Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
GELS AND NANOCOMPOSITES CONTAINING ARAMID NANOFIBERS
Document Type and Number:
WIPO Patent Application WO/2017/116598
Kind Code:
A1
Abstract:
Branched aramid nanofibers (ANFs) form hydrogel or aerogels with highly porous frameworks having mechanical properties exceeding those from non-branched nanomaterials with higher elastic crystalline moduli. ANF 3-D percolating networks (3DPNs) are made into different shapes by controlling the assembly conditions. In an embodiment, polymers such as epoxy are impregnated into the 3DPNs through a spin-coating assisted gelation layer-by-layer deposition (scg LBL) process. The nanocomposite fabrication process enables control of the film structure and thickness with nanometer resolution.

Inventors:
KOTOV, Nicholas A. (3233 Andora Drive, Ypsilanti, Michigan, 48197, US)
ZHU, Jian (1309 Maple Avenue, #3NEvanston, Illinois, 60201, US)
TUNG, Siu on (1860 Lindsay Lane, Ann Arbor, Michigan, 48104, US)
Application Number:
US2016/064186
Publication Date:
July 06, 2017
Filing Date:
November 30, 2016
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
THE REGENTS OF THE UNIVERSITY OF MICHIGAN (Office of Technology Transfer, 1600 Huron Parkway2nd Floo, Ann Arbor Michigan, 48109-2590, US)
International Classes:
C08J3/075; C08J5/00; C08J5/04; C08K7/02; C08L63/00
Domestic Patent References:
WO2015127115A12015-08-27
Foreign References:
US20130288050A12013-10-31
CN103802411A2014-05-21
Other References:
SHAO, Z. ET AL.: "Preparation of p-aramid aerogels using supercritical C02", SEN'I GAKKAISHI, vol. 70, no. 10, 2014, pages 233 - 239
IIJIMA, M. ET AL.: "Non-aqueous colloidal processing route for fabrication of highly dispersed aramid nanofibers attached with Ag nanoparticles and their stability in epoxy matrixes", COLLOIDS AND SURFACES A: PHYSICOCHEMICAL AND ENGINEERING ASPECTS, vol. 482, 5 October 2015 (2015-10-05), pages 195 - 202, XP029559986
Attorney, Agent or Firm:
BRAIDWOOD, G. Christopher et al. (Lempia Summerfield Katz LLC, 20 South Clark Street Suite 60, Chicago IL, 60603, US)
Download PDF:
Claims:
CLAIMS

We claim:

1. A method comprising a. providing a suspension of aramid nanofibers in an aprotic solvent; b. phase transforming the suspension by diffusing water into the suspension of step a. to make a hydrogel comprising water and aramid nanofibers.

2. The method according to claim 1, wherein the aprotic solvent comprises dimethyl sulfoxide (DMSO). 3. The method according to claim 1, wherein the suspension comprises 0.1 - 5 weight % aramid nanofibers.

4. The method according to claim 1, further comprising removing the water from the hydrogel to make an aerogel comprising aramid nanofibers.

5. The method according to claim 4, wherein removing the water comprises extracting with supercritical carbon dioxide.

6. The method according to claim 1, wherein the method is carried out continuously.

7. A composite material comprising a polymer matrix and aramid nanofibers dispersed in the polymer matrix.

8. The composite material according to claim 7, comprising an aerogel and a polymer in the pores of the aerogel, wherein the aerogel comprises aramid nanofibers.

9. The composite material according to claim 7, wherein the polymer comprises a cured epoxy resin.

10. A method of making a composite by a gelation assisted layer-by-layer deposition process, the method comprising a. providing a coating of a dispersion on a substrate, wherein the dispersion comprises aramid nanofibers and an aprotic solvent; b. phase transforming the coating of a. with a protic solvent to form a 3-D percolating network (3DPN); c. penetrating precursors of a polymer dissolved in solvent or a solution of a formed polymer into the 3DPN of step b. d. polymerizing the polymer precursors in situ in the case where precursors of the polymer are used in step c; and e. removing the solvent from the product of step d., whereupon the 3DPN of ANF collapses and blends uniformly with the polymer.

11. The method according to claim 10, comprising repeating steps a. - e. one or more times to build up a desired thickness of composite and delaminating the composite.

12. The method according to claim 10, wherein the aprotic solvent comprises dimethyl sulfoxide.

13. The method according to claim 10, wherein the polymer precursors comprise and epoxy resin and an epoxy hardener.

14. The method according to claim 10, wherein removing the solvent comprises extracting with supercritical carbon dioxide.

15. The method according to claim 10, comprising a. spin coating an ANF dispersion in DMSO on a substrate b. exchanging DMSO in the dispersion with water to make a spun ANF thin film; c. removing excess water by spinning; d. spin coating an epoxy solution on top of the spun ANF thin film; and e. optionally repeating the above steps to build up a desired thickness.

16. A hydrogel comprising water and aramid nanofibers.

17. The hydrogel of claim 16, in the form of a fiber.

18. The hydrogel of claim 16, in the form of a thin sheet.

19. An aerogel comprising aramid nanofibers.

Description:
GELS AND NANOCOMPOSITES CONTAINING ARAMID NANOFIBERS

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 62/272,877, filed on December 30, 2015. The entire disclosure of the above application is incorporated herein by reference.

GOVERNMENT SUPPORT

[0001 ] This invention was made with government support under DE-SC0000957 awarded by the Department of Energy; DMR-9871177, DMR-0315633, CBET1036672, CBET0932823, CBET0933384, ECS0601345 and EFRI0938019 awarded by the National Science Foundation. The Government has certain rights in the invention.

INTRODUCTION

[0002] Suspensions of aramid nanofibers (ANF or ANFs) can now be obtained by controlled reaction of microfibers at low temperatures for extended times in aprotic solvent in the presence of a base. Making high performance nanocomposites with ANFs or other nanocomponents relies on forming reliable percolated networks of well-dispersed nanomaterials and easy incorporation of matrix materials. But such networks and methods are still difficult to achieve.

[0003] Layer-by-layer (LBL) assembly produces nanocomposites with distinctively high volume fractions of nanomaterials and uniformity. Although depositing one nanometer scale layer at a time leads to high performance composites, this deposition mode is also associated with the slow multilayer build-up and represents the fundamental challenge of LBL. Methods of making high performance nanocomposites containing ANFs would be an advance.

SUMMARY

[0004] Aramid nanofibers prepared under disclosed conditions have branched morphology and therefore can assemble into a robust interconnected structure to make gels and composites with mechanical properties better than other networks made of stiff er components. By changing the conditions of the assembly, the nanofiber network can be shaped differently and produced continuously. The adaptability of network making process enables a consistent way to fabricate nanocomposites with nanoscale level control with the assistance of spin- coating. Exemplary ANF/epoxy composites are transparent, and have an ultimate strength of 505+47 MPa, toughness of 50.1+ 9.8 MJ/m , and close-to-zero thermal expansion, beating the properties of quasi-isotropic epoxy composites reinforced by aramid microfibers (like Kevlar®) or carbon microfibers.

BRIEF DESCRIPTION OF THE DRAWINGS

[0005] FIG. 1 shows a schematic of gelation-assisted layer-by-layer deposition process. [0006] FIG. 2A shows a thin layer of hydrogel peeled off from the glass slide in water;

FIG. 2B shows the cross-section of a thin layer of aerogel obtained by C0 2 supercritical drying; FIG. 2C shows a zoomed region in FIG. 2B showing the porous nanofiber network; FIG. 2D shows linear growth of the film examined by UV-vis spectroscopy - the inset shows the absorbance at 330nm; FIG. 2E shows a linear increase of thickness at the concentrations of 0.1% and 0.2%; FIG. 2F shows a linear increase of thickness at the concentration of 1%.

[0007] FIGS. 3A, 3B, 3C, 3D, 3E, 3F, 3G, 3H, 31, 3J, and 3K have the following captions: FIG. 3 A: The transparent [1%ANF/1%EPX] 6 on glass slide. FIG. 3B: The freestanding [1%ANF/1%EPX] 6 wrapped on a pen. FIG. 3C: UV-vis spectrum of [1%ANF/1%EPX] 6 . FIG. 3D: Cross-section of [1%ANF] 6 . FIG. 3E: Cross-section of [1%ANF/0.1%EPX] 6 . FIG. 3F: Cross-section of [1%ANF/1%EPX] 6. FIG. 3G: Cross-section of [1%ANF/2%EPX] 6 . FIG. 3H: Thickness of the film with different ANF fractions. FIG. 31: AFM image of [l%ANF]i. FIG. 3 J: [l%ANF] i surface morphology by SEM. FIG. 3K: [l%ANF/2%EPX]i surface morphology by SEM.

[0008] FIGS. 4A, 4B, 4C, and 4D have the following captions: FIG. 4A: Stress-strain curves for the composites in this work. FIG. 4B: The specific strength comparison of [1%ANF/1%EPX] 6 with various metal alloys. FIG. 4C: [1%ANF/1%EPX] 6 in comparison with various carbon and aramid micro fiber reinforced composites 16 in terms of ultimate strength. ISO stands for quasi-isotropic here. FIG. 4D: Toughness comparison.

[0009] FIGS. 5A, 5B, 5C, and 5D have the following captions: A comparison of various properties for ANF/EPX composites made in this work: FIG. 5A: Ultimate strength. FIG. 5B: Storage Modulus. FIG. 5C: Toughness. FIG. 5D: Damping ratio.

[0010] FIG. 6 has the caption: Normalized length change dL/Lo vs. temperature for several ANF/EPX composites, ANF film, EPX and aramid microfiber.

[0011 ] FIGS. 7A, 7B, 7C, 7D, 7E, 7F, 7G, 7H, 71, and 7J illustrate preparation and characterization of branched ANF (ANF). FIG. 7A: Molecular structure of the PPTA chains. FIG. 7B: A schematic drawing of the hierarchical structure of Kevlar™ microfibers (KMF). FIG. 7C: An SEM image of fractured KMF showing the fibrils and constitutive nanofibers. FIG. 7D: AFM image of ANFs deposited on a silicon substrate. FIG. 7E: Statistical analysis of ANF diameters obtained from multiple AFM images. FIG. 7F: TEM images showing the branched ANF. FIG. 7G: SEM images of branched ANF. FIG. 7H, FIG. 71, and FIG. 7J: FTIR spectra for KMF and ANF. The stretching and bending modes of different functional groups are indicated by γ and δ respectively.

[0012] FIGS. 8A, 8B, 8C, 8D, 8E, 8F, 8G, 8H, 81, 8J, 8K, and 8L illustrate fabrication of hydrogel and aerogel of ANF. FIG. 8 A: ANF dispersion in DMSO. FIG. 8B: A schematic drawing of the solvent exchange process for the formation of ANF hydrogel. FIG. 8C, FIG, 8D, and FIG. 8E: Photographs of FIG. C: ANF hydrogel, FIG. D: pieces of hydrogels cut and immersed in fresh deionized water, and FIG. 8E: ANF aerogels wedged in an opening of a beaker. FIG. 8F and FIG. 8G: SEM images of ANF aerogel in different magnifications. FIG. 8H: A schematic showing the ANF fibrous hydrogel production process. FIG. 81: A photo of the fibrous hydrogel jetting form a capillary glass tube. FIG. 8J: The produced fibrous hydrogel collected in a beaker. FIG. 8K: The SEM image of the fibrous aerogel converted from hydrogel by supercritical C02 drying and FIG. 8L: the zoomed image showing the internal nanofiber networks.

[0013] FIGS. 9A, 9B, 9C, 9D, 9E, 9F, 9G, 9H, 91, 9J, 9K, 9L, and 9M have the following captions: FIG. 9A: Oscillatory shear strain-stress curve of ANF hydrogel. FIG. 9B: Dependence of elastic moduli (G') and loss moduli (G") on the strain amplitude. FIG. 9C: Dependence of G' and G" on the frequency of the oscillatory shear. FIG. 9D: A comparison between the ANF hydrogel and other hydrogels made of typical reinforcing nanomaterials with high crystalline elastic moduli. FIG. 9E: Compressive stress-strain curves for ANF hydrogel and aerogel. FIG. 9F: ANF aerogel before compression. FIG. 9G: The same aerogel after compression. FIG. 9H: Side view photograph of the compressed aerogel. FIG. 91, FIG. 9J, and FIG. 9K: SEM images of the edge of fractured compressed ANF aerogel at different magnifications. FIG. 9L: Tensile stress-strain curves for ANF hydrogel and aerogel. FIG. 9M: Tensile stress-strain curves for ANF aerogel after being compressed into 1/6 ώ and 1/18 th of the initial height.

[0014] FIGS. 10A, 10B, IOC, 10D, 10E, 10F, 10G, 10H, 101, 10J, 10K, 10L, 10M, 10N, 10O, 10P, and 10Q illustrate the farbication and characterization of ANF/EPX composites. FIG. 10A: The schematic of the scgLBL process. FIG. 10B: A thin layer of hydrogel peeled off from the glass slide and suspended in water. FIG. IOC: The cross-section of a thin layer of aerogel obtained by C0 2 supercritical drying. FIG. 10D: A zoomed region in FIG. 10B showing the porous nanofiber network. FIG. 10E: 1%ANF/1%EPX] 6 on glass slide. FIG. 10F: Freestanding [1%ANF/1%EPX] 6 . FIG. 10G: Linear growth of the film examined by UV-vis spectroscopy. The inset shows the plot of absorbance at 330 nm vs. layer numbers. Linear increase of the film thickness at the ANF and EPX concentrations of FIG. 10H 0.1% , 0.2% and FIG. 101 1% determined by ellipsometry. FIG. 10J: UV-vis spectrum of [1%ANF/1%EPX] 6 film. FIG. 10K: Thickness of the film with different ANF fractions. FIG. 10L, FIG. 10M, FIG. 10N, and FIG. 10O: SEM images of the cross-section of ANF/EPX composites with different ANF content. FIG. 10P and FIG. 10Q: Surface morphologies of [l%ANF/l%EPX]i and [l%ANF/2%EPX]i by SEM respectively. [0015] FIGS. 11A, 1 IB, 11C, 1 ID, 1 IE, and 1 IF illustrate the mechanical properties of

ANF/EPX composites. FIG. 11A: Uniaxial tensile stress-strain curves for the ANF composites. FIG. 11B: Ultimate strength and FIG. 11C: Toughness comparison for the ANF/EPX composites. FIG. 11D: Specific strength comparison of [1%ANF/1%EPX] 6 with various metal alloys. FIG. HE: Ultimate strength and FIG. 11F: toughness of [1%ANF/1%EPX] 6 in comparison with various carbon and aramid (Kevlar®) microfiber reinforced composites. ISO here stands for quasi-isotropic.

[0016] FIGS. 12A, 12B, 12C, 12D, 12E, 12F, 12G, and 12H have the caption: FIG.

12A: A piece of hydrogel sheet obtained in water. FIG. 12B: PVA/ANF composite film. FIG. 12C: The cross-section of a PVA/ANF composite film. FIG. 12D: The surface morphology of a PVA/ANF composite film. FIG. 12E: Transparency of PVA/ANF composite film with and without epoxy coating. FIG. 12F, FIG. 12G, and FIG. 12H: FTIR spectra for PVA, ANF, and PVA/ANF composites. The stretching and bending modes of different functional groups were indicated by γ and δ respectively.

[0017] FIG. 13A and FIG. 13B illustrate dynamic mechanical properties testing of ANF/EPX composites.

[0018] FIG. 14 shows a photograph of the coated anode after step 6 described in the

Example 18 working example. The top layer is the ANF composite (or ANF ICM), the dark layer underneath is the carbon electrode, the copper foil is the copper current collector. The carbon and copper collector are typical of a lithium ion battery anode. DESCRIPTION

Embodiments of the invention

[0019] In one embodiment, a hydrogel, in the form of a bulk solid, a fiber, or a thin sheet is made of water and aramid nanofibers, wherein the nanofibers form a 3-D percolating network (3DPN). In these and other embodiments, the terms aramid nanofiber(s) and ANF(s) are used interchangeably. Aerogels are also provided that are made of the aramid nanofibers. The aerogels can be made by removing the water from the hydrogel while maintaining the 3-D structure. In various embodiments, the aramid nanofibers are characterized by a branching morphology. [0020] In another embodiment, a method comprises the steps of providing a suspension of aramid nanofibers in an aprotic solvent and then phase transforming the suspension by diffusing water into the suspension of to first step to make a hydrogel comprising water and aramid nanofibers. In non-limiting fashion, the aprotic solvent comprises dimethyl sulfoxide (DMSO). For example, the suspension comprises 0.1 - 5 weight % aramid nanofibers. [0021 ] Another method further involves the step of removing the water from the hydrogel to make an aerogel made of aramid nanofibers. One way to remove the water involves extracting with supercritical carbon dioxide.

[0022] The method can be carried continuously or in batch. Extrusion of the original dispersions of ANFs into water or other protic solvents results in the continuous production of hydrogel fibers. Extrusion from a coaxial needle results in a hollow fiber. In a continuous method, an ANF suspension is extruded through a syringe or extrusion die into a flowing water line so that the suspension is continuously contacted with the water. The suspension is phase transformed in real time as the suspension contacts the water. The extruded hydrogel can take the form of a fiber when extruded through a syringe, of a hollow tube when extruded through a coaxial die or needle, or can take the shape of an extrusion die, for example as a flat sheet of varying dimensions. In one embodiment, a syringe set up like one in FIG. 10H is used to extrude ANF suspension and join it with a water line to form the hydrogel at the same time the combined stream exits from the syringe. In an alternative, the ANF suspension is extruded from a die into a phase transformation bath (one containing enough protic solvent to accomplish phase transformation of the suspension) to form gel films. The extruded films are then dragged by a conveyor belt through the bath and into a subsequent drying stage, such as one that applies heated plates on top and bottom of the film. [0023] In another embodiment, a composite material comprises a polymer matrix and aramid nanofibers dispersed in the polymer matrix. For example, the composite material is made of an aerogel and a polymer in the pores of the aerogel, wherein the aerogel comprises aramid nanofibers. The polymer comprises a cured epoxy resin, in a non-limiting embodiment. [0024] Other nanomaterials can be added to the ANF dispersions. Examples include nanoparticles of metals to provide conductivity, ceramics to provide porosity and mechanical properties, magnetic to provide magnetic properties, carbon nanofibers to provide conductance, and cellulose nanofibers to provide mechanical properties and biocompatibility.

[0025] In a particular embodiment, the composite materials are made using a so-called gelation assisted layer-by-layer deposition process (gaLBL). When spin coating is used to deposit various materials in the process, it is also referred to as spin-coating assisted gelation layer-by-layer deposition, scgLBL, or similar nomenclature. In various embodiments, the method involves the steps of: a. providing a coating of a dispersion on a substrate, wherein the dispersion comprises aramid nanofibers and an aprotic solvent; b. phase transforming the coating of a. with a pro tic solvent to form a 3-D percolating network (3DPN); c. penetrating precursors of a polymer dissolved in solvent or a solution of a formed polymer into the 3DPN of step b. d. polymerizing the polymer precursors in situ in the case where precursors of the polymer are used in step c; and e. removing the solvent from the product of step d., for example by extraction with a supercritical fluid such as carbon dioxide, whereupon the 3DPN of ANF collapses and blends uniformly with the polymer. If desired, steps a. - e. can be repeated one or more times to build up a desired thickness of composite before delaminating the composite, with the understanding that the uniform blend of polymer and ANF becomes the substrate for the subsequent step a. of providing a coating of a dispersion on a substrate. The aprotic solvent comprises dimethyl sulfoxide (DMSO) in one embodiment. In this or other embodiments, the polymer precursors include an epoxy resin and an epoxy hardener.

[0026] In a particular embodiment, the method involves the steps of a. spin coating an ANF dispersion in DMSO on a substrate b exchanging DMSO in the dispersion with water to make a spun ANF thin film; c removing excess water by spinning; d spin coating an epoxy solution on top of the spun ANF thin film; and e. optionally repeating the above steps to build up a desired thickness

In an alternative process, two layers of epoxy are spin coated after each ANF dispersion coated.

[0027] In another embodiment, a spin coating gelation assisted layer by layer process involves one or more cycles of ANF 3DPN formation, wherein a cycle comprises the steps of phase transforming a spin coated ANF dispersion with water, and then infusing polymer by penetrating a solution comprising a solvent and either a polymer or polymer precursors into the 3DPN of the phase transformed ANF dispersion, then drying off the solvent to densify the ANF network. [0028] Various aspects of the limitations of these embodiments will now be more fully described, including a description of further options for each of the parameters of the invention. It is to be understood that the description of the various limitations can be mixed and matched, and recombined to enable and describe other embodiments not specifically exemplified. Substrate

[0029] The substrate used in the methods described herein provides a platform on which the composite materials can be built up. In some embodiments, the built-up composite is removed from the substrate before use. In other embodiments, the substrate becomes part of the device being made, such as is the case when ANF based insulation coatings are provided for implantable electronics. In various embodiments, the substrate can be selected from glass, metal, plastic, and so on. For electronics use, the substrate can be silica, optionally silica provided with a triple metal layer such as Cr/Au/Cr (20 nm/400 nm/20 nm).

Aramid Material

[0030] The aramid material described herein is made of aramid polymers, generally in the form of fibers, threads, or yarns. Aramid fibers of micro or macro dimensions are commercially available. Typically a commercially available aramid fiber is characterized by a diameter in the micro range, for instance by a diameter of one micron or greater. In one aspect, the current disclosure provides methods for transforming the commercially available microsized aramid fibers into ANFs.

[0031 ] Aramid polymers are defined as those that contain aromatic groups and amide linkages in the backbone. Normally, the amide groups provide linkages between adjacent aromatic groups. In one aspect, an aramid polymer is characterized as one in which at least 85% of the amide groups in the backbone are directly attached to two aromatic groups, especially where the aromatic groups are phenyl groups.

[0032] Two popular aramids are para-aramid fiber and meta-aramid fiber. The former is well known under the trademark of Kevlar® and others. Its backbone consists of phenyl groups separated by amide linkages, wherein the amides link the phenyl groups in a para configuration. A para-aramid represented by Kevlar is also called a poly (para-phenylene terephthalamide) (PPTA). Although the synthesis is not limited to reacting the particular monomers, in a simple form, a PPTA can be understood as the reaction product of para- phenylene diamine and terephthaloyl dichloride. In similar fashion, a meta-aramid such as illustrated by Nomex material can be understood as the product as meta-phenylene diamine and isophthaloyl dichloride.

[0033] Besides meta-aramids like Nomex and para-aramids like Kevlar, other aramid fibers are available of the copolyamide type. Copolyamides have structures that result from polymerizing other aromatic diamines with terephthaloyl or isophthaloyl chlorides, alternatively in the presence of /?ara-phenylene diamine. In whichever way the aramid fibers are produced, it is believed that the useful physical properties of the fibers derive from the stiff nature of the polymers and orientation along the fiber direction of the aromatic chains.

Nanofibers [0034] Aramid nanofibers (ANFs) are defined by their composition and by their dimensions. ANFs are made of aramid material such as discussed above. By nanofibers is meant that the diameter is in the nanometer range, and especially in the range of 3 to 100 nanometers, 3 to 50 nanometers, 4 to 40 nanometers, 3 to 30 nanometers, and 3 to 20 nanometers. In addition to diameters in the nanometer range, the ANFs are characterized by a high aspect ratio, meaning that the length of the fibers is at least 5 times, at least 10 times, or at least 20 times the diameter. In various embodiments, the length of the ANFs is greater than 0.1 microns or greater than 1 micron. Non-limiting examples of the length of the ANFs includes 0.1 to 10 microns and 1 micron to 10 microns.

[0035] In various embodiments, the ANFs have an empirically observed branched geometry. Branching in the nanofibers can be observed using electron microscopy or atomic force microscopy.

Making Suspensions Of Aramid Nanofiber(ANFs)

[0036] A process for making the aramid fibers is given in US publication number US

2013-0288050, the disclosure of which is incorporated by reference. The process involves preparing a suspension of ANFs in a solvent. An aprotic solvent such as DMSO has been found to be suitable for the synthetic method.

[0037] In various embodiments described herein, the process begins with commercially available aramid fibers having macro dimensions. Suitable aramid materials include those represented by the Kevlar® and Nomex® trade names and by aromatic co-polyamide materials. [0038] In an optional first step, an aramid fiber is treated with ultrasound in a solvent to swell the fiber as a preliminary step in the manufacture of a nanofiber. It has been found suitable to sonicate the starting material aramid fiber in a solvent such as NMP to swell the fiber before further reaction. If the ultrasound treatment is carried out, the swollen fiber can be removed from the ultrasound solvent after the treatment and combined in another solvent in the presence of a base. Alternatively, the optional ultrasound step is carried out in the same solvent used for making the nanofibers. Whether or not it was subjected to ultrasound, the fiber is then reacted with the base and the solvent - preferably an aprotic solvent or a mixture of solvents including an aprotic solvent - at low to ambient temperatures for a time until nanofibers are formed. Advantageously, the temperature of reaction is at 50° or less, at 40° or less, or at 30° or less. In various embodiments, the temperature of reaction is 20 to 50°C, 20 to 40°C, or 20 to 30°C. Reaction can be carried out at or about 20°C or at or about 25°C. Reaction at 25°C is considered to be at room temperature.

[0039] The aramid material is reacted in the base solution at a temperature below 50°C for times sufficient to produce ANFs having a diameter of 3 to 100 nanometers, 3 to 50 nanometers, or the like. Reaction times are generally on the order of days or weeks, and are illustrated further in working examples described herein, while the reaction is conveniently carried under ambient conditions such as at room temperature. Generally, the reaction temperature is above 0°C or above 10°C or above 15°C. In various embodiments the temperature is below 45°C, below 40°C, below 35°C, or below 30°C. In an illustrative embodiment, the reaction is carried out at about 20 - 30 °C. In this way the nanofibers are produced in situ and are suspended in the reaction solvent. It is found that under these conditions, the aramid material does not proceed to molecular dimensions where it could be considered as having been dissolved in the solvent. Rather, discrete nanofibers can be visualized in the suspension and in films produced from the suspension, which show the nanofibers having diameters less than 100 nm and high aspect ratios as described. The reaction in solution to prepare the suspension is carried out at low temperatures and can be carried out at room temperature or about 20 to 30°, for example, 25°C.

[0040] In a preferred embodiment, the solvent used to prepare the nanofibers is DMSO and the base is potassium hydroxide (KOH). In addition to the aprotic solvent, the reaction solvent can contain a minor amount, such as up to about 30 volume percent, of a protic solvent such as ethanol or isopropanol. Preferably, the base is provided in stoichiometric excess to the aramid material, and is conveniently applied in a one-to-one weight ratio. Relative to the amount of solvent, the combined weight of the base and the aramid material is from about 0.05 to 5%. In an exemplary embodiment, 500 mL of solvent is used, in which 1.5 grams of KOH are dissolved and in which 1.5 grams of aramid fiber material is suspended.

[0041 ] The reaction conditions are sufficient to transform the macro-dimensioned aramid starting material into ANFs having dimensions as discussed above and further exemplified below. The reaction conditions do not lead completely to dissolution of the aramid material. As a result, the reaction product is a suspension of nanosized ANFs in a solvent system that preferably contains DMSO.

[0042] In addition to DMSO, the solvent system can contain small amounts of water, for example wherein the volume ratio of water to the solvent (i.e., an aprotic solvent such as DMSO) is from 1/20 to about 1/1,000. In various embodiments, it has been observed that the presence of water leads to production of nanofibers having a smaller diameter. In this way, the diameter of the nanofibers produced by the method can be controlled to a desirable extent.

[0043] As a first step, a conventional aramid fiber (of micro or macro dimensions) is subjected to an optional first step involving exposing the aramid material to ultrasound energy while the aramid material of macro dimensions is in contact with a solvent. N- methylpyrrolidone (NMP) has been found to be a suitable solvent. The aramid material undergoing exposure to the ultrasound energy is made of fibers having a diameter of greater than 1 micrometers, in non-limiting example. A cartoon showing the relation of the microfiber to the nanofiber is available in FIG. 1.

[0044] After the optional pretreatment of the aramid material with ultrasound, the aramid material is combined with a solution containing a base and an aprotic solvent. The aprotic solvent can be the same as or different from the solvent in which the ultrasound treatment was carried out

[0045] In various embodiments, the KOH/DMSO solution can contain water in a volume ratio of water to DMSO of 1 to 20 to 1 to 1,000. In a non-limiting embodiment, the reacting solution contains equal parts by weight aramid material and KOH. Further, the aramid material and the base are present in relatively minor amounts in the solution, making up for example about 0.05 percent to about 5 percent by weight of the solution. In one embodiment, the weight of the KOH and the aramid material is about 7.6 percent by weight of the suspension.

[0046] The ANF suspensions optionally contain other nanomaterials. Non-limiting examples of other nanomaterials include nanoparticles of metals to provide conductivity, ceramics to provide porosity and improved mechanical properties, magnetic nanoparticles to provide magnetic properties, carbon nanofibers to provide conductance, and cellulose nanofibers (or cellulose nanowhiskers) to provide improved mechanical properties and biocompatibility. These can be added to the ANF dispersion, or they can be incorporated at other stages in the process. For example, a nanoparticle like cellulose nanofibers can be pre- dispersed into a solvent like DMSO, for example by sonication. The nanoparticle mixture is then mixed into the ANF dispersions described herein. In a non-limiting example, a cellulose nanofiber mixture (dispersion) in DMSO is mixed into a 2% ANF dispersion to create a 1.5% ANF solution. It is observed that mechanical properties like Young's modulus, tensile strength, and toughness increase with increasing amounts of incorporated cellulose nanofibers.

Phase transforming the suspension of nanofibers

[0047] The synthesis of the nanofibers results in a suspension of ANFs in an aprotic solvent, which is typically DMSO. To make a hydrogel from the suspension, the suspension is phase transformed by replacing the aprotic solvent with water or with water and another protic solvent. A visible material like a precipitate forms as the water replaces the aprotic solvent of the suspension. One way of phase transforming is to place the suspension in a cylindrical tube and to layer it with water by slowly dropping water on top of the suspension, while minimizing disturbance. The phase transformation begins at the water- solvent interface, and spreads as convection mixes the two solvents.

[0048] Another process called phase transforming involves adding water as a layer on top of a layer of nanofiber suspension that is coated on a substrate. The water in the added layer diffuses relatively quickly into the nanofiber suspension to cause what is being called a phase transformation.

[0049] The product of phase transforming is a hydrogel containing water and ANFs. As detailed below, it is characterized as a 3DPN. The designation of "percolating" means that there is a connected path through the network that extends in all three directions.

Removing water from the hydrogel to make an aerogel

[0050] Once the hydrogel is formed by any of the methods recited herein, an aerogel is prepared by removing the solvent in such a way as to maintain the 3DPN of the nanofibers. Examples include freeze drying and also extraction with a supercritical fluid such as supercritical carbon dioxide. Other methods to make aerogels include bubbling gas through the dispersion of ANFs, adding liquid that becomes gas at elevated temperature, and adding solid chemicals such as metal carbonates that produce gas upon chemical reactions.

[0051] The result is a porous light weight material that can be combined with polymer materials to make composites suitable for a number of applications. Alternatively, the aerogel can be ground or otherwise comminuted to provide filler material for other polymer composites.

Making composites ofANF's and polymers

[0052] Nanofibers described herein are compounded with polymeric materials to form various composites. In one aspect, the composites are describable as a porous aerogel, with a polymeric material and optionally other filler material disposed in the pores of the aerogel to form a solid composite. Alternatively, the composites are described as having a polymer matrix in which ANFs and optionally other materials are dispersed. They are dispersed in the polymer in the sense that the ANF and optional other materials are homogeneously blended in the polymeric matrix, which forms a continuous main phase. In various embodiments, enhanced physical properties result from the nature of the interaction of the nanofibers with the polymeric matrix. [0053] The polymers are choses from thermosets and thermoplastics. Thermosets include epoxies, polyurethanes, polyester resins, natural and synthetic rubbers, elastomers, fluorocarbon elastomers, phenol-formaldehyde resins, urea-formaldehyde resins, melamine- formaldehyde resins, diallylphthalate resins, polyimides, polycyanurates, and polyureas. Thermoplastics include acrylics such as poly (methyl methacrylate), acrylonitrile butadiene styrene (commonly known as ABS), polyamides such as Nylon 6 and Nylon 66, biodegradable polymers such as poly (lactic acid) (PLA), polybenzimidazole, polycarbonate, polyvinyl alcohol, polyvinyl acetate of various degrees of saponification, polyether sulfone, polyether ketone, polyetherimide, and aromatic polyamides such as the Kevlar® polymers from which the aromatic nanofibers are made.

[0054] In various embodiments, the composites are made by simply stirring, mixing, or otherwise agitating an combination of nanofiber and polymer until a dispersion is achieved. Since the method necessarily involves heating the mixture of nanofiber and polymer to the latter' s melting point, it is less preferred in some systems. Gelation assisted layer by layer deposition

[0055] Other methods of synthesis reported here for the first time involve gelation assisted layer by layer deposition. Instead of stirring or agitating to form a homogenous blend at or above the melting temperature of a polymer, a composite is built up layer by layer by alternately depositing a layer of ANF suspension and a layer of a solution of polymer precursors or a layer of a solution of polymer. Optionally, the layer of ANF suspension is subjected to a phase transformation before the polymer or polymer components are applied. After the first two layers of nanofiber and polymer are laid down, the bilayer forms a new substrate upon which subsequent bilayers of fiber and polymer can be laid, until a desired thickness of nanocomposite is achieved. [0056] In one embodiment, the aramid dispersion is phase transformed to a hydrogel before the polymeric components are added. After phase transformation with water or other protic solvent, the water-containing hydrogel is infiltrated with a solution of polymer components or a solution of polymer. The solvent containing the polymer or polymer components is any that readily dissolve the contents, and is preferably miscible with or readily soluble in the water or other solvent that is found in the hydrogel. Acetone is a non-limiting example. [0057] When polymer components are added by way of solution, the components can be subject to a subsequent polymerization step. In this way, composites of thermoset polymers can be prepared. Examples of such thermoset polymers include without limitation epoxies, polyurethanes, polyester resins, natural and synthetic rubbers, elastomers, fluorocarbon elastomers, phenol-formaldehyde resins, urea-formaldehyde resins, melamine-formaldehyde resins, diallylphthalate resins, polyimides, polycyanurates, polyisocyanurates, polyureas, and so on. The polymeric components include not only the monomers that react with one another to form the polymer, but also any necessary or desirable auxiliaries such as catalysts, curing accelerators, fillers, colorants, and so on. [0058] Composites with thermoplastic polymers are made by infiltrating a solution of the polymer into the aramid dispersion or into the phase transformed hydrogel made from the dispersion. Such can be accomplish by coating the solution onto the coating of aramid dispersion previously laid down, for example by spin coating. In some embodiments, polar polymers are preferred because they contain active hydrogens that can form hydrogen bonds to the aramid molecules, leading to better compatibility of the dispersed fibers and the polymer. Examples of polar polymers include acrylics such as poly (methyl methacrylate), acrylonitrile butadiene styrene (commonly known as ABS), polyamides such as Nylon 6 and Nylon 66, biodegradable polymers such as poly (lactic acid) (PLA), polybenzimidazole, polycarbonate, polyvinyl alcohol, polyvinyl acetate of various degrees of saponification, polyether sulfone, polyether ketone, polyetherimide, and aromatic polyamides such as the Kevlar® polymers from which the aromatic nanofibers are made. Non-polar polymers include polyethylenes such as polyethylene, polypropylene, polystyrene, polyvinyl chloride, and various fluoropolymers such as polytetrafluoroethylene (PTFE). Infiltration solutions of thermoplastic polymers can include various auxiliaries such as colorants, antioxidants, lubricants, and the like. [0059] As background, nanocomposites with the inclusion of high volume content well-dispersed nanomaterials have been successfully managed by the techniques of LBL, 17 ' 18 vacuum assisted flocculation (VAF) 19 ' 20 as well as the preformed nanoscale network infiltration (PNI). 1 1 1 11 21 22 LBL is a versatile bottom-up assembly approach, which makes use of the alternate nanoscale layering of macromolecular components based on the complementary interactions. 23 A major challenge of this technique in the classical non-exponential and non- spray version of LBL is the slow assembly process. A Ιμιη thick film can typically take 3-4 days to create. This issue is essentially related to the "two-dimensionality" of the absorption process. The strong interface obtained by equilibrated adsorption produces composites with excellent mechanical properties.17 Transition to the versions of LBL that take advantage of out-of-plane reactions or transport processes of LBL components have to sacrifice some energy of interactions of LBL components.

[0060] VAF is capable of somewhat faster assembly of nanocomposites than LBL. 19 However, this method brings about its own limitations exemplified by need for the two components to be dispersed in the same solvent, otherwise, bulk agglomeration and phase separation occur and preparation of VAF films become impossible. If the components are strongly attracted to each other, which is desirable for mechanical performance of the resulting

[0061 ] composite film until a complete stoppage occurs due to plugging of microporous filter membrane. PNI is based on infiltration of polymers into preformed 3DPN in the form of hydrogel or aerogel. 11 ' 21 ' 22 PNI method is capable of combining a variety of 3D nanoscale materials with infiltrating polymers to wide-range tunability of the composite functions. However, the long times associating with diffusion of the polymers, incomplete penetration of polymers into nanopores, and compression-induced stresses represent issues that one needs to consider for PNI. 21 ' 22 A gel-assisted layer-by-layer deposition (gaLBL) technique according to the current teachings combines the merits from both LBL and PNI. Moreover, the materials made by this method also display properties and especially their combinations that are unusual for composites made by these and other methods.

Epoxy nanocomposites [0062] For gaLBL a thin layer is deposited on a substrate by spin or spray coating

(FIG. 1) which is reminiscent of spin-assisted LBL. A polymer of choice is then allowed to diffuse into the 3D nanogel produced as the result of this step. This cycle can then be repeated as needed to get a desired thickness up to hundreds of nanometers of the composite deposited in each cycle. Since the layer is relatively thin, this process occurs quickly, which leads to materials accumulation faster than for LBL and materials integration better than that in PNI. Freedom to combine different components is potentially greater than for VAF.

[0063] Gel made from branched ANFs is used here for the gaLBL process due to (a) the simplicity and low cost of its preparation, (b) robustness of the network, and (c) high interfacial area with chemically active groups on ANF surface. In various embodiments, epoxy resin (EPX) is selected to infiltrate into the ANF gels for its generality of these thermosetting polymer materials displaying wide variety of mechanical properties and functionalities strength. EPXs are often used with microscale fibers, but their combination with nanoscale components 24- " 27 has until now been relatively difficult to make and relatively disappointing in performance, in part because of high viscosity of EPXs and difficulties with efficient dispersion of nanomaterials in them. The current teachings demonstrate that quasi-isotropic layered ANF/EPX composite can be made by gaLBL. Moreover, it shows excellent toughness exceeding 4 to 5 times that of conventional micro-composites. The ANF and EPX combination also displays close-to-zero thermal expansion and high transparency in the visible spectrum, which are the specific attributes imparted to the composite by ANFs.

EXAMPLES Example 1 - ANF epoxy composites - or ANF/EPX composites

[0064] ANF dispersion can be prepared by dispersing aramid microfiber (or commonly known as Kevlar fibers) in DMSO. A subsequent solvent exchange with water leads to the formation of ANF hydrogel in a process step called phase transforming. In order to make a thin ANF hydrogel three-dimensional percolated network, a thin liquid layer of ANF in DMSO is spin-coated on a glass substrate and subjected to solvent exchange. The resulting thin ANF hydrogel can be carefully peeled off from the substrate (FIG. 2A) and dried by supercritical C0 2 to examine the structure (FIG. 2B). In a non-limiting example, the obtained ANF aerogel is 57μιη thick with a wide range of pores defined by interconnected ANFs (FIG. 2C).

[0065] To make the ANF/EPX composites, the ANF hydrogel layer is left on the substrate to allow the diffusion of EPX. The hydrogel is adhesive enough to the substrate to allow EPX to diffuse into the three-dimensional percolated networks without causing wrinkles. 0.1— 2% EPX in a water miscible solvent such as acetone is used. Extra EPX solution can be removed by spinning off the substrate. After coating, the substrate is put in the 100 °C oven for 2 min to pre-anneal the film by removing the solvent. Then, another cycle can take place with ANF hydrogel deposited on the solidified coating. The cycle can be repeated continuously to obtain the needed thickness, which is similar to the conventional LBL assembly. 23 The film made after nth cycle is denoted as [ANF/EPX] n .

[0066] The linear growth of the film is confirmed by absorbance and thickness change

(FIGS. 2D, 2E, and 2F). An absorbance band centered at 330nm is shown for ANF/EPX film, and its intensity increases linearly with cycle number. A similar linear trend is observed for thickness, which can be finely tuned by the concentration of ANF and EPX. When concentrations are 0.1%, 0.2% or 1% for each component at the same spin rate of 1000 rpm, the average thickness per cycle is 7.5, 18, and 342 nm, respectively. This result demonstrates the GLBL technique has remarkable thickness control from several to hundreds of nanometers per layer.

[0067] The 1% ANF solution is then focused on for the following investigations for the thick layer formed each cycle. EPX concentration is changed from 0.1 to 2% to control the ANF volume fraction in the composites. As is determined by thermal gravimetrical analysis, ANF weight fraction is 90%, 87%, 64% and 38% when EPX varies at 0.1%, 0.5%, 1% and 2%. TGA also demonstrates a high decomposition temperature of 250°C.

[0068] The ANF/EPX composite is typically transparent (FIG. 3A). The transparency of [1%ANF/1%EPX] 6 is 88% at 700nm (FIG. 3C). High uniformity in the film is indicated by

Fabry-Perot patterns 17 ' 28 displayed in the absorbance spectrum. The ANF/EPX composite films can also be easily delaminated from glass substrate using dilute HF. 18 ' 29 The obtained transparent freestanding film is flexible enough to be wrapped around a pen (FIG. 3B). Fourier transform infrared spectroscopy (FTIR) confirms the chemical features of both ANF and EPX in the composite.

[0069] Under certain conditions, a stratification of ANFs (FIGS. 3D, 3E, and 3F) is observed in the composite, which is likely owing to the compression induced alignment during the collapse of the three-dimensional percolated network. With the increase of EPX in the composite, the layered structure becomes less distinct (FIG. 3G). This configuration of nanofibers is similar to nanosheets reinforced composites, such as clay 17 or graphene, 19 and quite different from other ID nanoreinforcement, such as carbon nanotube, 18 ' 29 or cellulose nanofibers. 30 This observation indicates strong ANF-ANF interactions in the composites as a result of abundant hydrogen bonds with amide functional groups. With such strong interactions, the strand of ANF under stress can easily transfer to other strands so that the whole neighboring network can be pulled out under load and then fractured. EPX here serves as crosslinks to ANF contacts when filling into the nanopores formed during the drying (FIG. 31). However, when EPX content is above a threshold, individual ANF can be completely surrounded by EPX. The strong ANF-ANF interaction is replaced by ANF-EPX-ANF interaction. The fracture then occurs at each individual ANF rather than the layered collective mode (FIG. 3G). The overfilling of EPX is also evident in the abrupt jump of the film thickness from approximately 2 μιη to 2.7 μιη when 2% EPX is used for infiltration. The surface of overfilled EPX composites is also less porous by appearance (FIG. 3J vs. FIG. 3K). Those structural differences can influence the mechanical and thermal expansion properties of the composites.

[0070] Rather than the brittle behavior of micro-composites, the ANF/EPX composites are rather ductile, demonstrating a plastic deformation after the initial elastic region (FIG. 4A). This characteristic is similar to those high performance aerospace alloys, such as titanium, steel

31

or aluminum. In particular, [1%ANF/1%EPX]6 shows an ultimate strength (o u ) of 505+47 MPa, an ultimate strain (e u ) of 0.16+0.03 with only a density (p) of 1.5+0.1 g/cm . The calculated toughness (K) by integrating the area under the stress/strain curve is 50.1+ 9.8 MJ/m . Similar to the micro-composites, the specific strength (σ„ / p) of this ANF/EPX composite is significantly larger than that of titanium, steel or aluminum alloys 14 (FIG. 4B). The absolute strength o u is much higher than those of SAE 1010 steel (365 MPa) and 6061-T6 aluminum alloy (310 MPa). 14 Although σ Μ οί [l%ANF/l%EPX] 6 is not comparable to the o u of unidirectional micro-composites in the alignment (0°) direction, it is 10 times and 16 times higher than the o u in the 90°direction in those micro-composites respectively 16 (FIG. 4C). In addition, both the quasi-isotropic laminas of carbon and aramid fiber micro-composites are inferior in o u (303 MPa and 141 MPa, respectively) 16 to the ANF/EPX examined in this work (FIG. 4C), which is more intrinsically isotropic with the nanofiber reinforcement.

[0071 ] Such a high o u for ANF/EPX composite, however, is not connected with the brittleness or low toughness that many composite materials have encountered. 13 ' 32 The toughness (K) of the ANF/EPX is 4 to 5 times higher than that of the unidirectional microcomposites measured in the 0° direction. The K is also much higher than the layered composites made by alumina nanoplatelets 10 or carbon nanotube 29 with greater or comparable o u . It is believed that o u and K are mutually exclusive properties in many cases. 32 ' 33 An optimum combination of both properties, however, is important for structural materials to avoid catastrophic failure under load. The solution to this dilemma relies on the design of hierarchical composite architecture similar to that in many natural materials. 32 ' 34 Here, the layered configuration of ANF, their strong interactions, and crosslinks formed by EPX lead to strong load-bearing ability. The collective layered failure mode plus the "stick-slip" interactions afforded by the hydrogen bonds 35 also facilitate energy dissipation during the stretching and thus high toughness.

[0072] More EPX or less in the composite leads to poorer mechanical performance than the [1%ANF/1%EPX] 6 (FIGS. 5A, 5B, 5C, and 5D). As is discussed above, this condition is a transition point at which nanopores start to get overfilled. This structural evolvement directly affects the o u and K, plus the storage modulus E', which is an indication of elastic stiffness of the material. All those properties show maxima for the film of [1%ANF/1%EPX] 6 (FIGS. 5 A, 5B, 5C, and 5D).

[0073] To be noted here, pure ANF film without EPX shows a u of 387+25 MPa, e u of 0.16+0.03, E' of 11.5+0.5 GPa, which are much higher than those from VAF made ANF film (ff M ~160MPa, e u ~0.1, E-7.1). 36 The mechanical performance enhancement can result from this gel-assisted LBL processing technique. In VAF, the long filtration step can lead to the deterioration of solution qualities, and thus the generation of defects in the final film. Additionally, hand-peeling the ANF film from the porous membrane can also break some film microstructures. In gel-assisted LBL, all those defects-introduction steps are avoided. Besides, some mechanical factors in the spin coating process, such as the centrifugal and air shear force can usually lead to some degree of lateral chain orientation and stratification. 37 ' 38 These factors can lead to an improved mechanical performance for the gel-assisted LBL made film with the same chemical composition, but with the different microarchitecture. [0074] The damping ratio, or tan δ, which is the ratio between the loss modulus and storage modulus. Tan δ measures the degree to which a material dissipates the vibration energy into heat. High damping capacity is useful in many automotive and sporting goods applications. 14 The carbon and aramid micro-composites typically have very low tan δ of 0.0024 and 0.018 respectively. 16 The ANF/EPX composites show higher tan δ (FIG. 5D). Interestingly, pure ANF film displays the largest tand over the range of 0.1- lHz. At 0.1 Hz, the tan δ of ANF film can be as high as 0.14, and is decreased to 0.06 at 1 Hz. With the addition of EPX, the tan δ gradually declines. The highest tan δ of pure ANF film can be related with the more freedom of ANFs in the periphery of abundant unfilled nanopores. The strong but unlocked ANF-ANF interfaces cause the high mechanical damping. 25 With the introduction of EPX into the PNN, the touching interface is gradually locked by the poor- damping EPX to result in a lower tan δ. [1%ANF/1%EPX] 6 has slightly lower tan δ of 0.11 at 0.1 Hz and 0.5 at 1 Hz. An significant drop of tan δ occurs for [1%ANF/2%EPX] 6 , which has even smaller tan δ than EPX . This finding actually agrees with the previously discussed structure transition. In this film, ANF is surrounded by EPX with the interface of ANF-EPX- ANF. The dissipating mechanism relying on the ANF-ANF friction disappears here, while the ANF serves as reinforcing agent for EPX to make its chain even less mobile.

[0075] Low coefficient of thermal expansion (CTE) is another key feature for conventional carbon or aramid micro-composites. 14 The low CTE can exhibit a better dimensional stability over a wide temperature range. Unidirectional carbon micro-composites have CTE of -0.44-0.16 ppm K _1 in 0° direction, and 0.36 to 4.02 ppm K "1 in 90° direction, while the quasi-isotropic ones have CTE of 0.36 to 4.02 ppm K "1 . 16 On the other hand, unidirectional aramid micro-composites have CTE of -2.57 to -1.74 ppm K "1 in 0° direction, and 21.4 to 27.5 ppm K "1 in 90° direction, while the quasi-isotropic ones have CTE of 9.5 to 12.9 ppm K "1 . 16 Here, [1%ANF/1%EPX] 6 can have the quasi-isotropic close-to-zero CTE of - 0.9 ppm K "1 until 220 °C (FIG. 6). With finer tuning, a real zero expansion can be achieved. It is interesting to notice that the aramid micro-composites have high positive CTE at 90° to the fiber direction, and this property contributes to the slightly lower but still positive CTE for quasi-isotropic composites. However, material made with ANF with the same chemical composition but much smaller diameter gives overall negative and even zero CTE.

[0076] This phenomenon is related with the existence of nanopores. Aramid microfiber is proved to have CTE of -4.9 ppm, agreeing with the previous studies. Pure ANF film shows two regimes of thermal expansion: one with CTE of -6 ppm K "1 up to 75 °C, and the other one with CTE of -0.5 ppm K "1 . Similar to graphene, 40 ' 41 the negative CTE in ANF is contributed by the transversal acoustic bending modes along the axis, 42 ' 43 or commonly known as the "membrane effect". 44 The nanopores formed by the overlapping of ANFs give rise to more free space to enhance this bending effect, thus more negative CTE. 43 With the increase of temperature, however, other phonon modes contributing to the positive CTE might take effect. 42 " The nanopores can also accommodate the positive radial expansion of ANF. In addition, since the mechanical load is carried mainly by the axial direction of the ANF, the overall CTE shows more axial behavior of ANF. With EPX in the composite, the thermal behavior is a combined effect of both components. When the nanopores are underfilled, the CTE increases a bit even with 36% EPX in the composites. When the nanopores are overfilled, such as in the film of [1%ANF/2%EPX] 6 , the CTE has increased to 11 ppm. In this film, not enough space exists to accommodate the positive radial expansion, and EPX as a matrix can uniformly distribute the load to various directions in ANF. Accordingly, the radial expansion can contribute more to the overall CTE.

[0077] To conclude, a gel-assisted LBL method makes transparent, strong and tough ANF/EPX composites with high damping and zero-expansion. The ultimate fracture strength of the composites is higher than that of quasi-intrinsic carbon or aramid microfiber reinforced composites (micro-composites). The toughness is even better than that of the unidirectional micro-composites. ANF/EPX composites with such combined functionalities can be used for bio-implants, packaging materials, electronic boards, bullet-proof windows and many more. This gel-assisted LBL approach can also be adopted for other gel-forming networks, such as cellulose nanofibers, carbon nanotube, and graphene to produce other functional materials for fields in need.

References for Example 1 :

1. Kojima, Y.; Usuki, A.; Kawasumi, M.; Okada, A.; Fukushima, Y.; Kurauchi, T.;

Kamigaito, O., Mechanical-Properties of Nylon 6-Clay Hybrid. J. Mater. Res. 1993, 8, 1185-1189.

2. Glotzer, S. C; Solomon, M. J., Anisotropy of Building Blocks and Their Assembly into Complex Structures. Nature Materials 2007, 6, 557-562.

3. Rogers, J. A.; Someya, T.; Huang, Y. G., Materials and Mechanics for Stretchable Electronics. Science 2010, 327, 1603-1607.

4. Bockstaller, M. R.; Thomas, E. L., Optical Properties of Polymer-Based Photonic Nanocomposite Materials. J Phys Chem B 2003, 107, 10017-10024.

5. Murugan, R.; Ramakrishna, S., Development of Nanocomposites for Bone Grafting.

Compos Sci Technol 2005, 65, 2385-2406.

6. Baughman, R. H.; Zakhidov, A. A.; de Heer, W. A., Carbon Nanotubes - the Route toward Applications. Science 2002, 297, 787-792.

7. De Voider, M. F.; Tawfick, S. H.; Baughman, R. H.; Hart, A. J., Carbon Nanotubes:

Present and Future Commercial Applications. Science 2013, 339, 535-9.

8. Lundstrom, M. S., Graphene the Long and Winding Road. Nature Mater. 2011, 10, 566-567.

9. Rao, C; Sood, A.; Subrahmanyam, K.; Govindaraj, A., Graphene: The New Two- Dimensional Nanomaterial. Angew. Chem. 2009, 48, 7752— 7777.

10. Bonderer, L. J.; Studart, A. R.; Gauckler, L. J., Bioinspired Design and Assembly of Platelet Reinforced Polymer Films. Science 2008, 319, 1069—1073.

11. Deville, S.; Saiz, E.; Nalla, R. K.; Tomsia, A. P., Freezing as a Path to Build Complex Composites. Science 2006, 311, 515— 518.

12. Erb, R. M.; Libanori, R.; Rothfuchs, N.; Studart, A. R., Composites Reinforced in Three Dimensions by Using Low Magnetic Fields. Science 2012, 335, 199— 204. 13. Podsiadlo, P.; Kaushik, A. K.; Arruda, E. M.; Waas, A. M.; Shim, B. S.; Xu, J.; Nandivada, H.; Pumplin, B. G.; Lahann, J.; Ramamoorthy, A.; Kotov, N. A., Ultrastrong and Stiff Layered Polymer Nanocomposites. Science 2007, 318, 80— 83.

14. Mallick, P. K., Fiber-Reinforced Composites: Materials, Manufacturing, and Design.

CRC Press: Boca Raton, FL, 2008; pp 619

15. Dzenis, Y., Materials Science - Structural Nanocomposites. Science 2008, 319, 419— 420.

16. Cambridge Materials Selector Database (Grantadesign.Com).

17. Podsiadlo, P.; Kaushik, A. K.; Arruda, E. M.; Waas, A. M.; Shim, B. S.; Xu, J. D.;

Nandivada, H.; Pumplin, B. G.; Lahann, J.; Ramamoorthy, A.; Kotov, N. A.,

Ultrastrong and Stiff Layered Polymer Nanocomposites. Science 2007, 318, 80— 83.

18. Zhu, J.; Shim, B. S.; Di Prima, M.; Kotov, N. A., Transparent Conductors from Carbon Nanotubes Lbl- Assembled with Polymer Dopant with Π-Π Electron Transfer. . Am. Chem. Soc. 2011, 133, 7450—7460. 19. Putz, K. W.; Compton, O. C; Palmeri, M. J.; Nguyen, S. T.; Brinson, L. C, High- Nanofiller-Content Graphene Oxide-Polymer Nanocomposites Via Vacuum- Assisted Self-Assembly. Adv. Fund. Mater. 2010, 20, 3322—3329.

20. Chen, H.; Muller, M. B.; Gilmore, K. J.; Wallace, G. G.; Li, D., Mechanically Strong, Electrically Conductive, and Biocompatible Graphene Paper. Adv. Mater. 2008, 20, 3557—3561.

21. Capadona, J. R.; Van Den Berg, O.; Capadona, L. A.; Schroeter, M.; Rowan, S. J.;

Tyler, D. J.; Weder, C, A Versatile Approach for the Processing of Polymer Nanocomposites with Self-Assembled Nanofibre Templates. Nat. Nanotechnol. 2007,

2, 765—769. 22. Kim, K. H.; Vural, M.; Islam, M. F., Single -Walled Carbon Nanotube Aerogel-Based Elastic Conductors. Adv. Mater. 2011, 23, 2865—2869.

23. Decher, G., Fuzzy Nanoassemblies: Toward Layered Polymeric Multicomposites.

Science 1997, 277, 1232—1237.

24. Rafiee, M. A.; Rafiee, J.; Wang, Z.; Song, H. H.; Yu, Z. Z.; Koratkar, N., Enhanced Mechanical Properties of Nanocomposites at Low Graphene Content. ACS Nano 2009,

3, 3884-3890. Suhr, J.; Koratkar, N.; Keblinski, P.; Ajayan, P., Viscoelasticity in Carbon Nanotube Composites. Nat. Mater. 2005, 4, 134-137. Zandiatashbar, A.; Picu, C. R.; Koratkar, N., Control of Epoxy Creep Using Graphene. Small 2012, 8, 1676—1682. Spitalsky, Z.; Tsoukleri, G.; Tasis, D.; Krontiras, C; Georga, S. N.; Galiotis, C, High Volume Fraction Carbon Nanotube-Epoxy Composites. Nanotechnology 2009, 20, 405702. Guan, Y.; Yang, S. G.; Zhang, Y. J.; Xu, J.; Han, C. C; Kotov, N. A., Fabry-Perot Fringes and Their Application to Study the Film Growth, Chain Rearrangement, and Erosion of Hydrogen-Bonded Pvpon/Paa Films. . Phys. Chem. B 2006, 110, 13484- 13490. Shim, B. S.; Zhu, J.; Jan, E.; Critchley, K.; Ho, S. S.; Podsiadlo, P.; Sun, K.; Kotov, N. A., Multiparameter Structural Optimization of Single-Walled Carbon Nanotube Composites: Toward Record Strength, Stiffness, and Toughness. ACS Nano 2009, 3, 1711—1722. Eichhorn, S. J.; Dufresne, A.; Aranguren, M.; Marcovich, N. E.; Capadona, J. R.; Rowan, S. J.; Weder, C; Thielemans, W.; Roman, M.; Renneckar, S.; Gindl, W.; Veigel, S.; Keckes, J.; Yano, H.; Abe, K.; Nogi, M.; Nakagaito, A. N.; Mangalam, A.; Simonsen, J.; Benight, A. S.; Bismarck, A.; Berglund, L. A.; Peijs, T., Review: Current International Research into Cellulose Nanofibres and Nanocomposites. . Mater. Sci. 2010, 45, 1—33. Meyers, M. A.; Chawla, K. K.; Hosford, W. F.; MyiLibrary, Mechanical Behavior of Materials. Cambridge University Press: Cambridge ; New York, 2009; pp 856. Ritchie, R. O., The Conflicts between Strength and Toughness. Nat. Mater. 2011, 10, 817—822. Launey, M. E.; Ritchie, R. O., On the Fracture Toughness of Advanced Materials. Adv. Mater. 2009, 21, 2103—2110. Studart, A. R., Towards High-Performance Bioinspired Composites. Adv. Mater. 2012, 24, 5024-5044. 35. Dikin, D. A.; Stankovich, S.; Zimney, E. J.; Piner, R. D.; Dommett, G. H. B.; Evmenenko, G.; Nguyen, S. T.; Ruoff, R. S., Preparation and Characterization of Graphene Oxide Paper. Nature 2007, 448, 457— 460.

36. Cao, K.; Siepermann, C. P.; Yang, M.; Waas, A. M.; Kotov, N. A.; Thouless, M. D.;

Arruda, E. M., Reactive Aramid Nanostructures as High-Performance Polymeric

Building Blocks for Advanced Composites. Adv. Fund. Mater. 2012, DOI: 10.1002/adfm.201202466.

37. Ariga, K.; Hill, J. P.; Ji, Q., Layer-by-Layer Assembly as a Versatile Bottom-up Nanofabrication Technique for Exploratory Research and Realistic Application. Phys. Chem. Chem. Phys. 2007, 9, 2319-2340.

38. Cho, J.; Char, K.; Hong, J. D.; Lee, K. B., Fabrication of Highly Ordered Multilayer Films Using a Spin Self-Assembly Method. Adv. Mater. 2001, 13, 1076—1078.

39. Jain, A.; Vijayan, K., Kevlar 49 Fibres: Thermal Expansion Coefficients from High Temperature X-Ray Data. Curr. Sci. 2000, 78, 331—335. 40. Bao, W. Z.; Miao, F.; Chen, Z.; Zhang, H.; Jang, W. Y.; Dames, C; Lau, C. N., Controlled Ripple Texturing of Suspended Graphene and Ultrathin Graphite Membranes. Nature Nanotech. 2009, 4, 562— 566.

41. Nelson, J. B.; Riley, D. P., The Thermal Expansion of Graphite from 15°C to 800°C.

Proc. Phys. Soc. 1945, 57, 477—486. 42. Mounet, N.; Marzari, N., First-Principles Determination of the Structural, Vibrational and Thermodynamic Properties of Diamond, Graphite, and Derivatives. Phys. Rev. B 2005, 71, 205214.

43. Grigoriadis, C; Haase, N.; Butt, H. J.; Mullen, K.; Floudas, G., Negative Thermal Expansion in Discotic Liquid Crystals of Nanographenes. Adv. Mater. 2010, 22, 1403—1406.

44. Lifshitz, I., Thermal Properties of Chain and Layered Structures at Low Temperatures.

Zh. Eksp. Teor. Fiz. 1952, 22, 475—486.

Example 2 - 3DPN's from aramid hydrogels

[0078] One-dimensional (ID) nanoscale components can form light-weight highly interconnected three-dimensional percolated networks (3DPNs) with considerably improved physical properties, such as high mechanical and electrical performance in comparison to their non-percolating analogs 1"9 , and are therefore essential for both load bearing structures and portable electronic devices 10 ' 11 . Typical ID nanomaterials, such as carbon nanotube, metal nanowires, or organic nanofibers, have a rodlike morphology. Their percolations are generally achieved through the direct contact with each other on the confined cross-area by limited molecule forces, which could be easily disrupted by mechanical disturbance, such as shearing or stretching. Nature, however, adopts a more effective strategy by using branched ID nanomaterials as network building blocks. For example, actin filament branched network ubiquitous in the cellular cytoskeleton shows exceptional elastic behavior thanks to the effective dissipation of the stress through the stretching and bending of the individual filament. 12"14 Here we follow this design principle by synthesizing the branched ultrastrong ANFs from Kevlar® microfibers and serve as components for 3DPNs. Branched ANFs readily form hydrogel or aerogels with highly porous frameworks with mechanical properties exceeding those from non-branched nanomaterials with higher elastic crystalline moduli. ANF 3DPNs can also be made into different shapes through controlling the assembly conditions. In an embodiment, polymers such as epoxy are impregnated into the 3DPNs through a spin- coating assisted gelation layer-by-layer deposition (scg LBL) process. The nanocomposite fabrication process enables control of the film structure and thickness with nanometer resolution. The prepared ANF/epoxy composites are transparent, and have an optimum ultimate strength of 505+47 MPa, toughness of 50.1+ 9.8 MJ/m , and close-to-zero thermal expansion, beating the properties of quasi-isotropic epoxy composites reinforced by Kevlar or carbon microfibers.

[0079] Kevlar® microfibers (KMF) consist of aligned poly(p-phenylene terephthalamide) (PPTA) chains connected by strong hydrogen bonds (FIG. 7A). The structural hierarchy of KMFs can be visualized in scanning electron microscopy (SEM) after tensile failure (FIG. 7B and FIG. 7C) and indicates the presence of constitutive nanofibers. KMF can be split chemically into nanofibers by deprotonation with saturated KOH in dimethyl sulfoxide (DMSO) 15 . The abstraction of protons from PPTA leads to the dissociation of weaker intermolecular bonds while the constitutive nanofibers remain intact. Atomic force microscopy (AFM) images of ANFs indicate that they have an average diameter of 4.5 nm (FIG. 7D and FIG. 7E), and a total length of several micrometers (FIG. 7D and FIG. 7E) with branching morphology, which were further corroborated by Transmission Electron microscopy (TEM) and SEM images of exfoliated ANFs (FIG. 7F and FIG. 7G). The ANFs's ability to bifurcate, or branch, several times along its length is a reflection of their structural organization in KMFs. The chemical signature that differentiates the ANF from the KMF is an upshift of N-H and C=0 stretching vibrations (FIG. 7H, FIG. 71, and FIG. 7J), 15 which are related to the hydrogen bonding environment. It is especially interesting to note that the C=0 vibration modes in the nanofiber is split into two peaks, which might be induced by the intra- and inter- nanofiber hydrogen bonds in the ANF. [0080] High aspect ratio plus the branched morphology makes ANF a promising nanoscale material for forming the 3DPNs. Indeed, the ANF three-dimensional percolated networks in the form of hydrogels can be formed using a solvent exchange process (FIG. 8A, FIG. 8B, and FIG. 8C). In a typical procedure, a layer of deionized water is gently sited on top of the DMSO phase containing 1 wt % ANF. With the slow diffusion of water, the deprotonated PPTA chains are gradually reverted to their initial chemical structures by abstracting protons from water, accompanied by a color change from dark red to light yellow (FIG. 8A vs. FIG. 8C). The attained hydrogel is highly robust, and can sustain structural integrity even after cutting and shearing with a razor blade (FIG. 8D). To further reveal the internal structure of ANF in the hydrogel, it is converted into an aerogel through the supercritical C0 2 extraction (FIG. 8E, FIG. 8F, and FIG. 8G). As expected, the aerogel contains a network of highly branched ANF nanofibers entangled with each other. The ANF aerogel is ultralight with a measured density of 11 mg/cm and adheres to electrified glass (FIG. 8E). Additionally, it has a Brunauer-Emmet-Teller (BET) surface area as large as 275 m /g, which could be compared to CellNF aerogel with a BET area ranging from 20 to 66 m 2 /g 16 .

[0081 ] Another important feature of ANF percolated nanofiber network is its ability to be spun continuously through a simple setup (FIG. 8H). With the constant flow of DI water, the continuously extruded 0.1 wt % ANF solution through an ultrafine needle can be rapidly converted to a fiber made of ANF 3DPNs (FIG. 8J, FIG. 8K, and FIG. 8L). Inside the fiber, ANF morphology is similar to those obtained by the static diffusion process. In addition, the dimension of the fiber approximates the size of hole in the needle, which indicates the minimal shrinkage during the phase transformation and supercritical drying process. This gel has an estimated density of 1.1 mg/cm , which is among the few materials existing in the ultralight regime. 17 ' 18 Considering no other ID nanomaterials could form robust networks at such a low concentration, we conclude that the interconnection of ANF through the branched structure is the key to the formation of 3DPN.

[0082] Further in-depth, the robustness of the ANF 3DPNs is substantiated by quantitatively evaluating their mechanical properties in respect to shear, compression and tension, which may further shed light on the role of the branched structure by comparing with other gels/3DPNs from non-branching nanomaterials.

[0083] The shear stress-strain curve of a 1% ANF hydrogel shows a linear viscoelastic region ending at strain amplitude of 10%, followed by a softening region where the hydrogel starts to break and flow (FIG. 9A). The maximum stress at the turning point - often known as the critical shear strength (x c ) - indicates the mechanical robustness of the gel. ANF hydrogel shows T c of 2.95+0.05 kPa, which is much larger than that of graphene hydrogel (x c =0.4kPa) at a similar solid content. 19 The dynamic shear test also separates the elastic and viscous contributions in terms of storage modulus (G') and loss modulus (G"). Agreeing with the linear region in FIG. 9A, the G' is initially independent on the strain, and then decreases strains above 10% (FIG. 9B). G" initially increases and then decreases, which reflects structural rearrangement in the materials for strains above 10%. G' varies little with the angular frequency from 0.06 to 60 rad/s remaining around 29 kPa at a fixed oscillatory strain of 1% (FIG. 9C). By contrast, G" exhibits a shallow region with larger value at higher and lower frequency regions. The low frequency rise implies the existence of slow structural rearrangement process, while the high frequency rise is attributed to the viscous relaxation of water in the gel.

[0084] The rheological properties of hydrogel are mainly dependent on the intrinsic mechanical behavior of the constituents and their short or long range interactions. FIG. 9D shows a comparison between the ANF hydrogel and other hydrogels made of typical reinforcing nanomaterials with high crystalline elastic moduli . 20- " 24 The hydrogels in comparison have a solid content around 1 wt%. Clearly, ANF and CellNF hydrogels are more rigid than carbon nanotube or graphene hydrogels despite the latter has stiffer nanocomponents, which indicates stronger molecular affinity among those polymer nanofibers in the hydrogel. Moreover, ANF hydrogel has 3 times as high shear G' as the CellNF hydrogel despite their similar crystalline moduli, or hydrogen bonding interactions at the molecule level. Considering their morphological resemblance including fiber diameter or length, we can safely attribute the higher G' of ANF hydrogels to the interconnecting branched nanofiber percolating structure. Additional comparison of ANF hydrogel with other chemically crosslinked polymeric hydrogels, such as those made of polyacrylamide (PAM) or polyethylene glycol (PEG) with larger solid content also indicates its higher rigidity. This observation indicates the moduli of the ingredients in the network also contribute to the shear properties of the hydrogel in addition to the connectivity. ANF hydrogel has both of the qualities. [0085] Uniaxial compression and tension tests of the ANF 3DPN also indicate its high mechanical strength (FIG. 9E, FIG. 9L, and FIG. 9M). 18 ' 25 ' 26 An interesting property of ANF 3DPN during compression is that both hydro- and aerogels can be compressed to strains over 90% without any cracks at the macro-, micro-, or nanoscale (FIGS. 9F-9K) while typical graphene or cellulose networks develop macroscopic cracks or even fracture at a much smaller compressive strain 16 ' 19 . More importantly, the ANF tends to align in perpendicular to the direction of compressing stress, while the large void space in the network can accommodate those deformations and ensure a zero Poisson's ratio (FIG. 9F, FIG. 9G, and FIG. 9K). The compressed aerogel also shows improved tensile properties due to the reduction of porosity (FIG. 9M). All those properties would be advantageous for infiltration of matrix materials into the 3DPNs for fabricating high performance materials.

[0086] 3DPNs can be served as the nano-skeletons for improving the mechanical properties, and the integration of 3DPNs and matrix materials are typically achieved through immersing the bulk 3DPNs into the polymer solution. 27 However, we consider this method not scalable as the diffusion of the polymer usually takes substantial time to reach deep inside especially when the gel dimension is millimeter or larger. This challenge can be circumvented by making thinner networks. Our first attempt is to develop a very thin hydrogel mold by conducting the ANF phase change between two glass-slides, and we achieved decent success in preparing polyvinyl alcohol/ANF composites with high strength and toughness. The scalability and uniformity of this fabrication method, however, is still not completely satisfying. We further made use of the adaptability of ANF gel formation, and developed a spin-coating assisted gelation and layer-by-layer deposition process (scgLBL). We also extended the matrix material to a more industrially popular matrix material, epoxy (EPX), and made higher performance ANF/EPX composites in comparison to other techniques.

[0087] A typical scgLBL process involves several cycles of thin ANF 3DPN formation by phase transforming a spin-coated ANF dispersion with DI water, and subsequent penetration of precursors dissolved in acetone (or other water soluble solvent) into the 3DPN (FIG. 10A). After drying off the solvent, the 3DPN of ANF collapsed and blended uniformly with the EPX. The repeated cycles of transformation-diffusion-densification of film making processes reminds of the conventional layer-by-layer (LBL) process based on electrostatic interaction. The film made after nth cycle is denoted as [ANF/EPX] n

[0088] The thin ANF 3DPN can be easily transferred from the substrate to DI water, and allow the examination of similarly porous internal structure defined by interconnected and branched ANFs (FIG. 10B, FIG. IOC, FIG. 10D). The linear growth of the film is confirmed by absorbance and thickness change (FIG. 10G, FIG. 10H, and FIG. 101). An absorbance band centered at 330nm is shown for ANF/EPX film, and its intensity increases linearly with cycle number. Similar linear trend can be observed for thickness, which can be finely tuned by the concentration of ANF and EPX. When concentrations are 0.1%, 0.2% or 1% for each component at the same spin rate of 1000 rpm, the average thickness per cycle is 7.5, 18, 342 nm (FIGS. 9G-9I). This result demonstrates scgLBL technique has remarkable thickness control from several to hundreds of nanometers per layer. The 1% ANF solution is then focused on for the following investigations. The ANF weight fraction in the composites can thus be controlled at 90%, 87%, 64% and 38%, which are determined by thermal gravimetrical analysis (TGA) when EPX concentration varies at 0.1%, 0.5%, 1% and 2%. The resultant [1%ANF/1%EPX] 6 composite has a transparent appearance with transmittance of 88% at 700 nm and high uniformity indicated by Fabry-Perot patterns 3 ' 28 After being delaminated by HF, the freestanding ANF composite is robust and quite flexible. FTIR confirms the chemical features of both ANF and EPX in the composite.

[0089] The uniform cross-section of all the fractured ANF composites over the thickness indicates the even distribution of ANF and EPX (FIGS. 10I-10O), which are also proved by the surface morphologies of the ANF network embedded with EPX (FIGS. 10P- 10Q). Close examination of the cross-sectional images show a transition from a stratified ANF network, which is similar to the compressed ANF 3DPNs (FIG. 9K), to a surface scattered with random individual ANF. The fractured surface of high content ANF composites (FIG. 10M and FIG. 10N) is analogous to that of the pure ANF film, indicating the similar failure mechanisms by breaking the strong ANF- ANF interactions. In other words, the intrinsic properties of highly branched ANF 3DPNs have a dominant contribution to the overall mechanical properties of the composite film with high ANF content. In addition, despite the addition of EPX, the thicknesses of those composites are almost invariant until a significant jump for ANF content of 38% (FIG. 10K). EPX here plays a role of binder to ANF and ANF contacts with continual filling of the nano-porous ANF network. Above a certain threshold, EPX overfills the nanopores and becomes a dominant phase in the composites with a complete surrounding around ANF. The fracture then occurs at each individual ANF rather than the collective ANF pullout (FIGS. 101- 10O). The surface of overfilled EPX composites is also less porous by appearance (FIG. 10Q). Those structural differences can influence on the mechanical and thermal expansion properties of the composites, which will be discussed below. [0090] Thanks to the high intrinsic mechanical properties of ANF 3DPNs and the uniform composite structure afforded by scgLBL, the ANF/EXP composite shows interesting mechanical properties (FIG. 11A). Overall, ANF/EPX composites are quite ductile with a plastic deformation after the initial elastic region. This characteristic is similar to those high performance aerospace alloys, such as titanium, steel or aluminum. Among those composites, [1%ANF/1%EPX]6 demonstrates optimum performance with an ultimate strength (σ„) of 505+47 MPa, an ultimate strain (¾) of 0.16+0.03 with only a density (p) of 1.5+0.1 g/cm . This observation in addition to the dynamical mechanical analysis (DMA) of those composites agrees well with the previous discussion of structural characterization of the composites. The initial improvement of mechanical properties is due to the lock-in effect of the EPX surrounding the ANF, and less defective sites as a result of filling the free space in the composites. The lock-in effect of EPX can be estimated from the decreasing damping ratio from the initial 0.14 for the pure ANF films. The excess EPX, however, leads to interface changes from ANF-ANF to ANF-EPX, which can weaken the material. In comparison to other common structural materials, the optimized ANF/EPX composites have higher specific strength (σ„ / p) than that of titanium, steel or aluminum alloys 30 (FIG. 10B). The absolute strength o u of the ANF/EXP composite is much higher than those of SAE 1010 steel (365 MPa) and 6061-T6 aluminum alloy (310 MPa). 30

[0091 ] ANF/EPX composites also excel when comparing the conventional high strength epoxy composite paragons, such as those made of Kevlar microfibers or carbon fibers. First, ANF/EPX composites are intrinsically homogenous in the nanoscale and transparent, and could made into very thin films important for electronics applications, which none of the microfibers could achieve. Second, ANF/EPX composites are isotropic at least in the plane direction. Although the optimized uniaxial Kevlar (40-45 wt%) or carbon nanofiber (60-70 wt%) composites possess very high strength in the alignment (0°) direction, they are far weaker in the 90° direction than the ANF/EPX nanocomposites. Their quasi-isotropic laminas are also inferior in o u (303 MPa and 141 MPa, respectively) 31 to the ANF/EPX examined in this work. Third, such a high o u for ANF/EPX composite, however, is not connected with the brittleness or low toughness that many composite materials have encountered. 32 ' 33 The toughness (K) of the ANF/EPX is 50.1+ 9.8 MJ/m 3 , 4 to 5 times higher than that of the unidirectional microcomposites measured in the 0° direction. The K is also much higher than the layered nanocomposites made by alumina nanoplatelets 34 or carbon nanotube 35 with greater or comparable o u . Fourth, many precise applications require zero thermal expansion of the composites. The [1%ANF/1%EPX] 6 can have the quasi-isotropic close-to-zero coefficient of thermal expansion (CTE) of -0.9 ppm K "1 till 220 °C by taking advantage of the negative thermal expansion of ANF in the axial direction. With fine tuning, a real zero expansion could be finally achieved. In comparison, the Kevlar/EPX composites have a CTE of 21.4-27.5 ppm K "1 in 90° direction, and the quasi-isotropic ones have CTE of 9.5-12.9 ppm K "1 . [0092] In conclusion, we have demonstrated the fabrication of ANF 3DPNs in the form of bulk hydrogels, fibers, and thin sheets through a simple solvent exchange process. The high mechanical properties of ANF 3DPNs can be ascribed to the high interconnectivity through the branched fibers. The loading of EPX into ANF 3DPNs lead to interesting mechanical performance of the composite through a scgLBL process, and better performed than the quasi- isotropic microfiber composites. This scgLBL approach can also be adapted for other gel- forming networks, such as CellNF, carbon nanotube, and graphene to produce other functional materials for fields in need. In addition, a variety of other functional materials able to be loaded/ assembled into the network. As a potentially biocompatible material, 36 the ANF 3DPNs could also find medical applications as a durable scaffold to replace silk or collagen networks for addressing challenges in tissue engineering. ANF composites with such combined functionalities can be used for bio-implants, packaging materials, electronic boards, bulletproof windows and many more.

References for Example 2

1. Hu, L., Hecht, D.S. & Gruner, G. Percolation in transparent and conducting carbon nanotube networks. Nano Lett. 4, 2513-2517 (2004).

2. Ramanathan, T., et al. Functionalized graphene sheets for polymer nanocomposites.

Nature Nanotech. 3, 327-331 (2008).

3. Podsiadlo, P., et al. Ultrastrong and stiff layered polymer nanocomposites. Science 318,

80-83 (2007). 4. Stankovich, S., et al. Graphene-based composite materials. Nature 442, 282-286 (2006).

5. Sahimi, M. Applications of percolation theory, (Taylor & Francis Inc. , Bristol, 1994).

6. Bigall, N.C., et al. Hydrogels and Aerogels from Noble Metal Nanoparticles. Angew Chem Int Edit 48, 9731-9734 (2009). 7. Mohanan, J.L., Arachchige, I.U. & Brock, S.L. Porous semiconductor chalcogenide aerogels. Science 307, 397-400 (2005). Aulin, C, Netrval, J., Wagberg, L. & Lindstrom, T. Aerogels from nanofibrillated cellulose with tunable oleophobicity. Soft Matter 6, 3298-3305 (2010). Gaponik, N., Herrmann, A.K. & Eychmuller, A. Colloidal Nanocrystal-Based Gels and Aerogels: Material Aspects and Application Perspectives. Phys Chem Lett 3, 8-17 (2012). Sun, Y. & Rogers, J.A. Inorganic Semiconductors for Flexible Electronics. Adv. Mater. 19, 1897-1916 (2007). Allen, B.L., Kichambare, P.D. & Star, A. Carbon Nanotube Field-Effect-Transistor- Based Biosensors. Adv. Mater. 19, 1439-1451 (2007). Pujol, T., du Roure, O., Fermigier, M. & Heuvingh, J. Impact of branching on the elasticity of actin networks. Proc. Natl. Acad. Sci. 109, 10364-10369 (2012). Gardel, M.L., et al. Elastic Behavior of Cross-Linked and Bundled Actin Networks. Science 304, 1301-1305 (2004). Broedersz, CP. & MacKintosh, F.C. Modeling semiflexible polymer networks. Rev Mod Phys 86, 995-1036 (2014). Yang, M., et al. Dispersions of ANFs: A New Nanoscale Building Block. ACS Nano 5, 6945-6954 (2011). Paakko, M., et al. Long and entangled native cellulose I nanofibers allow flexible aerogels and hierarchically porous templates for functionalities. Soft Matter 4, 2492- 2499 (2008). Schaedler, T.A., et al. Ultralight Metallic Microlattices. Science 334, 962-965 (2011). Sun, H.Y., Xu, Z. & Gao, C. Multifunctional, Ultra-Flyweight, Synergistically Assembled Carbon Aerogels. Adv. Mater. 25, 2554-2560 (2013). Xu, Y.X., Sheng, K.X., Li, C. & Shi, G.Q. Self-Assembled Graphene Hydrogel via a One-Step Hydrothermal Process. ACS Nano 4, 4324-4330 (2010). Wu, X., Moon, R. & Martini, A. Crystalline cellulose elastic modulus predicted by atomistic models of uniform deformation and nanoscale indentation. Cellulose 20, 43- 55 (2013). Tashiro, K., Kobayashi, M. & Tadokoro, H. Elastic-Moduli and Molecular-Structures of Several Crystalline Polymers, Including Aromatic Polyamides. Macromolecules 10, 413-420 (1977). Lee, C, Wei, X.D., Kysar, J.W. & Hone, J. Measurement of the elastic properties and intrinsic strength of monolayer graphene. Science 321, 385-388 (2008). Paci, J.T., Belytschko, T. & Schatz, G.C. Computational Studies of the Structure, Behavior upon Heating, and Mechanical Properties of Graphite Oxide. . Phys. Chem. Clll, 18099-18111 (2007). Robinson, J.T., et al. Wafer-scale Reduced Graphene Oxide Films for Nanomechanical Devices. Nano Lett.8, 3441-3445 (2008). Sehaqui, H., Salajkova, M., Zhou, Q. & Berglund, L.A. Mechanical performance tailoring of tough ultra-high porosity foams prepared from cellulose I nanofiber suspensions. Soft Matter 6, 1824-1832 (2010). Worsley, M.A., Kucheyev, S.O., Satcher, J.H., Hamza, A.V. & Baumann, T.F. Mechanically robust and electrically conductive carbon nanotube foams. Appl. Phys. Lett.94, 073115 (2009). Capadona, J.R., et al. A versatile approach for the processing of polymer nanocomposites with self-assembled nanofibre templates. Nat. Nanotechnol.2, 765-769 (2007). Guan, Y., et al. Fabry-Perot fringes and their application to study the film growth, chain rearrangement, and erosion of hydrogen-bonded PVPON/PAA films. . Phys. Chem. B 110, 13484- 13490 (2006). Meyers, M.A., Chawla, K.K., Hosford, W.F. & MyiLibrary. Mechanical behavior of materials, (Cambridge University Press, Cambridge ; New York, 2009). Mallick, P.K. Fiber-reinforced composites: materials, manufacturing, and design, (CRC Press, Boca Raton, FL, 2008). Cambridge Materials Selector database (grantadesign.com). Podsiadlo, P., et al. Ultrastrong and Stiff Layered Polymer Nanocomposites. Science 318, 80-83 (2007). 33. Ritchie, R.O. The conflicts between strength and toughness. Nat. Mater. 10, 817-822 (2011).

34. Bonderer, L.J., Studart, A.R. & Gauckler, L.J. Bioinspired Design and Assembly of Platelet Reinforced Polymer Films. Science 319, 1069-1073 (2008). 35. Shim, B.S., et al. Multiparameter Structural Optimization of Single-Walled Carbon Nanotube Composites: Toward Record Strength, Stiffness, and Toughness. ACS Nano 3, 1711-1722 (2009).

36. Henderson, J.D., Mullarky, R.H. & Ryan, D.E. Tissue biocompatibility of kevlar aramid fibers and polymethylmethacrylate, composites in rabbits. . Biomed. Mater. 21, 59-64 (1987).

Example 3 - Preparation of ANF gel with cylindrical shape:

[0093] 1% ANF dispersion was prepared by stirring Kevlar 69 (from Thread Exchange, right twist) in DMSO for two weeks in the presence of KOH (lg/lOOml DMSO). 4ml of the as- prepared solution was put into a cylindrical tube, and 10ml water was slowly dropped on top of Kevlar solution minimizing disturbance. The phase separation started immediately and was completed within 12 hours. Fresh water was added two times a day for four days to completely replace DMSO in the Kevlar hydrogel. In order to prepare ANF aerogel, the water in hydrogel was exchanged with ethanol, and then extracted with supercritical C0 2 .

Example 4 - Preparation of ANF continuous gel fiber:

[0094] 0.1% ANF dispersion was extruded from a 28G stainless steel needle at a rate of

3ml/h into a flow of DI water with a rate of 12ml/h. The continuous gel fiber was immediately formed at the tip of the needle and was guided into a 0.58mm (ID) glass capillary tube, and then collected in a DI water reservoir. The flow rate control was achieved by syringe pumps, and soft silicone tubing was used to make the connection.

Example 5 - Preparation of thin sheets from ANF gel:

[0095] The 1% ANF dispersion was confined between two pieces of 2" by 3" clean glass slides at a distance of -0.2 mm, and then put into water. The thickness of a gel film was controlled by a spacer between the glass slides or adapted to the weight placed on top of the glass slide balanced by the viscosity of ANF dispersion. Within 12 hours, ANF thin sheet gel can be peeled off from the glass slides under water. The gel was then transferred into fresh water for storage. Example 6 - Preparation of PVA/ANF composite:

[0096] The ANF thin sheet hydrogel was immersed in lwt % PVA (Aldrich, Mowiol ®

56-98, Mw about 195000) for 12 hours, and then rinsed with fresh water for 5mins. The thin sheet was then carefully transferred onto a Teflon sheet and dried in 70°C oven for 30 minutes. Example 7 - Spin-coating assisted gelation and layer -by -layer deposition (scg LBL) process:

[0097] 2" by 2" glass slides were cleaned by immersion in Piranha solution (3: 1

H 2 SCVH 2 O 2 ) for 12 h, then thoroughly rinsed with DI water prior to use. 1ml 1% ANF dispersion was poured onto the substrate, and then spread uniformly all over the surface by spinning with a rate of 1000 rpm and an acl of 45 for 30 s. Then DI water was quickly dropped onto the surface, the color of the coating was immediately changed from orange to white, indicating the formation of thin layer hydrogel. The substrate then went through another spin at the same settings for 30 s to remove extra water. 1ml 0.1-1% EPX in acetone was subsequently put on the hydrogel layer to allow infiltration, and a 30 s spin removed the extra EPX solution. After that, the glass slide was taken from the spin coater and put in the 100°C oven for 2 min to allow pre- annealing. This complete cycle usually took 4 min. The above procedures could be repeated to put another ANF/EPX layer on top. Typically, films made after 6 cycles of deposition were used for property measurements. The samples finally went through an overnight annealing at 70°C to completely cure the EPX and remove the solvent. The freestanding films were delaminated from glass substrate with the aid of 1% HF. For thickness measurement by ellipsometry, silicon rather than glass was used, and other procedures were kept the same.

Example 8 - Estimation of porosity in the aerogel:

[0098] The porosity was estimated by the following equation:

Porosity = 1— ^ gel

Psolid

in which, p ge i and p SO M are the density of the aerogel and its constituent respectively. The following density is used for the comparison purpose in the main text: Kevlar (1.44g/cm3), Cellulose (1.5g/cm3), Carbon nanotube (1.3 g/cm3), graphene (2.26 g/cm3), Polyethylene Glycol (1 g/cm3), Agarose (1.2 g/cm3).

Example 9 - Characterization:

[0099] The transparency of film was determined by an 8453 UV-vis ChemStation spectrophotometer from Agilent Technologies. Cross-section and morphology of the film were examined by FEI NOVA Nanolab Scanning electron microscopy (SEM) or JEOL 2100F S/TEM. Tapping mode atomic force microscopy (AFM) images were obtained using a NanoScope Ilia Atomic force microscope (AFM) from Veeco Instruments.

[00100] Differential scanning calorimetry (DSC) was carried out on a TA instrument Discovery DSC under nitrogen atmosphere at a temperature ramp rate of 20°C/min. To eliminate thermal history, the samples went through steps of heating-cooling-heating according to the protocol in ASTM D3418-08. The second heating step was used for analysis. PVA content can then be estimated by comparing the PVA melt enthalpy in the composite with that in pure PVA. Thermal gravimetrical analysis (TGA) was run on a TA instrument Discovery TGA with a heating rate of 10°C/min in nitrogen. Coefficient of thermal expansion (CTE) of films was measured using extension mode in Perkin Elmer TMA7, following ASTM Test Method for Linear Thermal Expansion of Solid Materials by Thermomechanical Analysis (E 831) and slightly modified to measure the thin film. The extension probe and grips were customized by RT instruments, Inc. to minimize the expansion of grips during the measurement. A ramp rate of 5°C/min was used and the second heating step was used for analysis.

[00101 ] The rheological (shear) measurement was conducted on a TA Instruments' AERS rheometer with a 25mm cone-plate geometry at 25 °C. Dynamic frequency sweep experiments were set from 0.06 to 60 rad/s at a fixed oscillatory strain of 1%. The strain sweep experiments were set from 0.1% to 100% at a fixed frequency of 6 rad/s. The samples were covered with a thin layer of silicon oil to prevent evaporation of water. Uniaxial tensile testing was done on RSAIII Rheometrics Systems Analyzer from TA instruments. The tensile tests confirm to the ASTM standard ASTM D882. In a typical measurement, a 1mm wide and 6mm long sample strip was fixed onto the steel grips. The Kevlar microfiber was fixed by super- gluing the ends onto two pieces of stainless steel metal sheets separated by a distance of 6mm. The metal sheets were then put between the grips for measurement. The test speed was O.Olmm/s. A total of 6 measurements were made for 2 batches of PVA/ANF composite. A total of 10 measurements were made for 2 batches of ANF/EPX composites. A total of 3 measurements were made for the Kevlar microfiber. Example 10 - Mechanical Properties ofANF 3DPNs:

[0102] ANF hydrogels and aerogels also exhibit satisfying performance in the compression test. The compressive strain-stress curves show three stages typical for porous materials (FIG. 3E) 1 . The linear elastic stage was observed initially, then the material reached its elastic limit, at which point the 3DPNs started to yield at a nearly constant stress. This plateau stage is followed by a densification region where the porous network starts to collapse. Both gels can be compressed to strains over 90% without any cracks at the macro-, micro-, or nanoscale (FIGS. 91, 9J, and 9K). Other 3DPNs exhibit substantially higher brittleness. For example, graphene hydrogels show micro-cracks and stress discontinuity at a strain of 42% , and cellulose aerogels become completely fractured at 68% strain .

[0103] The compressive modulus E and yield stress a y for the ANF hydrogel are 57+3 kPa and 8+1 kPa. ANF aerogel has twice higher values, E=90+5 kPa and σ,,=18+1 kPa. Note, however, the plasticization effect of water on ANF hydrogel is less pronounced as for other 3DPNs. For example, highly porous (>99%) CellNF and CNT hydrogels are so compliant that they are reminiscent of a viscous fluid. 3- " 5 Their 3D structure is disturbed by swelling in water and/or agitation. Instead, the ANF gels are stable in water for over a year without any visible fluidization. Even intensive sonication cannot destroy their structural integrity, while other un- crosslinked CNT or CellNF cannot survive the same treatment. [0104] We also made a comparison of the compressive properties of ANF 3DPNs to other commonly used reinforcing networks in the form of aerogels. The E of ANF aerogel is similar to CellNF or CNT aerogel, but 2 orders of magnitude higher than graphene aerogel at similar density. The similarity is likely due to the insufficient load transfer between the ligaments during the compression, 6 unlike the network deformation during the shearing. The nanofibers or nanotubes are likely to buckle or dislocate under the stress due to their small diameters, as the bending modulus is of the order of Efi ber d 4 , in which £/ ¾er and d are elastic modulus and diameter of the fiber . The graphene aerogel is especially prone to be bent due to its extreme thinness (~1 nm).

[0105] Toughness is one of the key properties determining most applications of load- bearing and other materials. Importantly, ANF 3DPNs show high toughness in compression with 25 kJ/m 3 and 78 kJ/m 3 for hydrogel and aerogel, respectively. This property is higher than SWNT and CellNF aerogels, and some polymeric gels with much higher solid content (Table 1). Table 1 Comparison of mechanical properties of various hydrogels.

[0106] The tensile properties of ANF hydro- and aero- gels are further studied. The

Young's modulus E y , ultimate stress a u , and ultimate strain s u of ANF hydrogel in tension are 230+18 kPa, 24+4 kPa, and 13+2%. The same parameters for ANF aerogel are E y =750+10 kPa, σ Μ =90+7 kPa, and ¾=12+3%. The Young's modulus (E y ) of the ANF 3DPNs in extension is much higher than that obtained in compression. Due to the fluidity of CNT, CellNF, and graphene hydrogels at similar solid content, their tensile properties are not available to compare. Still, the E y of ANF hydrogel is 10 times higher than that PEG hydrogels with much higher solid content (Table S2).

[0107] From the practical perspective, it might be useful to increase tensile properties sacrificing the porosity. After ANF aerogel is compacted into l/6th of its initial volume, it displays E y , s u , and a u of 16+2 MPa, 11+2%, and 1.3+0.7 MPa, respectively. These properties can be further improved to 136+1 IMPa, 7+2%, and 6.2+0.5 MPa when the aerogel is further compressed. Such enhancement is likely to originate from the increased density of hydrogen- bonding crosslinks and the nanofiber alignment in the film. The densified ANF 3DPNs have ultimate stress comparable to o u of CNT buckypaper but with seven times higher ultimate strain.

Example 11. PVA/ANF composites through diffusion ofPVA into thin ANF gel: [0108] Polymers can be impregnated into the 3DPNs from ANFs to facilitate stress transfer and improve the defect tolerance and toughness. 9 Such impregnation can be easily implemented for ANF 3DPNs by immersing the aero/hydrogel into various solutions of required components. Polyvinyl alcohol (PVA) was selected as a soft matrix component due to its abundant -OH groups capable of hydrogen-bonding with ANFs. When the ANF gel was treated with 1 wt % PVA solution, the polymer chains adsorbed strongly onto the exposed surface of ANFs and long-term rinsing had little influence on the PVA content.

[0109] The PVA-saturated gel was dried at 70 °C into a transparent solid film (FIG. 12A, 12B). The morphology of the resulting film can be described as interpenetrating PVA and ANF networks collapsed from the capillary force during the water removal (FIG. 12C and 12D). The irregular micro/nano pores (FIG. 12D) in the structure left by this drying process, however, can act as scattering interfaces to reduce the light transmission. The light scattering can be minimized by filling these pores with transparent epoxy; 10 the 1.25 μιη thick film show to 86% transparency at 600 nm (FIG. 12E).

[0110] ANFs content in the composite is 35wt% of the as determined by TGA and

DSC analysis. Their uniform distribution in the material can be easily observed in the cross- section image of the fractured composite (FIG. 12C). FTIR spectra confirm the interactions between ANF and PVA (FIGS. 12F-12H). The presence of hydrogen bonds are revealed in the change in the y(C=0) position. The band at 1646 cm "1 for intra- ANF hydrogen-bond- influenced C=0 does not change appreciably, but the other C=0 band for inter-fiber hydrogen bonds is upshifted by 0.8 cm "1 (FIG. 12H). This observation suggests that -OH groups from PVA compete with C=0 as hydrogen acceptors, thus increasing the electron density in the C=0 unit. The bending (5(CH 2 )) and rocking modes (5 R (CH 2 )) of -CH 2 - groups disappear in the composite spectra. 11 It is likely that the strong van der Waals interactions from phenylene groups in ANF limit the movement of -CH 2 - units in PVA.

[0111] From the stress-strain curve, the PVA/ ANF composite has ou = 257+9 MPa and su = 27+5%. The toughness of the PVA/ ANF composite is 46+3 MJ/m which is almost twice as high as that of KMF - the benchmark for tough materials. In the previous studies, reasonably high o u and s u parameters were attained through the more time- and labor-intensive bottom-up method, such as the laminated chitosan/alumina platelet composites 12 with o u of

315+95 MPa and s u oi 21+5%, and layer-by-layer assembled PVA and CNT composites 13 with o u of 225+25 MPa and s u of 19+7%. The mechanical properties of PVA/ANF films described herein, made through a simple impregnation process, are comparable or surpass the properties of the existing composites just described.

[0112] The inclusion of ANF, which has an unusual negative coefficient of CTE in the axial direction, can greatly reduce the overall CTE of the composite. 14 Below glass transition temperature (r g ), the PVA/ANF composite has a CTE of 1.9 ppm K "1 , which is smaller than most of the ceramics, such as glass, silicon and boron carbide 1 , above T g , the composite has a CTE of 32 ppm K "1 , close to that of neat PVA in the glassy state.

Example 12 - Dynamic Mechanical Analysis of ' ANF /EPX:

[0113] Another interesting property in dynamic mechanical properties measurement (FIG. 13A and 13B) is the damping ratio or tan δ, which is the ratio between the loss modulus and storage modulus. Tan δ measures the degree to which a material dissipates the vibration energy into heat. High damping capacity is useful in many automotive and sporting goods applications. 15 The carbon and aramid micro-composites typically have very low tan δ of 0.0024 and 0.018 respectively. 16 The ANF/EPX composites show higher tan δ (FIG. 13B). Interestingly, pure ANF film displays the largest tan δ over the range of 0.1- lHz. At 0.1 Hz, the tan δ of ANF film can be as high as 0.14, and is decreased to 0.06 at 1 Hz. With the addition of EPX, the tan δ gradually declines. The highest tan δ of pure ANF film can be related with the more freedom of ANFs in the periphery of abundant unfilled nanopores. The strong but unlocked ANF-ANF interfaces cause the high mechanical damping. 17 With the introduction of EPX into the 3DPN, the touching interface is gradually locked by the poor- damping EPX to result in a lower tan δ. [1%ANF/1%EPX] 6 has slightly lower tan δ of 0.11 at 0.1 Hz and 0.05 at 1 Hz. An significant drop of tan δ occurs for [1%ANF/2%EPX] 6 , which has even smaller tan δ than EPX (FIG. 13B). This finding actually agrees with the previously discussed structure transition. In this film, ANF is surrounded by EPX with the interface of ANF-EPX-ANF. The dissipating mechanism relying on the ANF-ANF friction disappears here, while the ANF serves as reinforcing agent for EPX to make its chain even less mobile.

Example 13 - Thermal Expansion Properties of ANF/EPX:

[0114] Low CTE is another key feature for conventional carbon or aramid micro- composites. 15 The low CTE can exhibit a better dimensional stability over a wide temperature range. Unidirectional carbon micro-composites have CTE of -0.44-0.16 ppm K _1 in 0° direction, and 0.36-4.02 ppm K "1 in 90° direction, while the quasi-isotropic ones have CTE of 0.36-4.02 ppm K "1 . 16 On the other hand, unidirectional aramid micro-composites have CTE of -2.57—1.74 ppm K "1 in 0° direction, and 21.4-27.5 ppm K "1 in 90° direction, while the quasi- isotropic ones have CTE of 9.5-12.9 ppm K "1 . 16 Amazingly here, [1%ANF/1%EPX] 6 can have the quasi-isotropic close-to-zero CTE of -0.9 ppm K "1 till 220 °C. With finer tuning, a real zero expansion could be achieved. It is interesting to notice that the aramid micro-composites have high positive CTE at 90° to the fiber direction, and this property contributes to the slightly lower but still positive CTE for quasi-isotropic composites. However, material made with ANF with the same chemical composition but much smaller diameter gives overall negative and even zero CTE.

[0115] This phenomenon is related with the existence of nanopores. Aramid microfiber is proved to have CTE of -4.9 ppm, agreeing with the previous studies. 18 Pure ANF film shows two regimes of thermal expansion: one with CTE of -6 ppm K "1 till 75 °C, and the other one with CTE of -0.5 ppm K "1 . Similar to graphene, 19 ' 20 the negative CTE in ANF is contributed by the transversal acoustic bending modes along the axis, 21 ' 22 or commonly known as the

"membrane effect". 23 The nanopores formed by the overlapping of ANFs give rise to more free space to enhance this bending effect, thus more negative CTE. 22 With the increase of temperature, however, other phonon modes contributing to the positive CTE might take effect 21 1 The nanopores can also accommodate the positive radial expansion of ANF. In addition, since the mechanical load is carried mainly by the axial direction of the ANF, the overall CTE shows more axial behavior of ANF. With EPX in the composite, the thermal behavior is a combined effect of both components. When the nanopores are underfilled, the CTE increases a bit even with 36% EPX in the composites. When the nanopores are overfilled, such as in the film of [1%ANF/2%EPX] 6 , the CTE has increased by a lot to 11 ppm. In this film, no enough space exists to accommodate the positive radial expansion, and EPX as a matrix can uniformly distribute the load to various directions in ANF. Accordingly, the radial expansion can contribute more to the overall CTE.

Example 14 - scgLBL in comparison to other technique:

[0116] To be noted here, pure ANF film without EPX shows a u of 387+25 MPa, e u of

0.16+0.03, E' of 11.5+0.5 GPa, which are much higher than those from VAF made ANF film The mechanical performance enhancement can result from this elegant gLBL processing technique. In VAF, the long filtration step can lead to the deterioration of solution qualities, and thus the generation of defects in the final film.

Additionally, hand-peeling the ANF film from the porous membrane can also break some film microstructures. In gLBL, all those defects-introduction steps are avoided. Besides, some mechanical factors in the spin coating process, such as the centrifugal and air shear force can usually lead to some degree of lateral chain orientation and stratification. 25 ' 26 These factors can lead to an improved mechanical performance for the gLBL made film with the same chemical composition, but with the different microarchitecture. References for Examples 3 - 14

1. Ashby, M.F. Materials selection in mechanical design, (Elsevier, Maryland Heights, 2005).

2. Xu, Y.X., Sheng, K.X., Li, C. & Shi, G.Q. Self-Assembled Graphene Hydrogel via a One-Step Hydrothermal Process. ACS Nano 4, 4324-4330 (2010).

3. Paakko, M., et al. Long and entangled native cellulose I nanofibers allow flexible aerogels and hierarchically porous templates for functionalities. Soft Matter 4, 2492- 2499 (2008).

4. Hough, L.A., Islam, M.F., Hammouda, B., Yodh, A.G. & Heiney, P. A. Structure of Semidilute Single-Wall Carbon Nanotube Suspensions and Gels. Nano Lett. 6, 313-317

(2006) .

5. Paakko, M., et al. Enzymatic Hydrolysis Combined with Mechanical Shearing and High-Pressure Homogenization for Nanoscale Cellulose Fibrils and Strong Gels. Biomacromolecules 8, 1934-1941 (2007). 6. Meza, L.R., Das, S. & Greer, J.R. Strong, lightweight, and recoverable three- dimensional ceramic nanolattices. Science 345, 1322-1326 (2014).

7. Broedersz, CP. & MacKintosh, F.C. Modeling semiflexible polymer networks. Rev Mod Phys 86, 995-1036 (2014).

8. Xu, G.H., Zhang, Q., Zhou, W.P., Huang, J.Q. & Wei, F. The feasibility of producing MWCNT paper and strong MWCNT film from VACNT array. Appl. Phys. A: Mater.

Sci. Process. 92, 531—539 (2008).

9. Capadona, J.R., et al. A versatile approach for the processing of polymer nanocomposites with self-assembled nano fibre templates. Nat. Nanotechnol. 2, 765-769

(2007) . 10. Yano, H., et al. Optically Transparent Composites Reinforced with Networks of Bacterial Nanofibers. Adv. Mater. 17, 153-155 (2005).

11. Krimm, S., Liang, C.Y. & Sutherland, G.B.B.M. Infrared spectra of high polymers. V.

Polyvinyl alcohol. J. Polym. Sci. 22, 227-247 (1956).

12. Bonderer, L.J., Studart, A.R. & Gauckler, L.J. Bioinspired Design and Assembly of Platelet Reinforced Polymer Films. Science 319, 1069-1073 (2008). 13. Shim, B.S., et al. Multiparameter Structural Optimization of Single-Walled Carbon Nanotube Composites: Toward Record Strength, Stiffness, and Toughness. ACS Nano 3, 1711-1722 (2009).

14. Zhu, J., et al. Pseudonegative Thermal Expansion and the State of Water in Graphene

Oxide Layered Assemblies. ACS Nano 6, 8357-8365 (2012).

15. Mallick, P.K. Fiber-reinforced composites: materials, manufacturing, and design, (CRC Press, Boca Raton, FL, 2008).

16. Cambridge Materials Selector database (grantadesign.com).

17. Suhr, J., Koratkar, N., Keblinski, P. & Ajayan, P. Viscoelasticity in carbon nanotube composites. Nat. Mater. 4, 134-137 (2005).

18. Jain, A. & Vijayan, K. Kevlar 49 fibres: Thermal expansion coefficients from high temperature X-ray data. Curr. Sci. 78, 331—335 (2000).

19. Bao, W.Z., et al. Controlled ripple texturing of suspended graphene and ultrathin graphite membranes. Nature Nanotech. 4, 562— 566 (2009). 20. Nelson, J.B. & Riley, D.P. The thermal expansion of graphite from 15°C to 800°C.

Proc. Phys. Soc. 57, 477—486 (1945).

21. Mounet, N. & Marzari, N. First-principles determination of the structural, vibrational and thermodynamic properties of diamond, graphite, and derivatives. Phys. Rev. B 71, 205214 (2005). 22. Grigoriadis, C, Haase, N., Butt, H.J., Mullen, K. & Floudas, G. Negative thermal expansion in discotic liquid crystals of nanographenes. Adv. Mater. 22, 1403— 1406 (2010).

23. Lifshitz, I. Thermal properties of chain and layered structures at low temperatures. Zh.

Eksp. Teor. Fiz. 22, 475—486 (1952). 24. Cao, K., et al. Reactive Aramid Nanostructures as High-Performance Polymeric Building Blocks for Advanced Composites. Adv. Funct. Mater. 23, 2072-2080 (2013).

25. Ariga, K., Hill, J.P. & Ji, Q. Layer-by-layer assembly as a versatile bottom-up nanofabrication technique for exploratory research and realistic application. Phys. Chem. Chem. Phys. 9, 2319-2340 (2007). Cho, J., Char, K., Hong, J.D. & Lee, K.B. Fabrication of highly ordered multilayer films using a spin self-assembly method. Adv. Mater. 13, 1076-1078 (2001). Sun, H.Y., Xu, Z. & Gao, C. Multifunctional, Ultra-Flyweight, Synergistically Assembled Carbon Aerogels. Adv. Mater. 25, 2554-2560 (2013). Worsley, M.A., Kucheyev, S.O., Satcher, J.H., Hamza, A.V. & Baumann, T.F. Mechanically robust and electrically conductive carbon nanotube foams. Appl. Phys. Lett. 94, 073115 (2009). Sehaqui, H., Salajkova, M., Zhou, Q. & Berglund, L.A. Mechanical performance tailoring of tough ultra-high porosity foams prepared from cellulose I nanofiber suspensions. Soft Matter 6, 1824-1832 (2010). Aamer, K.A., Sardinha, H., Bhatia, S.R. & Tew, G.N. Rheological studies of PLLA- PEO-PLLA triblock copolymer hydrogels. Biomaterials 25, 1087-1093 (2004). Storm, C, Pastore, J.J., MacKintosh, F.C., Lubensky, T.C. & Janmey, P. A. Nonlinear elasticity in biological gels. Nature 435, 191-194 (2005). Hobbie, E.K. & Fry, D.J. Rheology of concentrated carbon nanotube suspensions. . Chem. Phys. 126(2007). Fan, Z.H. & Advani, S.G. Rheology of multiwall carbon nanotube suspensions. J Rheol 51, 585-604 (2007). Lee, C, Wei, X.D., Kysar, J.W. & Hone, J. Measurement of the elastic properties and intrinsic strength of monolayer graphene. Science 321, 385-388 (2008). Mihranyan, A., Edsman, K. & Str0mme, M. Rheological properties of cellulose hydrogels prepared from Cladophora cellulose powder. Food Hydrocolloids 21, 267- 272 (2007). Wu, X., Moon, R. & Martini, A. Crystalline cellulose elastic modulus predicted by atomistic models of uniform deformation and nanoscale indentation. Cellulose 20, 43- 55 (2013). Wang, Q., et al. High-water-content mouldable hydrogels by mixing clay and a dendritic molecular binder. Nature 463, 339-343 (2010). 38. Tashiro, K., Kobayashi, M. & Tadokoro, H. Elastic-Moduli and Molecular-Structures of Several Crystalline Polymers, Including Aromatic Polyamides. Macromolecules 10, 413-420 (1977).

39. Kumar, P., Maiti, U.N., Lee, K.E. & Kim, S.O. Rheological properties of graphene oxide liquid crystal. Carbon 80, 453-461 (2014).

40. Paci, J.T., Belytschko, T. & Schatz, G.C. Computational Studies of the Structure, Behavior upon Heating, and Mechanical Properties of Graphite Oxide. . Phys. Chem. C lll, 18099-18111 (2007).

41. Robinson, J.T., et al. Wafer-scale Reduced Graphene Oxide Films for Nanomechanical Devices. Nano Lett. 8, 3441-3445 (2008).

42. Nakayama, A., et al. High Mechanical Strength Double-Network Hydrogel with Bacterial Cellulose. Adv. Fund. Mater. 14, 1124-1128 (2004).

43. DeKosky, B.J., et al. Hierarchically designed agarose and poly(ethylene glycol) interpenetrating network hydrogels for cartilage tissue engineering. Tissue Eng Part C Methods 16, 1533-1542 (2010).

44. Temenoff, J.S., Athanasiou, K.A., Lebaron, R.G. & Mikos, A.G. Effect of poly(ethylene glycol) molecular weight on tensile and swelling properties of oligo(poly(ethylene glycol) fumarate) hydrogels for cartilage tissue engineering. . Biomed. Mater. 59, 429-437 (2002).

45. Bryning, M.B., et al. Carbon nanotube aerogels. Adv. Mater. 19, 661-664 (2007).

46. Kim, K.H., Oh, Y. & Islam, M.F. Graphene coating makes carbon nanotube aerogels superelastic and resistant to fatigue. Nat Nano 7, 562-566 (2012).

Example 15 - Materials: [0117] 1% ANF dispersion was prepared by stirring Kevlar 69 (from Thread Exchange, right twist) in DMSO for two weeks at the presence of over saturated KOH (Ig/lOOml DMSO). System 2000 epoxy resin and 2020 epoxy hardener were bought from Fibre Glast. The two parts were mixed together in the ratio of 3: 1, and then diluted with acetone (Aldrich) to the needed concentration. Example 16 - Typical procedures for gelation assisted layer-by-layer deposition: [0118] 2" by 2" glass slides were cleaned by immersion in Piranha solution (3: 1

H 2 S VH 2 O 2 ) for 12 h, then thoroughly rinsed with DI water prior to use. 1ml 1% ANF dispersion was poured onto the substrate, and then spread uniformly all over the surface by spinning with a rate of 1000 rpm and an acl of 45 for 30 s. Then DI water was quickly dropped onto the surface, the color of the coating was immediately changed from orange to white, indicating the formation of thin layer hydrogel. The substrate then went through another spin at the same settings for 30 s to remove extra water. 1ml 0.1-1% EPX in acetone was subsequently put on the hydrogel layer to allow infiltration, and a 30 s spin removed the extra EPX solution. After that, the glass slide was taken from the spin coater and put in the 100°C oven for 2 min to allow pre-annealing. This complete cycle usually takes 4 min. The above procedures could be repeated to put another ANF/EPX layer on top. Typically, films made after 6 cycles of deposition were used for property measurements. The samples finally went through an overnight annealing at 70°C to completely cure the EPX and remove the solvent. The freestanding films were delaminated from glass substrate with the aid of 1% HF. For thickness measurement by ellipsometry, silicon rather than glass was used, and other procedures were kept the same.

Example 17 - Characterization:

[0119] The transparency of film was determined by an 8453 UV-vis ChemStation spectrophotometer from Agilent Technologies. Cross-section and morphology of the film were examined by FEI NOVA Nanolab Scanning electron microscopy (SEM). Tapping mode atomic force microscopy (AFM) images were obtained using a NanoScope Ilia Atomic force microscope (AFM) from Veeco Instruments. Thicknesses of thin films were measured by a BASE- 160 spectroscopic ellipsometer produced by J.A.Woolam Co., Inc.

[0120] Thermal gravimetrical analysis (TGA) was run on a TA instrument Discovery TGA with a heating rate of 10°C/min in nitrogen. Coefficient of thermal expansion (CTE) of films was measured using extension mode in Perkin Elmer TMA7 following ASTM Test Method for Linear Thermal Expansion of Solid Materials by Thermomechanical Analysis (E 831) and slightly modified to measure the thin film. The extension probe and grips were customized by RT instruments, Inc. to minimize the expansion of grips during the measurement. A ramp rate of 5°C/min was used and the second heating step was used for analysis.

[0121 ] Uniaxial tensile testing was done on RSAIII Rheometrics Systems Analyzer from TA instruments. The tensile tests confirm to the ASTM standard ASTM D882. In a typical measurement, 1mm wide and 6mm long sample strip was fixed onto two pieces of stainless steel metal sheets. The metal sheets were then put between the grips for measurement. The test speed is 0.01 mm/s. A total of 10 measurements were made for 2 batches of ANF/EPX composite. Example 18 - Ion conducting membranes:

[0122] The use of ANF to make ion conducting membranes is disclosed in US provisional serial number 61/941785 filed on February 19, 2014, and in international application PCT/US2015/016675 filed February 19, 2015. The entire contents of the two applications is hereby incorporated by reference. [0123] In the incorporated applications, an ion conducting membrane (ICM) is built from ANF and used in a wide variety of batteries as a separator between two electrodes, being an anode and a cathode, or being a positive electrode and a negative electrode. An advance over the art involves direct coating of an ANF material onto one of the electrodes or onto a existing separator. The coating basically provides an ANF based separator or ICM. Fabrication of an electrode assembly is simplified because the coated electrode can be directly assembled adjacent the other electrode so that only two parts are needed for assembly instead of all three (i.e., two electrodes and the separator).

[0124] The direct coating method is to apply an ICM solution/mixture/slurry containing

ANF onto an existing electrode. In addition to ANF, the solution/mixture/slurry compositions optionally include organic fibers (such as cellulose microfibers, cellulose nanofibers, and poly(p-phenylene-2,6-benzobisoxazole) (PBO) nanofibers), inorganic nanomaterials (such as ceramic nanoparticles, nanowires, and the like), as well as other components disclosed in the incorporated documents. These include components like polyelectrolytes, polymer nanoparticles, and so on. [0125] The goal of this method is to provide a thermally stable, electrically insulating, ion conducting and mechanically stiff ICM to provide dendrite suppression, thermal runaway prevention, and other benefits. The close contact between the electrode and ICM will also provide benefits in lowered interfacial resistance, easier handling of thin separators (100 nm - ΙΟμιη) and simplified assembly of batteries. This direct coating can act as a thermal stability enhancing layer for electrodes, where the layer is to reduce the current generated during a hard shorting situation (e.g., nail penetration), reducing the heat generated and thus preventing thermal runaway. [0126] A coated electrode made with these methods can be combined with another electrode to form an electrode assembly either with or without a free-standing separator. In the case of an electrode assembly with a free-standing separator, the ANF coating on the electrode will act solely as a thermal stability enhancing layer to prevent thermal runaway. In the case of an electrode assembly without a free-standing separator, the ANF coating on the electrode will act as an ICM and will assume the role both of a separator and of a thermal stability enhancing layer.

[0127] The electrodes can be any commercially available electrodes. Some are made of carbon with polymeric binders. Others include electrodes of lithium metal or any lithium active material like lithium titanate. Silicon containing anodes can also be used. The electrodes and the assemblies made from them find use in fuel cells and in batteries such as lithium batteries.

[0128] Since the ANF coating is applied directly to the electrode, the separator/electrode is a one piece design, giving close contact between the separator and electrode. In a traditional battery assembly, where there is cathode, separator, and anode, there are three components to assemble and align together. With a coated electrode, it is only necessary to assemble two components since the separator is already coated on one of the electrodes.

[0129] A non-limiting way of fabricating an ANF coated electrode involves the following illustrative steps: [0130] Mix ANF ICM solution/mixture/slurry with desired components (AI 2 O 3 nanoparticles, cellulose fibers, polyelectrolytes etc.) The ANF composition may range from 100% to 10% of the solids in the solution, mixture, or slurry.

[0131 ] Fix Electrode onto a glass substrate.

[0132] Pour ANF ICM solution/mixture/slurry onto the electrode and coat evenly using a doctor blade.

[0133] Wash the assembly in protic solvent such as water, ethanol, isopropyl alcohol etc.

[0134] After complete solvent exchange, dry the ANF ICM/Electrode assembly under compression in a vacuum oven at 60°C. [0135] The completed assembly should have a ANF ICM adhered evenly to the electrode. The ANF ICM may detach from the electrode when electrolytes are added to the battery cell during assembly, however, in the dry state, the ANF ICM should be adhered to the electrodes.

[0136] Basically, the process of coating an electrode involves using the electrode material as a substrate in the coating methods described herein and in the incorporated documents. An ANF slurry is deposited on the electrode substrate and then is phase transformed to make a gel, which is a hydrogel when water is used for the phase transformation. After optional solvent exchange the ANF coating is dried to form a dried gel as a coating on the electrode. Compression vacuum is use during drying to prevent wrinkles and enable the coating to come out flat and even.

[0137] This process can be scaled-up with several continuous coating methods, such as dip coating, slot die coating, micro gravure coating. The choice of coating methods will depend on the required output (m/s), and coating thickness.

Working Example:

1. Prepare a 2 wt% ANF suspension in an aprotic solvent such as DMSO

2. Add alumina nanoparticles (Sigma Aldrich, 50nm) 1.4g into 3 lg of the 2% ANF suspension by rigorous planetary mixing at ~1500rpm for 15 minutes.

3. Carbon anode is fixed onto a glass substrate and flattened.

4. The Alumina/ ANF mixture is poured over the anode and spread evenly using a doctor blade. The coating thickness ranges from 300μιη to 50μιη.

5. The coated anode is submerged into an ethanol bath for 30 minutes to allow gelation and solvent exchange.

6. The coated anode is removed from the ethanol bath and sandwiched between two glass slides and dried in a vacuum oven overnight at 60°C.

7. After the coated anodes are dried, circular electrodes are punched out and further dried for 2 hours at 110°C under vacuum.

8. The coated electrodes are then assembled into coin cells with a NMC cathode with no separator needed.

9. FIG. 14 is a photograph of a coated anode after step 6 described above. A yellowish top layer (light color in black and white) is the ANF composite (or ANF ICM), the dark layer underneath is the carbon electrode. The copper foil is the copper current collector. The carbon and copper collector are typical of a lithium ion battery anode.