was added. After 1.5 hours, the PI-b-PS-b-P2VP chains were end- functionalized with excess ethylene oxide before quenching with methanolic HCl (forming PI-b-PS-b-P2VP-OH, ISVOH). After THF was removed via rotary evaporator, the ISVOH was redissolved in chloroform and washed with water three times before being precipitated in methanol. The vacuum dried (48 hours at 50°C) polymer was redissolved in THF, followed by the addition of excess potassium chloride. Re-initiation of PI-b-PS-b-P2VP-OH was achieved through the addition of potassium naphthalenide. Distilled ethylene oxide was added and allowed to react for 4 days at 50°C before the resulting polymer was quenchedwith methanolic HCl and worked up as previously described for ISVOH. The final product was quantitively characterized using a combination of THF gel permeation chromatography (GPC, Waters, equipped with a 410 differential refractive index detector) and1H-nuclear magnetic resonance (1H-NMR) spectroscopy (Bruker AV-500, solvent: CDCl3). For a summary of all individual BCP characteristics see Table 6.

[0249] Porous film preparation. Single and multicomponent porous films were generated via the combination of solvent evaporation induced block copolymer (BCP) self-assembly with non-solvent induced phase separation also referred to as SNIPS (self-assembly and non- solvent induced phase separation). To that end, BCP solutions (60-150 μL depending on viscosity) were pipetted onto glass substrates and spread into thin films by an automated blade casting system comprising a casting knife (Elcometer) and a step-motor-controlled stage (Velmex, Inc.) situated in a humidity-controlled dry box. For solutions containing ISVO, a 25 mm diameter hydrophilic polyester substrate (0.2 μm pore size, Sterlitech) was placed atop the glass substrate for additional mechanical stability. Before each cast, gate height was manually adjusted to 0.203 – 0.229 mm and dry box relative humidity purged to 40 – 45% (ambient temperature 18 – 20 °C) with dry nitrogen gas. The step motor was programmed to (1) cast at 1185 spins per second (sps) and (2) allow for 80 s of preferential solvent evaporation to drive BCP self-assembly near the film surface, before (3) submerging the partially evaporated film into a non-solvent DI water bath (~5 L) to precipitate out the polymer and generate porosity as described in the Example. Following precipitation (~ 5 minutes), films were dabbed dry and prepared for subsequent characterization. All solutions were prepared by individually dissolving respective BCPs in a DOX / THF mixture (detailed concentrations and solvent compositions in Table 7) and ambiently stirring at 300 rpm for a minimum of two hours. For multicomponent (N ≥ 2) materials, said BCP solutions were combined by weight and allowed to ambiently stir at 200 rpm for 15 minutes. prior to casting. Mixing time was chosen to minimize degree of inter-micellar chain exchange as outlined in the Example (e.g., chain exchange between an ISV and SVPS micelle). To demonstrate that 15 minutes << the timescale of chain-exchange, a 60:40 ISV121:SVPS67solution was stirred for 15 minutes, after which 100 μL was cast according to the above protocol. The remaining solution was sealed via parafilm and stirred at 200 rpm for 3 more days, at which point a further 100 μL was cast.

[0250] Scanning electron microscopy (SEM) characterization. Porous films (evaporated top surface facing upwards) were affixed to flat SEM stubs with double-sided carbon tape. To minimize effects of charging, material edges were painted with conductive carbon adhesivebefore samples were sputter coated with Au-Pd (7 seconds, Denton Desk II vacuum sputter coater). Imaging was performed on a Tescan Mira3 FESEM (5 kV accelerating voltage) equipped with an in-lens secondary electron detector.

[0251] SEM image analysis. All image analysis was performed using the FIJI distribution of ImageJ2. Unless otherwise stated, sample statistics (e.g., averages and standard deviations) were derived from three separate 2500 nm x 2500 nm sample regions. As a first step in isolating the regions of interest (pore space) from the solid matrix, images were bandpass filtered (3 < pixels < 40), autoscaled and saturated (e.g., FIG.41A,G,K). This image was then imported into the Trainable Weka Segmentation (TWS) plugin for pixel-based segmentation.

[0252] Pixel-based segmentation. For single component (e.g., pure ISV) and non- macrophase separated two-component ISV121+ISVO165 samples, the interactive user interface was used to manually label select sample regions as either pore space or matrix based on visual identification. Default segmentation settings were used to train a FastRandomForest classifier (200 trees and 2 random features per node, where the random features are chosen from Gaussian blur, Sobel filter, Hessian, difference of gaussians, and membrane projections). More sample regions were iteratively labeled until reasonable segmentation results were achieved (as determined by the user). Images stemming from the same (combination of) BCPs were segmented using the same trained classifier. To remove image artefacts (regions erroneously classified as pore space due to minor intensity perturbations), objects < 78.54 nm2(10 nm diameter, assuming perfect circularity) were removed for all segmentations (e.g., FIG.43D). For visualization purposes, images were color coded, with red, blue, green and magenta pixels representing ISV, ISVO, SVPS and matrix domains, respectively (see FIG.30D-F). For ISV121+ISVO165samples, visual inspection revealed that the segmentation had picked up on domains not part of the primary 2D square lattice (lattice spacing > 40 nm; see FIG.43D inset and associated description). To aid FFT analysis in capturing the effects of blend composition (i.e., x% ISV121 + (100-x)% ISVO165) on primary lattice spacing, an attempt was made to systematically remove these non-primary domains according to certain nearest neighboring distance (NND) and domain size criteria. The NND distribution of ISV121 (the smaller of the BCP pair; FIG.43B) was used as a reference (nearest neighbor distance plugin for FIJI) when designating the minimum NND cut-off (NND = 35 nm). To differentiate between the primary and non-primary domains fulfilling the above NND criterion, a 14 nm domain size threshold was chosen based on trial-and-error. Domains removed due to these criteria are outlined in red (FIG.43E). While this approach was conservative in nature (i.e., it did not remove all non-primary domains), the improvementwas sufficient for subsequent FFT analysis. Classification of the remaining ISV121+ISVO165sample domains as either ISV or ISVO was realized by applying a size threshold of 471.44 nm2(24.5 nm diameter, assuming perfect circularity; FIG.43F) derived from the domain size distributions of pure ISV or ISVO films. Statistics suggest that, on the experimental timescale (<< the timescale for chain-exchange), over 90% of the domains in the binary blend exceeding this threshold would be ISVO (FIG.43A).

[0253] For two- and three-component systems displaying evidence of macrophase separation (e.g., the presence of two distinct lattice spacings or lattice geometries), the interactive user interface was used to manually select traces of three classes (ISV(O), SVPS, and matrix) for classifier training. When selecting representative traces, ISV(O) and SVPS domains were visually identified on the bases of domain size (ISV(O) > SVPS), lattice geometry (2D square vs.2D hexagonal), and / or pixel value (ISV(O) domains were closer to the black end of the spectrum; see FIG.41A,G). Traces were iteratively added until satisfactory segmentation results (e.g., FIG.41B,H) were achieved. FIG.41C highlights the domains removed by size thresholding (remaining domains > 78.54 nm2), more clearly unveiling the underlying square / hexagonal lattices. Due to the pixel-based nature of TWS, the collection of pixels in a single domain could comprise more than one class (e.g., red domains with a few green pixels lining the edge, or vice versa). As chain-exchange was shown to be minimal on the experimental timescale (FIG.40), domains (micelles) should only reflect a single BCP chemistry (e.g., only ISV / red or only SVPS / green). To have the segmentation results reflect this, the TWS-output pixel-based segmentation was transformed to an object- based segmentation via a “winner-takes-all” approach (FIG.41F). Assigning individual red and green pixels values of 0 and 1, respectively, the mean value across an entire domain was taken. Thereafter, domains were recolored such that 0 ≤ ^^ ^^ ^^ ^^ ≤ 0.5 ∈ ^^ ^^ ^^ and 0.5 < ^^ ^^ ^^ ^^ ≤ 1 ∈ ^^ ^^ ^^ ^^ ^^ (FIG.41D). The vast majority (> 70%) of domains fall within 5% of the two extremes (FIG.41E), suggesting a clear pixel-based distinction between the ISV and SVPS classes. As such, the error introduced by this adjustment is minimal. An analogous approach was used for the ISVO165-SVPS53samples (see FIG.41H-J). For three-component systems, a combination of the ISV-ISVO and ISV-SVPS approaches was utilized. Starting with a bandpass filtered image (FIG.41K), the above ISV-SVPS approach yielded FIG.41L. Thereafter, all ISV domains > 471.44 nm2(24.5 nm diameter, assuming perfect circularity) were reassigned to ISVO (FIG.41M). Visual inspection revealed that domains not part of the primary square ISV / ISVO lattice (spacing > 40 nm) had been erroneously assigned as green (SVPS) (see circled sections of FIG.41M). These features, despite their size and contrast, arenot believed to be SVPS due to their existence in binary ISV-ISVO samples (compare square packed regions in FIG.41M with those in FIG.43D). Considering the native lattice spacings of SVPS (~ 26 nm) and ISV / ISVO (> 40 nm; from FIG.20I), the SVPS component was isolated from the square-packed components (analogous to FIG.40I), and domains whose NND exceeded 40 nm were removed, yielding FIG.41N.

[0254] Fast Fourier Transform (FFT). Starting with a binarized image (e.g., FIG.43E or, for split-component FFT analysis, a binarized version of FIG.40G), the maxima (prominence > 10.00, essentially the domain centroids) were found (Process >> Find Maxima…), and the output was used to run the Fiji FFT command. The resulting FFT was subjected to the radial profile plugin, yielding an intensity vs. pixel value plot, where the x-axis was subsequently spatially calibrated to nm-1. Lattice spacing was calculated using2 ^^^^∗ where q* is the position of the first correlation peak.

[0255] Domain size distribution. To obtain domain size distribution statistics, the “Analyze Particles…” function was run on a binarized image (e.g., FIG.43E). The output area (A) for each domain was transformed into a diameter value (D) assuming perfect circularity (i.e., ^^ = 2√^^^^ ).

[0256] scattering (DLS). DLS measurements were performed on aMalvern Zetasizer ZS equipped with a 4mW 633 nm laser and a 173° backscatter measurement angle setup.60 mg BCP was dissolved in 7:3 DOX:THF (~2 mL) under overnight stirring (final concentration: 3 wt.% BCP). Solutions were syringe filtered (0.45 μm), after which 1 mL was transferred to a lidded quartz cuvette for characterization. Due to its large micelle size, the ISVO165 solution could not be filtered as such. The viscosity and refractive index for the binary solvent mixture were calculated from pure solvent values at 25°C, yielding a viscosity of 0.88 cP (classic Arrhenius mixing rule) and a refractive index of 1.42 (volume-fraction mixing rule, Newton formula). Samples were allowed to equilibrate at 25°C for 120 s prior to measurement. Reported hydrodynamic diameter (DH) statistics were derived from the intensity-based distributions (e.g., FIG.44B) of three consecutive measurements (10-20 runs per measurement, automated by Zetasizer software). Correlation data was exported from the Zetasizer software and plotted without further modification (FIG. 44A).

[0257] Voronoi analysis. To obtain a detailed understanding of the square and hexagonal motifs in the ISV+SVPS film surface structures, the freud library was used to perform aModified Voronoi Analysis (MVA). Adaptations were made from the standard Voronoi tessellation analysis (VTA) to correct for the minor structural perturbations inherent to the square packed ISV121 structures. The input for MVA is the array of domain centroids (FIJI >> Process >> Find Maxima…) extracted from a binary segmentation of the film surface (TWS- generated output without class information; see FIG.46A). The insets in FIGS.46B and C demonstrate how the modified Voronoi cells are derived from the VTA. MVA excludes “small” Voronoi cell edges resulting from minor lattice perturbations (gray), treating only the remaining cell edges (black) as the central domain’s nearest neighbors. This transformation reveals the 4-fold symmetry (red cells in FIG.46C) underlying the predominantly 6-fold symmetry (green cells in FIG.46B) suggested by VTA. Small edges can be identified by the angle, δ, subtended by the cell edge to the central domain. After examining the δ distribution of pure ISV121 (lattice: 2D square) and SVPS67 (lattice: 2D hexagonal) top surfaces (FIG. 46D), δ =30° was identified as the cut-off that balances: (1) screening the small edges (δ ≤ 30°) in ISV Voronoi representations while (2) retaining the 6-fold symmetry formed by the SVPS domains. FIG.46E compares the ISV121 domains’ number of nearest neighbor distributions according to VTA and MVA (i.e., before and after the imposition of the δ =30° cut-off).

[0258] Cluster analysis. Cluster analysis was utilized to quantify the extent to which SVPS53disrupts differently sized ISV121+ISVO165square lattices (lattice size: domain spacing according to FFT analysis) on ternary ISV121+SVPS53+ISVO165 film surfaces. A more disrupted lattice is expected to consist of, on average, smaller 4-fold symmetry clusters (cluster: a collection of domains exhibiting strong structural correlation). To identify such clusters, the segmentation derived array of domain centroids (FIG.50A) was used in conjunction with its global radial distribution function g(r) (FIG.50B) to identify the four nearest neighbors of each domain (a radial distance of 100 nm was used to search for neighbors, from which the four nearest neighbors were selected). Thereafter, the local 4-fold bond orientational order parameter for each domain k, ^^4( ^^^^) (Eq.1), was calculated. 41i is the imaginary number andconnecting domain k to its neighbor j (see inset in FIG.50A). For each of the four nearest neighbors j of domain k, the dot product of the local 4-fold bond orientational vectors, d4(j,k)(Eq.2), was calculated to estimate their degree of correlation, with a higher value reflecting stronger correlation (distribution of all such pairs in FIG.50C). ^^4( ^^, ^^) = ^^4( ^^) ^^4∗( ^^) (2)

[0259] The * denotes the complex conjugate. Spatially connected clusters displaying 4- fold order were constructed by considering neighbors of k that satisfy d4(j,k) ≥ 0.60, and the number of nearest neighbor connections, Nc ≥ 2. FIG.50D shows the ten largest clusters color-coded based on their number of domains. The choices of d4(j,k) and Ncwere optimized to detect the connectivity of local 4-fold ordered clusters and also distinguish the cluster domains separated by grain boundaries.

[0260] Methods: Simulations.

[0261] Computational Coarse Graining. The goal of the computational analysis in this work was to provide structural insights into the self-assembly of the ISV-ISVO-SVPS micelle systems submitted to the SNIPS process. While one could approach this by fully modeling the assembly of unimers into micelles and subsequently into lattices, achieving lateral scales comparable to the experimental SEM images (e.g., thousands of micelles) is computationally infeasible. As such, a coarse-grained perspective was adopted to approximate system behavior.

[0262] Coarse graining assumed negligible molecular exchange between chemically distinct micelle populations on the SNIPS timescale (< 1 hr), as substantiated by FIG.40. Therefore, the BCP micelles should behave as soft, dynamical units whose internal polymer dynamics can be ignored. Given the size of the micelles, a colloidal perspective was utilized where effective pair potentials were used to simulate the interactions between the micellar units, and the solvent was treated implicitly (FIG.37). The dynamics of the system were probed using standard Brownian Dynamics (BD) methods.

[0263] Particle Interaction Potential. Micelles were treated as colloids whose interaction potential was chosen based on the following criteria. The chosen potential must: (1) allow assembly into square (for ISV / ISVO) and hexagonal (for SVPS) geometries; (2) allow for some degree of interpenetration to reflect the soft micellar interactions; (3) be primarily repulsive; (4) be monotonic, as there is no complex interactive pattern in the micelle block copolymers.

[0264] The Yoshida-Kamakura (YK) potential is a suitable choice that satisfies the above criteria. Its functional form is: ^ ^^ ^^ −^^− −^ 2^^ ^^

[0265] Here, ^^ and ^^ respectively denote the classical scales for energy and length, and ^^ is the softness parameter whose main role is to control whether the potential leads to square or hexagonal structures. The structural landscape of this potential across different thermodynamic conditions has been studied previously. In this work, ^^ = 3.0 and ^^ = 3.7 were respectively used to assemble square (ISV / ISVO) and hexagonal (SVPS) structures at the interface.

[0266] Different micelle sizes were accounted for by incorporating a scaling factor ^^ to the potential as follows: 2 ^^^^ ^^(^^)= ^^ exp { ^^ (1 −^^^^ ^^) − 6 (1 −^^ ^^^^ ^^) ln (^^ ^^)}(4)

[0267] Thisexperimental . are Table 8. To represent dispersity, a histogram with a wide bin size (bin size = 2; see FIG.32I) was used to capture the general shape of the pore size distribution. Each distribution was split into five weighted, discrete populations (μ - 2σ, μ - σ , μ, μ + σ, μ + 2σ) based on its average (μ) and standard deviation (σ). The associated size factors are listed in Table 10.

[0268] Table 10. Scaling factors (f) for polydisperse BD simulations. Left column represents the 5 discrete micelle populations used for each species as detailed in Methods. fISV121-SVPS53fISV138-SVPS53fISV164-SVPS53ISV 2 0.841 0.803 0.770 ISV 3 1.000 1.000 1.000 ISV 4 1.159 1.198 1.231 ISV 5 1.317 1.396 1.461 SVPS 1 0.496 0.506 0.446 SVPS 2 0.616 0.628 0.553 SVPS 3 0.736 0.751 0.661 SVPS 4 0.855 0.873 0.769 SVPS 5 0.975 0.995 0.877

[0269] To describe the interaction between chemically distinct micelles, classical geometric mixing rules were used: ^^12=√^^1× ^^2^^12= √ ^^1× ^^2

[0270] For micelle pairs with chemically distinct coronas (e.g., ISVO-SVPS, ISV-SVPS), the additional enthalpic repulsion is accounted for via a mixing value for ^^ that exceeds 1. For this work, the optimal value was found to be ^^^^ ^^ ^^= 1.075. FIG.38 shows representative examples of the potentials.

[0271] Interfacial Potential. Solvent evaporation during the SNIPS process drives changes in the interfacial (film-environment) conditions, prompting micelle rearrangement. This interface was approximated by introducing repulsive walls above and below the simulation box that repel all species. These walls follow a half-harmonic form, where only the repulsive part is applied. For each species in the simulation (distinct in either chemistry or size), the interfacial potential follows this functional form: ^^^^^^= ^^^^ ^^+ ^^^^(6) ^^ ^^ ^^ ^^ ^^ ^^ 2 2

[0272] sets the strength of the repulsion, ^^ is the scaling factor and ^^ is the Heaviside function, which serves the role of cutting the interaction at a set length ^^^^.

[0273] The potential for each component is composed of two walls. The first, denoted by ^^ ^^, is a hard and short repulsive potential whose exclusive role is to prevent the particles from leaving the simulation box. It is the same for all species. The second term, denoted by ^^, is a chemically specific potential that serves to model the conditions at the interface for each species. Species size is accounted for by scaling both the strength and the cutoff of each wall with the scaling factor ^^ for each species.

[0274] The strength of the second wall changes dynamically throughout the simulation to reflect the changes in the interfacial conditions (vide supra). At the start, it is the same for all chemical species (normalized by micelle size), but as the simulation goes on, it is increased for species expected to be repulsed by the interface (ISV and ISVO) and decreased for others (SVPS). This change drives the surface transition and leads to the observed structures. See FIG.39 for a schematic of the wall potential and its evolution.

[0275] The values for the interfacial properties were chosen to best match theexperimental evolution of the system. For this work, the following values were used: ^^ ^^ ^^^^ =0.5, ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ / ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^ ^^^^ = 2.0, ^^ ^^ ^^ = 20, ^^ ^^ = 0.2, ^^ ^^ = 1.1 and ^^ ^^ = 0.0.

[0276] Mass Calculations. As previously stated, experimental pore size distributions (PSDs) were used to determine the relative simulation sizes (f) of each BCP micelle. Additionally, to match simulation compositions to those in the experiments, relative micellemasses (m) were computed. For each simulation, unless stated otherwise, the mass of the largest component was set to 1.0. The following relation was then used to derive the masses of the remaining components: ^^1 = ^^2 × (^^ 31 ^^^^ ) ×182 ^^2( )

[0277] For the densities, thedensities and volume fractions of^^ ^^ ^^_121=^^ ^^,^^ ^^ ^^ ^^_165=^^ ^^^^^^ ^^ ^^ ^^_53= 1.069^^^^ ^^.component masses have been calculated, their number ratios can beon weight fraction of each component in the experimental solution (Table 7). Since there are three components ( ^^, ^^ and ^^), the following equation can be used to compute how much of each component, ^^^^, is needed: ^^ ^^ ^^^^ = ^^ ^^ ^^ ^^ ^^ ^^ ^^(9)

[0279] system (300,000 for BD and 1,500 for Monte Carlo (MC) simulations), ^^ corresponds to the mass of each component and ^^ is the goal mass percentage of each component.

[0280] Note that this equation fails if any ^^ is zero. In that case, it becomes a two- component problem, and the following equation applies: ^^ ^^^^ ^^ =^^^^(10)

[0281] For polydispersethe total number of each chemical species needed, and then these were split using the relative distributions for each size (e.g., FIG.32I).

[0282] Brownian Dynamics (BD) Simulations. The evolution of the system through BD simulations with implicit solvent was investigated. The simulations were performed on the Large-Scale Atomic / Molecular Massively Parallel Simulator (LAMMPS). For each simulation, the system was composed of 300,000 micelles of different species according to experimental conditions (vide supra). This size led to interfaces with 6,000-10,000 particles, comparable to the number of pores in a 2500 x 2500 nm SEM image of the membrane top surface.

[0283] All values presented below are in Lennard Jones units as implemented in LAMMPS. The particles are first initialized randomly at a low concentration in a fully periodic 3D system. The system is then subjected to a Langevin thermostat at a temperature of 0.025 and pressurized slowly from 1.0 to 5.0 using a Berendsen thermostat for 5 × 105timesteps of 0.05. These conditions were chosen to match the thermodynamic regime in which the YK potential reaches the desired structures.

[0284] Once equilibrated, the system is introduced to the interfacial walls in the z direction, while remaining periodic in the x and y directions. To equilibrate the systemfollowing the introduction of the walls, it is annealed at a temperature of 0.05 and cooled to0.02 for the dynamic interfacial run. This cooling run is for 1.5 × 106 steps with a timestepof 0.01. After this equilibration run, the timestep is increased to 0.05 and the dynamic interfacial changes start, with walls slowly evolving over 106steps. After this procedure is completed, the system undergoes a final equilibration run of 5 × 105steps.

[0285] During the dynamical wall evolution, particle positions are stored every 5,000 steps and the trajectory is read using the pyscal python module. The top layer composition is used for structural quantification and comparison to experiments. Due to the size dependent scaling of the wall potential (FIG.39A), which drives the system evolution, simulation times between systems representing different ISV BCPs combined with the same SVPS53 BCP are not comparable (i.e., because the different sized ISV components are all scaled to 1.0, the SVPS component ends up ranging from 0.661 to 0.751, even though it is experimentally the same BCP; Table 8). To remedy this, the ratio of average pore sizes from the PSD (FIG.36) were used to scale time. The time dependent changes in ISV surface composition before and after scaling are shown in FIG.49A and B, respectively. Scaling for both the monodisperse and polydisperse BD simulations used the ISV121+SVPS53system as a baseline. The characteristic time for the ISV121+SVPS53 BD simulation was chosen to match the ISV121+SVPS53experimental surface composition (FIG.32B and Table 7). Equivalent scaled times were chosen for the remaining systems (FIG.49B), resulting in the surface snapshots shown in FIG.32F to H and J to L.

[0286] Equilibrium Monte Carlo (MC) Simulations. In addition to the BD simulations, it was of interest to determine how close the experimentally observed surface structures – derived from a non-equilibrium process – resemble equilibrium structures.

[0287] To find the equilibrium behavior of the mixtures, classic metropolis NPT Monte Carlo (MC) simulations were performed for the ternary mixtures of ISV, ISVO and SVPS atdifferent compositions. The same coarse-grained model and interaction potential (as described above for BD simulations) were used. These simulations were done in 2D, modelling only the equilibrium monolayer morphologies of the corresponding systems.

[0288] For each simulation, the system starts in a random dilute configuration. The system is then equilibrated for 1.5 × 106MC cycles, each corresponding to a mixture of displacement, swap and area moves. Displacement moves are calibrated for a 40% acceptance rate, while area moves are calibrated to a 30% acceptance rate.

[0289] Supplementary Discussion 3: Calculated “corrected” ISVO surface fraction in ISV+ISVO films. Instating the 24.5 nm cut-off to distinguish between ISV and ISVO pores provides a straightforward way of comparing the relative number of ISVO pores on the film surface between different film compositions. This reveals that even when ISVO comprises more than 50% of the BCP solution composition (by wt.), its surface presence is still overshadowed by that of ISV (see FIG.34, Table 7). Given the knowledge that the FFT lattice spacing of these films reasonably reflects the BCP solution composition (Table 9), it is of interest to calculate the statistically corrected number fraction of ISVO pores on the film surface to investigate the extent to which ISVO is the minority surface component. To provide a better sense of how these corrections were made, one region of the 27 ISV121: 73 ISVO165sample will be used as an example. Segmentation results obtained following the protocol described in Methods revealed that the top surface (FIG.42D) comprised 651 ISVO domains (number of pores exceeding a 24.5 nm diameter, assuming perfect circularity). As demonstrated by the domain size distributions of pure ISV and ISVO (FIG.43A), only ~92% of the domains exceeding a 24.5 nm diameter can be attributed to ISVO (~ 598 domains), and that number only captures ~ 64% of the ISVO domain size distribution (total: ~ 936 ISVO domains). To ensure a fair pore size comparison between the samples displayed in FIG.42, all samples were segmented with the same classifier and post-processed identically. Nonetheless, closer inspection of FIG.42D reveals more missing domains (non-porous space that one would expect to be porous) than exhibited in FIG.42E and F. These domains, as revealed in FIG.42A, are not necessarily absent, but have lower grayscale contrast (e.g., due to imperfect lattice formation in the substantial presence of the dynamically hindered ISVO165), and consequently closely resemble the matrix. Therefore, instead of using the total domains as determined via segmentation to determine surface composition (1778 domains), the theoretical total number of domains was calculated using the dimensions of the image (2490 nm x 2490 nm), and the FFT derived domain spacing (51.9 nm), yielding a total of ~ 2304 domains. All ‘missing’ domains were attributed to ISV, given its smaller pore size, andtherefore higher likelihood of blending into the matrix. The theoretical ISVO numerical surface fraction according to these assumptions was ~ 41%, still much lower than the solution composition (73 wt.% relative to ISV). Comparing the slope of the theoretical ISVO surface fraction to that of y=x, reveals that the lag in surface composition increases with ISVO solution composition, suggesting the slower dynamics of the larger ISVO micelles (FIG.44) could be playing a role.

[0290] Supplementary Discussion 4: Interpretation of surface assemblies in the context of micelle movement away from the interface.

[0291] This Example discusses the assembly of film surface structures in the context of micelles moving towards the film-air interface (i.e., micelles trying to assemble at that interface). Alternatively, in the limit where the Péclet number (Pe) >> 1 (evaporation rate >> diffusion rate; applicable to the current system), the results could be interpreted in terms of micelle movement away from the film-air interface due to evaporation induced chemical potential gradients. Larger colloids are known to experience a stronger “downward” force (due to more excluded volume) that overcomes their slower diffusion, leading to a surface enriched in the smaller colloid, especially when the volume fraction of the smaller colloid is high. Although the experimental results support this narrative (i.e., the ISV surface coverage at ISV121:ISVO165 compositions of 75:25 and 50:50 is higher than that at 25:75), normalizing the surface fractions to the amount of ISV in solution suggests the opposite (i.e., the surface enrichment of ISV increases with decreasing volume fraction).

[0292] Although the present disclosure has been described with respect to one or more particular embodiment(s) and / or example(s), it will be understood that other embodiment(s) and / or example(s) of the present disclosure may be made without departing from the scope of the present disclosure.

Claims

CLAIMS:

1. A method of forming a porous functionalized multiblock copolymer film comprising: forming a first film from a deposition solution comprising two or more multiblock copolymers, wherein each multiblock copolymer comprises one or more corona-forming block(s) and one or more core-forming block(s), and i) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or ii) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s), wherein, a first multiblock copolymer is present in a plurality of first micelles and a second multiblock copolymer is present in a plurality of second micelles and there is no substantial exchange of the polymer chains of the first of the multiblock copolymer from the first micelles with the polymer chains of the second multiblock copolymer of the second micelles, a solvent system comprising two or more solvents, optionally, one or more homopolymer(s), optionally, one or more metal salt(s), optionally, one or more inorganic nanoparticle(s), and optionally, one or more additive(s); removing at least a portion of the solvent system from the first film to form a second film; and contacting the second film with a phase separation solvent system, such that the porous functionalized multiblock copolymer film is formed.

2. The method of claim 1, wherein the deposition solution is formed from a first multiblock copolymer solution comprising one or more first multiblock copolymer(s) and one or more first solvent(s) and a second multiblock copolymer solution comprising one or more second multiblock copolymer(s) and one or more second solvent(s), such there is no substantial exchange of the polymer chains from the first micelles with the polymer chains of the second micelles.

3. The method of claim 1, wherein ~χABN (the product of the Flory-Huggins interaction parameter and the degree of polymerization for A, a corona region of a first multiblock copolymer, and B, a corona region of a second multiblock copolymer) is about zero.

4. The method of claim 1, wherein ~χABN (the product of the Flory-Huggins interaction parameter and the degree of polymerization for A, a corona region of a first multiblock copolymer, and B, a corona region of a second multiblock copolymer) is greater than about zero.

5. The method of claim 1, wherein ~χABN (the product of the Flory-Huggins interaction parameter and the degree of polymerization for A, a corona region of a first multiblock copolymer, and B, a corona region of a second multiblock copolymer) is less than about zero.

6. The method of claim 1, wherein at least a portion, substantially all, or all of the plurality of first micelles and at least a portion, substantially all, or all of the plurality of second micelles exhibit attraction.

7. The method of claim 1, wherein at least a portion, substantially all, or all of the plurality of first micelles and at least a portion, substantially all, or all of the plurality of second micelles exhibit repulsion.

8. The method of claim 1, wherein at least a portion, substantially all, or all of the plurality of first micelles and at least a portion, substantially all, or all of the plurality of second micelles do not exhibit attraction or repulsion.

9. The method of claim 1, wherein at least one multiblock copolymer or all of the multiblock copolymers comprises one or more hydrogen-bonding block(s).

10. The method of claim 1, wherein at least one multiblock copolymer or all of the multiblock copolymers further comprises one or more low Tg block(s).

11. The method of claim 1, wherein the homopolymer(s) is / are chosen from polysulfanes, polypropylenes, polylactic acids, polysulfones, poly(ether sulfone)s, poly(phenyl sulfone)s,cellulose acetates, poly(acrylonitrile)s, poly(amide)s, structural analogs thereof, and any combination thereof.

12. The method of claim 1, wherein the one or more multiblock copolymer(s) comprise about 10 percent by weight or less of the deposition solution (based on the total weight of multiblock copolymer(s) and homopolymer(s), if present, in the deposition solution) and / or the one or more homopolymer(s) comprise about 90 percent by weight or more of the deposition solution (based on the total weight of multiblock copolymer(s) and homopolymer(s), if present, in the deposition solution).

13. The method of claim 1, wherein the deposition solution further comprises inorganic nanoparticles.

14. The method of claim 1, wherein the concentration of the multiblock copolymer(s) and the homopolymer(s), if present, in the deposition solution is about 5 to about 30 percent by weight (based on the total weight of the solvents and the multiblock copolymer(s) and the homopolymer(s), if present).

15. The method of claim 1, wherein the solvent(s) is / are chosen from 1,4-dioxane, tetrahydrofuran, methanol, ethanol, toluene, chloroform, dimethylformamide, acetone, dimethylsulfoxide, N-methyl pyrrolidone, structural analogs thereof, and any combination thereof.

16. The method of claim 1, wherein the deposition solution further comprises one or more small molecule(s) and the film further comprises the one or more small molecule(s).

17. The method of claim 1, further comprising depositing one or more inorganic material(s) on at least a portion of a surface of the film.

18. The method of claim 17, wherein the inorganic material is a plurality of metal nanoparticles.

19. The method of claim 1, wherein the first film is formed on a substrate.

20. The method of claim 19, further comprising separating the film from the substrate to form a free-standing film.

21. The method of claim 1, wherein the two or more multiblock copolymers comprise polyisoprene-b-polystyrene-b-poly(4-vinylpyridine) (ISV) and / or one or more structural analog(s) thereof and polystyrene-b-poly(4-vinylpyridine)(P4VP)-b-propylene sulfide(PPS) (SVPS) and / or one or more structural analog(s) thereof, and, optionally, the solvents comprise 1,4-dioxane (DOX) and tetrahydrofuran (THF).

22. The method of claim 1, wherein the porous functionalized multiblock copolymer film comprises one or more or all of the following: (i) an isoporous region or layer thickness of about 20 nm (nm = nanometer(s)) to about 500 nm, a plurality of pores having a size of about 5 nm to about 100 nm and / or an asymmetric pore region or layer has a thickness of about 5 microns (micron(s) = micrometer(s)) to about 500 microns, and pores having a size of about 10 nm to about 50 microns; (ii) an isoporous region or layer pore density of at least about 1 x 1014pores / m2; (iii) an isoporous region or layer pore size distribution of about less than about 3, wherein the pore size distribution is the ratio of the maximum pore diameter (dmax) to the minimum pore diameter (dmin); (iv) at least a portion of the pores of the asymmetric pore region or layer increases in size along a dimension of the film moving away from a surface of the isoporous region or layer in contact with the asymmetric pore region or layer towards a surface of the asymmetric pore region or layer in contact with the substrate; (v) a plurality of the pores in the asymmetric pore region or layer lying on a plane about 20 microns from the surface layer and parallel to interface between the isoporous region or layer and asymmetric pore region or layer have a size in the plane of at least about 5 microns; (vi) at least a portion of the pores in the asymmetric pore region or layer have an aspect ratio of greater than about 2; or (vii) at least a portion of the pore surface in the asymmetric pore region or layer is mesoporous.

23. A composition suitable for use as a deposition solution comprising: two or more multiblock copolymers, whereineach multiblock copolymer independently comprises one or more corona- forming block(s) and one or more core-forming block(s), and i) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or ii) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s), optionally, one or more homopolymer(s), optionally, two or more solvents, optionally, one or more metal salt(s); optionally, one or more inorganic nanoparticle(s), and optionally, one or more additive(s).

24. A film comprising at least two multiblock copolymers, wherein the film comprises: i) an isoporous region or layer comprising a) two or more multiblock copolymers, wherein each multiblock copolymer comprises one or more corona-forming block(s) and one or more core-forming block(s), and i) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) or ii) at least two of the multiblock copolymers comprise one or more different corona-forming block(s) and at least two of the multiblock copolymers comprise one or more different core-forming block(s), b) a thickness of about 20 nm to about 500 nm, ii) an asymmetric region or layer comprising a) a thickness of about 5 microns to about 500 microns, and b) pores, independently, having a size of about 5 nm to about 100 microns, wherein, the isoporous surface region is disposed on at least, substantially all, ora a portion of an exterior surface of the asymmetric pore region, at least a portion, substantially all, or all of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of the core blocks of a second multiblock copolymer coreis / are isotropically or statistically distributed in at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer, or is / are disposed in a plurality of distinct domains forming at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer, or form(s) a three-dimensional pore lattice in at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer, and at least a portion, substantially all, or all of the functional groups of the core blocks of a first multiblock copolymer and / or at least a portion, substantially all, or all of a functional group or functional groups of the core blocks of a second multiblock copolymer core are disposed on at least a portion, substantially all, or all of the pore wall surface(s) of the isoporous region or layer.

25. The film of claim 24, wherein the isoporous surface region or layer comprises a) a plurality of pores about 5 nm to about 100 nm in size, and / or b) a pore density of at least about 1×1014pores / m2.

26. The film of claim 24, wherein the pores of the isoporous region or layer have a pore size distribution of about less than 3, wherein the pore size distribution is the ratio of the maximum pore diameter (dmax) to the minimum pore diameter (dmin).

27. The film of claim 24, wherein the distinct domains, independently, have a linear dimension of about 10 nm to about 3 microns.

28. The film of claim 24, wherein the asymmetric region or layer comprises a) a “spongy” substructure or a “finger-like” asymmetric substructure, and b) a plurality of pores having a size of about 5 nm to about 100 microns.

29. The film of claim 24, wherein at least a portion of the pores of the asymmetric region or layer increases in size along a dimension of the film moving away from a surface of the isoporous region or layer in contact with the asymmetric substructure towards a surface of the asymmetric pore region or layer in contact with the substrate.

30. The film of claim 24, wherein a plurality of the pores in the asymmetric region or layer lying on a plane about 20 microns from the surface layer and parallel to interface between theisoporous region or layer and asymmetric substructure have a size in the plane of at least about 5 microns.

31. The film of claim 24, wherein at least a portion of the pores in the asymmetric region or layer comprise an aspect ratio of greater than about 2.

32. The film of claim 24, wherein at least a portion of the pore surface in the asymmetric substructure is mesoporous.

33. The film claim 24, wherein the asymmetric region or layer comprises one or more homopolymer(s).

34. The film of claim 33, wherein the asymmetric region or layer does not comprise about 10 percent by weight or more multiblock copolymer(s) (based on the total weight of multiblock copolymer(s) and homopolymer(s) in the film.

35. A device comprising one or more film(s) of claim 24.

36. The device of claim 35, wherein the device is a filtration device, a separation device, or a catalysis device.

Images