POROUS NANOSTRUCTURED POLYIMIDE NETWORKS AND METHODS OF MANUFACTURE
Porous three-dimensional networks of polyimide and porous three-dimensional networks of carbon and methods of their manufacture are described. For example, polyimide aerogels are prepared by mixing a dianhydride and a diisocyanate in a solvent comprising a pyrrolidone and acetonitrile at room temperature to form a sol-gel material and supercritically drying the sol-gel material to form the polyimide aerogel. Porous three-dimensional polyimide networks, such as polyimide aerogels, may also exhibit a fibrous morphology. Having a porous three-dimensional polyimide network undergo an additional step of pyrolysis may result in the three dimensional network being converted to a purely carbon skeleton, yielding a porous three-dimensional carbon network. The carbon network, having been derived from a fibrous polyimide network, may also exhibit a fibrous morphology.
Latest Aerogel Technologies, LLC Patents:
This application claims the benefit of U.S. Provisional Patent Application Ser. No. 61/375,656, filed Aug. 20, 2011, entitled “Porous Nanostructured Polyimide Networks and Methods of Manufacture,” by Leventis et al.FEDERALLY SPONSORED RESEARCH
Research leading to various aspects of embodiments presented herein were sponsored, at least in part, by the National Science Foundation, Grant No. CHE-0809562 and Grant No. CMMI-0653919. The United States Government may have certain rights in the invention.BACKGROUND
Aspects herein relate to nanostructured polyimide networks, carbonized derivatives, related materials, uses thereof, and methods of manufacture.
2. Discussion of Related Art
Three-dimensional porous architectures are a desirable form factor for many materials as they allow installation of new properties into a material not possessed by the non-porous form of a material. Porous architectures possessing nanostructured features, such as nanopores or nanoparticulate solid frameworks, are further desirable in many cases as they can possess new and/or more extreme properties than porous architectures without nanostructured features.
Aerogels are an example of a porous architecture possessing nanostructured features. Aerogels are materials comprised of three-dimensional assemblies of nanoparticles or nanostructures that exhibit high porosity materials and ultra-low densities Aerogel materials are typically produced by forming a gel that includes a porous solid component and a liquid component and then removing the liquid component by supercritically, subcritically, or freeze drying the wet gel to isolate the porous solid component. This porous solid component is an aerogel. Supercritical drying involves the liquid being transformed into a fluid above its critical point and removing the fluid while leaving the porous solid structure generally intact. Subcritical drying involves evaporation of the liquid below its critical point in a way that leaves the porous solid structure generally intact. Freeze drying involves freezing of the liquid component and sublimation of the resulting solid in a way that leaves the porous solid structure generally intact.
The large internal void space in aerogels and other nanostructured and non-nanostructured three-dimensional porous networks generally provides for a low dielectric constant, a low thermal conductivity, and a high acoustic impedance. These materials have been considered for a number of applications including thermal insulation, lightweight structures, and impact resistance.SUMMARY
Articles and methods for manufacturing three-dimensional porous polyimide networks and three-dimensional porous carbon networks are described.
Three-dimensional porous polyimide networks can be prepared by mixing a dianhydride and a diisocyanate in a solvent comprising a pyrrolidone and acetonitrile at room temperature to form a sol-gel material and subsequently drying the sol-gel material to form a polyimide aerogel. Drying may be performed supercritically, subcritically, or by freeze drying. In some embodiments, a polyimide aerogel manufactured by methods described herein may have a fibrous morphology which may or may not also include particulate features.
Three-dimensional porous carbon network aerogels may also be manufactured through a conversion step from polyimide aerogels. Once a polyimide aerogel is formed, the aerogel may be subjected to a pyrolysis step, giving rise to a carbon skeleton in the aerogel, hence, a carbon aerogel. In some embodiments, a polyimide aerogel having a fibrous morphology that is subject to the pyrolysis step may give rise to a carbon aerogel also having a fibrous morphology.
In some cases, three-dimensional porous polyimide networks not considered aerogels may be produced. Likewise, three-dimensional porous carbon networks not considered aerogels may be derived from such three-dimensional polyimide networks.
Various embodiments of the present invention provide certain advantages. Not all embodiments of the invention share the same advantages and those that do may not share them under all circumstances.
Further features and advantages of the present invention, as well as the structure of various embodiments of the present invention are described in detail below with reference to the accompanying drawings.
The accompanying drawings are not intended to be drawn to scale. For purposes of clarity, not every component may be labeled in every drawing. Various embodiments of the invention are described, by way of example, in the accompanying drawings. In the drawings:
Porous three-dimensional networks of polyimide and porous three-dimensional networks of carbon and methods of their manufacture are described herein. In some embodiments, three-dimensional porous polyimide networks (e.g., aerogels) are prepared by mixing a dianhydride and a diisocyanate in a solvent comprising a pyrrolidone and an acetonitrile at room temperature to form a sol-gel material. The sol-gel material may be dried supercritically or subcritically so as to form the three-dimensional porous polyimide network. The three-dimensional porous polyimide network may also exhibit a fibrous morphology. In some embodiments, the three-dimensional porous polyimide network may be subject to an additional step of pyrolysis resulting in the three dimensional network of polyimide being converted to a carbon skeleton, yielding a three-dimensional porous carbon network. The three-dimensional porous carbon network, having been derived from a fibrous three-dimensional porous polyimide network, may also exhibit a fibrous morphology.
Three-dimensional porous polyimide networks may be prepared via a one-step room-temperature synthesis route involving reaction of a dianhydride with a diisocyanate. Using an isocyanate to form the polyimide network using methods described herein provides certain advantages: a) CO2 is the only substantial byproduct; b) sacrificial dehydrating agents (e.g., acetic anhydride/pyridine) are not required for gelation; c) gelation and imidization take place at room temperature and, thus, post-gelation curing at high temperatures (e.g., 190° C.) in order to complete imidization is not required; d) use of isocyanates is cheaper than use of amines from the traditional route; e) higher density bifunctional aerogels for dual-purpose applications (e.g. thermal insulators that are also used as structural components; impact dampening materials that are also used as structural components) are easily accessible with no solubility issues in turn extending the accessible density range upwards, while the polyamic acid route encounters solubility issues at higher concentration sols; and f) an entangled fiber nanomorphology stabilizes the nanostructure, extending the accessible density range downwards.
Three-dimensional porous polyimide networks such as polyimide aerogels may be manufactured using a route that involves condensation of a dianhydride and a diamine to form a polyamic acid solution that is dehydrated with an anhydride to form polyimide. In some cases, use of a sacrificial dehydration step is generally expensive with respect to the cost of materials and removal of the byproducts. A further complication may be the kinetic formation of isoimides which can be converted to more thermodynamically stable imides by high-temperature treatment. To accommodate this, the dehydration step may be carried out at elevated temperatures directly (>130° C.). Preparation of certain polyimide aerogels are described in U.S. Pat. No. 7,074,880 entitled “Polyimide aerogels, carbon aerogels, and metal carbide aerogels and methods of making same” and is incorporated herein by reference in its entirety.
Polyimide aerogels synthesized via a conventional route using diamine (PI-AMN) as compared to routes contemplated by the inventors and described herein using isocyanate (PI-ISO) may have similar surface areas (e.g., for 20% solid formulation PI-ISO: 351 m2/g, density=0.196 g/cm3; PI-AMN: 377 m2/g, density=0.291 g/cm3). However, the nanomorphologies of polyimide aerogels prepared using isocyanate may be completely different from nanomorphologies of polyimide aerogels prepared using diamine. For example, for a similar chemical composition, PI-ISO may have a fibrous morphology, while PI-AMN have a particulate morphology. In general, PI-ISO may also be more robust and may shrink less compared to PI-AMN. Upon pyrolysis, polyimide aerogels can be converted to carbon aerogels which may retain the main morphological characteristics (fibrous versus particulate) of their parent polyimide aerogels.
Three-dimensional porous polyimide networks such as polyimide aerogels synthesized through the isocyanate route are less prone to shrinkage during processing compared to polyimide aerogels synthesized through the diamine route, densities as low as 0.04 g/cm3 have been obtained. Further, owing to solubility issues, PI-AMN aerogels generally cannot be made with sols above 20% w/w solids in the sol. Resulting PI-AMN aerogels had a density of 0.291 g/cm3. However, more concentrated sols are possible via the isocyanate route. For example, PI-ISO aerogels from sols with 30% w/w solids had a density of about 0.304 g/cm3. Upon carbonization at 800° C., PI-ISO retain their parent nanomorphology, which may be fibrous. The ability to prepare higher density polyimide aerogels and other three-dimensional porous polyimide networks extends the range of compressive strength and stiffness attainable for polyimide networks and their carbonized derivatives.
Three-dimensional porous polyimide networks such as polyimide aerogels prepared in accordance with methods described herein may exhibit a suitable degree of shrinkage. In some embodiments, three-dimensional porous polyimide networks (e.g., polyimide aerogels) prepared with isocyanate may exhibit a shrinkage of less than 30%, less than 20%, less than 10%, less than 5%, less than 2%, less than 1%. In some cases, the shrinkage of a polyimide aerogel may be affected by the percent solids w/w. For example, when the percentage of solids in the sol is lower, the degree of shrinkage for the polyimide aerogel may be greater.
The density of three-dimensional porous polyimide networks or three-dimensional porous carbon networks prepared from pyrolysis of polyimide networks using methods described herein may be between, for example, about 1 mg/cc and about 550 mg/cc, or between about 100 mg/cc and about 400 mg/cc. In some embodiments, the density of three-dimensional porous polyimide networks or three-dimensional porous carbon networks prepared from pyrolysis of polyimide networks may be greater than about 300 mg/cc.
Aerogels described herein may include a three-dimensional porous network of secondary particles and primary particles. Porous secondary particles having a diameter of between about 5-10 nm are made up of nonporous primary particles having a diameter of less than about 1 nm. While the primary particles are arranged in a manner that forms micropores within the secondary particles, larger mesopores arise through the arrangement of secondary particles relative to one another. A silica nanoparticle network in the form of an aerogel may include mesopores having voids that are between 2-50 nm in diameter between secondary particles. Primary particles include voids that are less than 2 nm in diameter.
Three-dimensional porous networks discussed herein may have fibrous morphologies where the networks may include nanofibers having various diameters and lengths. In some embodiments, fibrous morphologies of such networks include fibers having an average diameter ranging between about 1 nm and about 500 nm (e.g., between about 10 nm and about 400 nm, between about 100 nm and about 300 nm) or less than 500 nm (e.g., less than 400 nm, less than 300 nm, less than 200 nm). In some embodiments, fibrous morphologies of porous networks described herein include fibers having an average length of at least 50 nm and may extend into the micron length scale. The average aspect ratio of fibers in a three-dimensional porous networks having a fibrous morphology, such as an aerogel may be greater than 2, greater than 5, greater than 10, greater than 20, greater than 50, etc. Three-dimensional porous networks described herein that exhibit a fibrous morphology may include both fibrous and non-fibrous features (e.g., particulate), but, in some embodiments, could also contain only fibrous features, free of non-fibrous features.
Any appropriate isocyanate, such as a diisocyanate or triisocyanate, may be used for forming a sol-gel material to produce a three-dimensional porous polyimide network, such as a polyimide aerogel. For example, suitable non-limiting diisocyanates may include 4,4′-diisocyanatodiphenylmethane or methylene di-p-phenyldiisocyanate (MDI).
Any appropriate dianhydride may be used to react with an isocyanate (e.g., diisocyanate, triisocyanate) in forming a sol-gel material to produce a three-dimensional porous polyimide network, for example, a polyimide aerogel. In an embodiment, a suitable non-limiting dianhydride includes pyromellitic dianhydride (PMDA).
A suitable solvent may be used along with the isocyanate and the dianhydride. In various embodiments, a pyrrolidone (e.g., N-methyl-2-pyrrolidone, NMP), acetonitrile, acrylonitrile, acetone, and/or combinations thereof may be used to form a three-dimensional porous polyimide network, such as a polyimide aerogel.
In some embodiments, monolithic polyimide aerogels (PI-ISO) may be prepared by drying wet gels synthesized via a rather underutilized room-temperature reaction of pyromellitic dianhydride (PMDA) with 4,4′-methylene di-p-phenyldiisocyanate (MDI). The reaction may be observed using liquid-phase 13C-NMR in DMSO-d6 which, for some cases, has indicated that the reaction proceeds through a seven-member ring intermediate that collapses to the imide by expelling CO2. PI-ISO may be characterized and compared with aerogels obtained via the classic reaction of PMDA and 4,4′-methylenedianiline (MDA), referred to as PI-AMN. The two materials, PI-ISO and PI-AMN may be chemically identical, showing similar degrees of crystallinity (30-45%, by XRD), and both including similarly sized primary particles (6.1-7.5 nm, verified by SANS). For some embodiments, they both may contain meso- and macroporosity by N2-sorption porosimetry and may exhibit similar BET surface areas (300-400 m2 g−1).
A major difference between PI-AMN and PI-ISO prepared using methods described herein, however, is that PI-AMN are particulate while PI-ISO may be fibrous in morphology. Without being limited to any one particular mechanism, the difference in morphology may be due to the rigidity of the seven-member ring intermediate of PI-ISO. PI-AMN shrinks significantly during processing (up to 40% in linear dimensions), but can exhibit stronger mechanical properties than PI-ISO of the same density. Upon pyrolysis at 800° C., both PI-ISO and PI-AMN have been observed to convert to porous carbons. PI-AMN lose their nanomorphology and more than ⅔ of their surface area, as opposed to PI-ISO, which retains their nanomorphology and surface area. Etching with CO2 at 1000° C. increases the surface area of both PI-AMN and PI-ISO, and improves the electrical conductivity of the latter. In one embodiment, etching resulted in an increase of surface area in PI-AMN to 417 m2/g and an increase in PI-ISO to 1010 m2/g and an increase in electrical conductivity in PI-ISO by a factor of 70. Etching for longer periods of times, with higher flow rates of CO2, and higher temperatures can further increase these values.
Among engineering plastics, polyimides demonstrate good chemical resistance as well as excellent mechanical properties and stability at high temperatures. For example, a special formulation referred to as PMR-15 is emerging as an aerospace industry standard for replacing metal components in jet engines and is rated at 290° C. for 10,000 h. In that regard, polyimide aerogels could be ideal materials for high-temperature low-k dielectrics for fast electronics (where k stands for the dielectric constant), and for high-temperature thermal insulation with advantages over blown closed-cell macroporous polyimide foams already in use for that purpose. As mentioned, polyimide aerogels may have synthesized employing a conventional two-step DuPont route utilizing dianhydrides and diamines. The two monomers react at room temperature yielding a polyamic acid solution that subsequently is dehydrated to the polyimide with acidic anhydride and a base-catalyst (e.g., pyridine or triethylamine). A post-gelation high-temperature treatment of wet gels ensures complete imidization and conversion of undesirable isoimides to imides. Though chemical dehydration of the polyamic acid is energy-intensive and introduces by-products, industrially, dehydration and imidization are carried out by direct heating of the polyamic acid at high temperatures (˜200° C.). In an attempt to adapt this direct heating method for the synthesis of aerogels, however, precipitates were obtained rather than gels.
Polyimide aerogels described herein may also be synthesized via a one-step room-temperature route that involves reaction of anhydrides with isocyanates. In some embodiments, pyromellitic dianhydride (PMDA) is reacted with 4,4′-methylene diphenyl diisocyanate (MDI). For contrast, a route involving the reaction of PDMA with 4,4′-methylenedianiline (MDA) will be considered. Resulting aerogels from both routes are chemically identical (by IR and solids CPMAS 13C-NMR). However, despite similar degrees of crystallinity and primary particle sizes (by XRD and SANS, respectively), their nanomorphology (by SEM) may be quite different: polyimides from the amine route (PI-AMN) are particulate, while those from the isocyanate route (PI-ISO) are fibrous. One of ordinary skill in the art would be able to recognize the difference between a particulate and fibrous morphology. Upon pyrolysis, both kinds of polyimide aerogels may be converted to carbon aerogels, which, in the case of PI-ISO retain the fibrous morphology of their parent polyimides.EXAMPLES Synthesis of PI-ISO Versus PI-AMN
PI-AMN were prepared through a polyamic acid step (see
Room-temperature PI-AMN gels were solvent-exchanged with ethanol and dried to aerogels (abbreviated as PI-AMN-RT) using liquid CO2 extracted supercritically. Subcritical drying of the aerogels may also be performed. According to IR, PI-AMN-RT aerogels do not contain detectable amounts of isoimide (with expected absorption at 1800 cm−1), but the absorption at ˜3500 cm−1 indicates that they may contain amides and therefore unreacted polyamic acid. Heating PI-AMN-RT aerogels at 190° C. causes extensive shrinkage and results into xerogel-like materials with a loss in surface area. Hence, PI-AMN was heated in a solvent-swollen state, and thus imidization was completed before drying by transferring wet gels into NMP followed by heating at 190° C. Subsequently, wet gels may be cooled to RT, NMP was exchanged with ethanol, and ethanol-filled wet-gels may be supercritically dried with CO2 yielding PI-AMN-190 aerogels. In some cases, even though absorption at ˜3500 cm−1 decreases, it did not disappear. Further, even by heating in the solvent-swollen state, PI-AMN samples shrink significantly relative to their molds (up to 40% in linear dimensions) irrespective of their bulk density; shrinkage is not caused by the heat treatment at 190° C., as aerogel samples either dried directly after gelation (PI-AMN-RT) or after heating at 190° C. in NMP according to
On the other hand, PMDA/MDI solutions gel directly at room temperature without use of additional reagents. The reaction was followed up to the gelation point using liquid-phase 13C-NMR in DMSO-d6. As shown in
In some embodiments, the seven-member intermediate may be considered an in-situ foaming agent leading to foam-like PI-ISO architectures defined by porous walls. Macroscopically void-free (i.e., bubble-free) PI-ISO aerogel monoliths were obtained by two methods (shown in
IR spectra of PI-ISO-RT look essentially identical to IR spectra of PI-AMN-190, but the solid-state 13C-NMR spectra of the two materials show differences in the relative peak intensities which are attributed to differences in polymer length and therefore to the polarity of the sol (as a reminder, PI-ISO-RT were prepared in NMP/acetonitrile or NMP/DMSO mixtures, while PI-AMN-190 were prepared in pure NMP).
Pyromellitic dianhydride (PMDA) (e.g., obtained from Chriskev Company, Inc.); 4,4′-diisocyanatodiphenylmethane (4,4′-methylene diphenyl diisocyanate, MDI), 4,4′-methylenedianiline (MDA), and acetic anhydride ((Ac)2O) (e.g., obtained from Acros Chemicals); and pyridine (e.g., obtained from Fisher Scientific) may be used. Anhydrous N-methyl-2-pyrrolidone (NMP), acetonitrile, dimethylsulfoxide (DMSO), and absolute ethanol were used (e.g., obtained from Aldrich Chemical Co). Deuterated DMSO (DMSO-d6) was also used (e.g., obtained from Cambridge Isotope Laboratories, Inc).
Preparation of polyimide aerogels via the anhydride/diamine route (PI-AMN) will now be described.
Pyromellitic dianhydride (15.696 g, 0.072 mol) and 4,4′-methylenedianiline (14.256 g, 0.072 mol) were added in variable amounts of NMP in a 250 mL round bottom flask under magnetic stirring. For example, for the 15% w/w solids formulation (samples referred to as PI-AMN-15) the amount of NMP was 169.7 g. The solution was stirred at room temperature under nitrogen for 12 h and the color changed from colorless to yellow. At that point, acetic anhydride (in 6.3 mol excess to pyromellitic dianhydride) and pyridine (1:1 mol ratio to acetic anhydride) were added to the resulting polyamic acid solution, and the new solution (sol) was poured into molds and was allowed to gel at room temperature. As molds for samples used for general chemical and physical characterization, Wheaton 4-mL Polypropylene Omni-Vials with a 1.04-cm inner diameter were used (Fisher part No. 225402), and for samples used for mechanical testing 30-mL Fisherbrand Class B Amber Glass Treaded Vials were used, 2.1-cm inner diameter (Fisher part No. 03-339-23E). The gelation time depends on the formulation: for 2.5%, 5%, 10% and 20% w/w solids (PMDA+MDA) in the original NMP solution the gelation time was ˜4 h, ˜1.5 h, 15 min and <15 min, respectively. The gels were aged in the molds for 12 h, subsequently they were washed with NMP (3×, 8 h each time, using 4× the volume of the gel each time), cured at 190° C. for 3 h in the last NMP wash solution, cooled to room temperature, washed with ethanol (4×, 8 h each time, using 4× the volume of the gel for each wash) and dried into polyimide aerogels in an autoclave with liquid CO2, taken out at the end supercritically.
Preparation of polyimide aerogels via the isocyanate route (PI-ISO) at room temperature and at 90° C. will now be described.
At room temperature, MDI (2.50 g, 0.01 mol) was added under stirring to a mixture of NMP and acetonitrile (3:1 w/w), or NMP and DMSO (1:1 w/w) in a three-neck round bottom flask under N2 at room temperature. Once the isocyanate was dissolved, pyromellitic dianhydride (2.18 g, 0.01 mol) was added. For example, for a 15% w/w solids formulation the amount of NMP and acetonitrile were 19.88 g (19.34 mL) and 6.62 g (8.43 mL), respectively. Similarly for a 10% w/w solids formulation, the amounts of NMP and DMSO were 21.06 g and 21.06 g respectively. The reaction mixture was stirred under nitrogen for 1 h. The resulting sol was poured in molds (see above), which were allowed to stand at room temperature. Gelation takes usually 6 h-48 h depending on the formulation. For example, in NMP:CH3CN (3:1 w/w), the 10%, 15%, 20% and 30% w/w solids formulations take 2 days, 36 h, 18 h, and 12 h, respectively. By comparison, a 15% w/w solids formulation in NMP takes 48 h for gelation at RT. Gels were aged in the molds for 24 h-8 days depending on the gelation time, typically four times the gelation time. Subsequently, gels were removed from the molds and were initially washed (solvent-exchanged) with NMP:acetonitrile (3:1 v/v), then with NMP:acetonitrile:acetone (2:1:1 v/v/v), NMP:acetone (1:1 v/v), and finally with pure acetone. Similarly, gels synthesized in NMP/DMSO were placed in pure NMP, the second wash was carried out with NMP:acetone (3:1 v/v) the third with NMP:acetone (1:1 v/v), and finally gels were placed in pure acetone. After 4 acetone washes (8 h each time, using 4× the volume of the gel for each wash) wet gels were dried into polyimide aerogels in an autoclave with liquid CO2, taken out at the end supercritically.
At 90° C., a more time-efficient synthesis of PI-ISO was carried out at slightly (by polyimide standards) elevated temperatures as follows: pyromellitic dianydride (2.18 g, 0.01 mol) and MDI (2.50 g, 0.01 mol) were added in variable amounts of NMP in a three-neck round bottom flask under nitrogen at room temperature. For example, for a 15% w/w solids formulation the amount of NMP was 26.52 g. The flask was placed in a 60° C. bath and the solution was stirred under N2 for 0.5 h. The resulting sol was poured into molds (see above), which were heated in an oven successively at 60° C., 70° C. and 80° C. for 3 h at each temperature. Although gelation of the PMDA/MDI/NMP mixture may occur even at room temperature over longer periods of time as described above, according to this protocol, gelation usually occurs at the early stages of the 60° C. heating. Gels were first aged at 90° C. for 12 h in their molds, and subsequently they were removed from the molds and they were placed directly in fresh ethanol. After four ethanol washes (8 h each time, using 4× the volume of the gel for each wash) wet gels were dried into polyimide aerogels in an autoclave with liquid CO2, taken out at the end supercritically.
Conversion of polyimide aerogels into carbon aerogels will also be described.
PI-AMN and PI-ISO aerogels were pyrolyzed at 800° C. for 3 h in a tube furnace under a flowing stream of Ar. Before heating, the tube was purged with Ar for 10 min, and the heating rate was set at 5° C. min−1. At the end of the heating period, the power to the furnace was disconnected and the tube was allowed to cool slowly back to room temperature under flowing Ar.
Carbon aerogels were etched by placing the aerogels in a tube furnace under flowing argon and were heated at 1000° C. The flowing gas was switched to CO2 and the temperature was maintained at that level for 3 h. Subsequently the flowing gas was switched back to Ar and the power to the furnace was disconnected, allowing for slow cooling back to room temperature.
Drying with supercritical fluid CO2 was conducted in an autoclave (SPI-DRY Jumbo Supercritical Point Dryer, SPI Supplies, Inc. West Chester, Pa.). Aerogel bulk densities (ρb) were calculated from the weight and physical dimension of the samples. Skeletal densities (ρs) were determined with helium pycnometry using a Micromeritics AccuPyc II 1340 instrument. Porosities, Π, were determined from ρb and ρs according to: P=100×[(1/ρb)−(1/ρs)]/(1/ρb). BET surface areas (σ) and pore size distributions were measured with nitrogen adsorption/desorption porosimetry using a Micromeritics ASAP 2020 Surface Area and Porosity Analyzer. Samples for surface area and skeletal density determination were outgassed for 24 h at 80° C. under vacuum before analysis. Average pore diameters were determined by the 4×VTotal/σ method, where VTotal is the total pore volume per gram of sample. VTotal is calculated either from the single highest volume of N2 adsorbed along the adsorption isotherm or from the relationship VTotal=(1/ρb)−(1/ρs). The single point N2 adsorption method tends to underestimate VTotal significantly when macropores are involved, and thus numerical proximity of the values determined by the two methods is used as a semi-quantitative criterion for evaluating macroporosity. Scanning electron microscopy (SEM) was conducted using a Hitachi S-4700 Field Emission instrument. Chemical characterization of all polyimide aerogels was based on IR and solid state 13C NMR. Infrared spectra were obtained in KBr pellets using a Nicolet-FTIR model 750 Spectrometer. Solid-state 13C NMR spectra were obtained with samples ground in fine powders on a Bruker Avance 300 spectrometer (75.475 MHz carbon frequency), using magic angle spinning (5 kHz) with broadband proton suppression and the CPMAS TOSS pulse sequence for spin sideband suppression. 13C NMR spectra were externally referenced to the carbonyl of glycine (176.03 ppm relative to tetramethylsilane). 13C-NMR peak assignment was aided by NMR spectra simulations using ChemDraw Ultra 7.0. Thermogravimetric analysis (TGA) was conducted under nitrogen with a TA Instrument, model Hi-Res-TGA 2950 using ˜10 mg samples and a heating rate of 10° C. min−1. The reaction between PMDA and MDI was monitored in DMSO-d6 at room temperature up to the gelation point by liquid 13C NMR using a 400 MHz Varian Unity Inova NMR instrument. The crystallinity of the polyimide samples was determined by x-ray diffraction (XRD) using a Scintag 2000 diffractometer with Cu Kα radiation and a proportional counter detector equipped with a flat graphite monochromator. The identity of the fundamental building blocks of the two materials was probed with small angle neutron scattering (SANS) using ˜2 mm thick discs cut with a diamond saw from cylinders, on a time of flight, low-Q diffractometer, LQD, at the Manuel Lujan Jr. Scattering Center of the Los Alamos National Laboratory. The scattering data are reported in the absolute units of differential cross section per unit volume (cm−1) as a function of Q, the momentum transferred during a scattering event. Quasi-static mechanical testing under compression was conducted on an Instron 4469 universal testing machine frame, following the testing procedures and specimen length/diameter ratio (equal to one) in ASTM D1621-04a (Standard Test Method for Compressive Properties of Rigid Cellular Plastics). The recorded force as a function of displacement (machine-compliance corrected) was converted into stress as a function of strain. Four-point-probe conductivity measurements were conducted on flat surface of rectangular block of carbon aerogel samples (made with a fine sand paper) using an Alesis contact probe station model CPS-06 with a Cascade Microtech electrode model C4S-44/5S. The reliability of the probe was confirmed with silicon wafers and indium-tin-oxide coated glass slides of known sheet resistance.Macroscopic Characterization of PI-ISO Versus PI-AMN
General materials properties of polyimide aerogels synthesized by methods shown in
Partly owing to the similar chemical composition of the two materials, partly to their similar degree of crystallinity (30-45% by XRD, see Table 1), and partly to similar packing distances within the crystalline phases (expressed by similar 2θ values, see Table 1), skeletal densities, ρs, of all samples are in the 1.4-1.5 g cm−3 range, which is comparable with the density of bulk polyimides obtained from PMDA and MDA (1.357 g cm−3). The similar ρs values but the different ρb's are reflected in the porosities, which are higher for the PI-ISO aerogels relative to the corresponding PI-AMN samples. However, BET surface areas, σ, from N2-sorption data show an opposite trend from the porosities: despite higher shrinkage, higher bulk densities and lower porosities, PI-AMN samples have about equal or higher surface areas than PI-ISO samples. These data point to significant differences in the nanomorphology of the PI-AMN versus the PI-ISO samples, which are discussed in the next section.
The microstructure of polyimide aerogels was evaluated in terms of their pore-size distribution and the nanomorphology of their skeletal frameworks. The pore-size distribution at the meso/macro scale was evaluated semi-quantitatively by analysis of the N2-sorption data in combination with SEM, while the elementary building blocks of the skeletal framework were probed with SANS. Data shown concern samples obtained with the 15% w/w solids formulation. Similar data and trends have been observed with samples prepared with all other solids formulations (Table 1).
For both PI-AMN and PI-ISO samples, N2-sorption isotherms show a rapid increase of the volume adsorbed at relative pressures above 0.9, which in combination with the narrow desorption loop indicates the presence of both meso- and macroporosity. Indeed, pore size analysis via the relationship (pore diameter)=4×VTotal/σ, where VTotal is calculated either from the maximum adsorption point in the isotherm, or the relationship VTotal=(1/ρb)−(1/ρs), gives quite different values (Table 1), which get progressively closer as the bulk density increases, as expected by the fact that more dense materials should have smaller pores. The BJH-desorption method reflects the mesoporosity and yields pore diameters closer to those obtained by the single point absorption method (Table 1), but it also shows quite broad pore size distributions (evaluated by the width at half maxima of the BJH plots, see Table 1). Overall, the N2-sorption method indicates that both PI-AMN and PI-ISO aerogels are meso/macroporous materials. SEM, however, shows that their pore structures are quite different: at all densities PI-AMN are particulate while PI-ISO are fibrous. At high magnifications, there was a discernible primary/secondary particle structural hierarchy in the case of PI-AMN, while in some cases it can be also claimed that the PI-ISO ribbons consist of particles.
The make-up of the skeletal frameworks in PI-AMN and PI-ISO was probed quantitatively with SANS. To exclude the effect of the solvent, which controls phase-separation, and therefore affects the size of the particles, the specific materials compared were both prepared in NMP (PI-ISO-90 and PI-AMN-190). While scattering from PI-AMN and PI-ISO is distinct, there are similarities. To facilitate analysis, and as shown in
Overall, chemically identical (case of PI-AMN-190 and PI-ISO-90) and structurally very similar primary particles (in terms of crystallinity and size) seem to form secondary particles in the case of PI-AMN, and fibers in the case of PI-ISO. The size of the primary particles may be controlled by the common solvent (NMP). Then, the variable that remains different in the two systems is the actual chemistry of the two processes, which is translated into the surface functionality of the primary particles. It is suggested that the flexible amic acid bridges between PI-AMN primary particles allows pivoting, closer packing, and a 3D growth resulting into secondary particles; on the other hand, the rigidity of the seven-member ring between PI-ISO primary particles, in combination with steric hindrance from neighboring particles, imposes growth at the exposed ends of the assembly resulting in directional growth and fibers.
The 3D growth in PI-AMN create numerous crosslinks between secondary particles while in the case of PI-ISO, crosslinking may take place only at the contacts between fibers. This model for PI-AMN versus PI-ISO aerogels is supported by the higher shrinkage of PI-AMN, and may serve as an explanation for their high compressive mechanical strength, which compares favorably with that of structurally analogous polymer crosslinked silica aerogels at the same densities. On the other hand, at similar percent solids formulations, PI-ISO aerogel monoliths are mechanically much weaker materials, undergoing premature catastrophic failure to large pieces with much smaller loads. In some embodiments, entangled nanofibrous structures are generally mechanically stronger (the bird-nest effect).
Pyrolysis of Polyimide Aerogels and Conversion to Carbon
Porous carbons are pursued as electrodes for fuel cells and batteries. Polyimides generally have good carbonization yields, and in fact the first PI-AMN aerogels reported were also investigated for their conversion to carbon aerogels (pyrolytically) and metal carbide aerogels (carbothermally, after doping with selected sol-gel derived metal oxides). Carbonizable polymers are capable of either cyclizing, or undergoing ring fusion and chain coalescence by heating. For this the chain should either contain aromatic moieties or be aromatizable (usually by oxidation). In the former case, there is one carbon atom between aromatic rings; otherwise, pyrolytic chain scission will prevail leading to loss of fragments. The PMDA/MDI or MDA polyimides herein fulfill the last criterion. By thermogravimetric analysis under N2 (TGA) both PI-AMN and PI-ISO are stable up to about 550° C., subsequently loosing 40-50% of their mass before 650° C. (presumably by loss of small molecules like CO and CO2).
Polyimide aerogels remain as quite sturdy monoliths after pyrolysis. Carbon aerogels are also black in color. PI-AMN-derived carbons, however, have a metallic luster, while PI-ISO derived samples are dull. The carbonization yield generally is between 51-56% w/w. Carbon aerogels shrink further beyond their initial imidization shrinkage (Table 1), but PI-ISO-RT samples shrink more (˜60%) than the PI-AMN samples (˜50%), so that the total shrinkage calculated from the initial molds (Table 2) is approximately the same for both kinds of samples (63-66%). The PI-ISO-90 samples seem to have a small advantage over the rest in terms of shrinkage (overall 52% relative to the molds).
Chemically, pyrolytic samples consist only of carbon (as determined by EDS). XRD shows very broad diffractions. Typical Raman spectra show both the G (graphitic) and D (disordered) peaks at 1352 cm−1 and 1597 cm−1, respectively.
Microscopically (by SEM) PI-AMN-derived carbons are different from their parent polyimide aerogels.
The presence of micropores indicated by the N2-sorption isotherms suggests that a significant gain in surface area could be achieved by etching. That was carried out under flowing CO2 at 1000° C. (carbon and CO2 comproportionate to CO.) The results are included in Table 2. CO2-treated samples lose 20-40% of their mass, shrink 3-9% and remain monolithic. Consequently, bulk densities decrease, however, skeletal densities increase to the 2.1-2.3 g cm−3 range (density of graphite at 2.26 g cm−3). The porosity is dominated by the bulk density decrease, reaching up to ˜70% v/v of empty space. Microscopically, PI-ISO samples remain fibrous, while PI-AMN samples show macropores similar in shape to those observed before etching. The number of those macropores has increased and surrounding walls seem rougher; considering these data together suggests that before etching many pores are masked by a thin porous crust of carbon. After CO2-etching, the N2-sorption isotherms indicate that the majority of the empty space is attributed to micropores, while the BET surface areas of the samples increases dramatically, reaching the levels of the parent polyimide aerogels in the case of PI-AMN-derived carbons (417 m2 g−1), or far surpassing those levels in the case of PI-ISO-derived samples (1010 m2 g−1). For reasons not well understood yet, despite the mass loss CO2-etching increases the electrical conductivity of PI-ISO-derived carbon aerogels by ˜70×, from 0.013 mho cm−1 (at ρb=0.967 g cm−3) to 8.697 mho cm−1 (at ρb=0.670 g cm−3). (By comparison the electrical conductivity of CO2-etched PI-AMN-derived carbon aerogels is 4.491 mho cm−1 at ρb=0.701 g cm−3.) Those values render polyimide-derived carbon aerogels particularly attractive as electrochemical electrodes.
Three-dimensional porous polyimide and carbon networks derived therefrom, such as polyimide and carbon aerogels, are high value-added materials and may be useful for a number of applications. For example, such porous polyimide networks may be useful in high-temperature thermal insulation (e.g., architectural, automotive industrial applications, aircraft, spacecraft, clothing), lightweight structures, acoustic insulation (e.g., buildings, automobiles, aircrafts), impact dampening, dielectrics (e.g., for fast electronics), supports for catalysts, and as hosts of functional reactants for chemical, electronic and optical applications. In some embodiments, three-dimensional porous networks described may be used as insulation material for apparel, such as clothing, astronaut suits, and armor.
Described herein are three-dimensional porous polyimide networks synthesized via a low temperature process through a reaction of dianhydrides with diisocyanates. Although MDI has been used, other monomers such as 4,4′-oxybis(phenylisocyanate) may behave similarly and resulting polyimides may have the chemical composition of Kapton. Three-dimensional porous polyimide networks (e.g., polyimide aerogels) derived from PMDA/MDI may exhibit fibrous morphologies and can be converted pyrolytically to isomorphic three-dimensional porous carbon networks (e.g., carbon aerogels), which, owing to their large surface area, may be useful for their reactivity with nanoparticulate oxidizing agents.
Three-dimensional porous polyimide networks including polyimide aerogels may be useful in applications that involve, for example, absorption of oil or other hydrophobic materials. In some instances, such materials may be capable of absorbing 5, 15, 20, 25, or more times their weight in oil or other hydrophobic material. In some instances, the majority of the substance absorbed may be retrieved by chemical and/or mechanical methods. Three-dimensional porous carbon networks derived from three-dimensional porous polyimide networks may also be useful for a number of applications including, for example, manufacture of electrodes, batteries, supercapacitors, high-temperature insulators, high-temperature ballistics materials, ablative materials and infrared blocking armor. In some embodiments, doping PI-ISO with cerium oxide and pyrolysis to carbon under argon may lead to materials that are self ignited when exposed to air, giving rise to potential energetic materials applications.
Having thus described several aspects of at least one embodiment of this invention, it is to be appreciated various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modification, and improvements are intended to be part of this disclosure, and are intended to be within the spirit and scope of the invention. Accordingly, the foregoing description and drawings are by way of example only.
1. A method of manufacturing a three-dimensional nanostructured polyimide network, the method comprising:
- mixing an anhydride and an isocyanate in a solvent to form a sol-gel material; and
- drying the sol-gel material to form the polyimide network.
2. The method of claim 1, wherein the solvent comprises at least one of pyrrolidone and N-methyl-2-pyrrolidone.
3. The method of claim 1, wherein the anhydride is pyromellitic dianhydride.
4. The method of claim 1, wherein the isocyanate is 4,4′-diisocyanatodiphenylmethane.
5. The method of claim 1, wherein the solvent further comprises at least one of acetonitrile, acrylonitrile, or acetone.
6. The method of claim 1, further comprising a step of subjecting the sol-gel material to solvent-exchange using at least one of a pyrrolidone, acetonitrile, or acetone.
7. The method of claim 1, wherein the drying the sol-gel material includes supercritical drying of the sol-gel material.
8. The method of claim 1, wherein the drying the sol-gel material includes subcritical drying of the sol-gel material.
9. A fibrous aerogel comprising:
- a three dimensional network of particles including polyimide, the three dimensional network having a fibrous morphology.
10. An insulator, a lightweight structural material, an acoustic insulating material, or an impact dampening material comprising the aerogel of claim 9.
11. The method of claim 1, further comprising pyrolyzing the polyimide network to form a carbon network.
12. The method of claim 11, wherein the carbon network is chemically etched to produce a carbon network having increased microporosity.
13. The carbon aerogel of claim 11.
14. A fibrous aerogel comprising:
- a three dimensional network of nanoparticles including a carbon skeleton, the three dimensional network having a fibrous morphology.
15. An electrode, a battery, a supercapacitor device, an insulator, a ballistic material, an ablative material, an armor comprising the aerogel of claim 14.
16. A method of manufacturing an aerogel, the method comprising:
- mixing an anhydride and an isocyanate in a solvent in the presence of a ceramic oxide to form a sol-gel material;
- drying the sol-gel material to form the hybrid polyimide aerogel; and
- pyrolyzing the hybrid polyimide aerogel to form at least one of a metal aerogel, a metalloid aerogel, a metal carbide aerogel and a metalloid carbide aerogel.
17. The method of claim 16, in which the ceramic oxide forms an interpenetrating network throughout a polyimide network.
18. The method of claim 16, in which the ceramic oxide comprises one or more of a metal oxide, a metalloid oxide, or silicon oxide.
19. The aerogel of claim 16.
20. The method of claim 1, wherein the isocyanate comprises at least one of a diisocyanate, triisocyanate, and a polyisocyanate.
Filed: Aug 22, 2011
Publication Date: May 31, 2012
Applicant: Aerogel Technologies, LLC (Glendale, WI)
Inventors: Nicholas Leventis (Rolla, MO), Chariklia Sotiriou-Leventis (Rolla, MO), Chakkaravarthy Chidambareswarapattar (Rolla, MO)
Application Number: 13/214,633
International Classification: C08G 73/10 (20060101); C01B 31/02 (20060101); C01B 31/00 (20060101); B44C 1/22 (20060101); C08J 9/28 (20060101); C01B 31/30 (20060101); B82Y 40/00 (20110101); B82Y 30/00 (20110101);