TITLE OF THE INVENTION

COMPOSITIONS OF CONDUCTIVE POLYMERS AND METHODS FOR

MAKING AND USING SAME

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority from U.S. Provisional Application Serial No. 62/898,195 filed on 10 September 2019, U.S. Provisional Application Serial No. 62/939,265 filed on 22 November 2019, and U.S. Provisional Application Serial No. 63/000,870 filed on 27 March 2020, each of which are incorporated herein by reference in their entireties.

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

Not applicable.

MATERIAL INCORPORATED-BY-REFERENCE Not applicable.

FIELD OF THE INVENTION

The present disclosure generally relates to methods and compositions for making nanostructured conducting polymer particles and layers (e.g., films, mats) using vapor polymerization methods, as well as electric components and devices comprising the nanostructured conducting polymer particles and layers.

SUMMARY OF THE INVENTION

Among the various aspects of the present disclosure are provisions for compositions and methods for rust-based formation of nanostructured polymers, polymer/oxide composites, or polymer/carbon composites.

An aspect of the present disclosure provides for a A method for forming a nanostructured poly (3,4-ethylenedioxythiophene) (PEDOT) layer on a solid substrate, the method comprising: providing the substrate, at least a portion of the substrate comprising a-Fe2O3; and contacting the substrate with a vapor mixture comprising HCI and 3,4-ethylenedioxythiophene (EDOT) monomer at a temperature of at least 140 °C to form the nanostructured PEDOT layer, the nanostructured PEDOT layer comprising a plurality of nanofibers, each nanofiber comprising a PEDOT outer layer and an underlying ferrous core..

Another aspect of the present disclosure provides for a method of forming a 3D nanofibrillar PEDOT micro-supercapacitor (m-SC), the method comprising: providing an electrically insulated substrate; applying a photoresist layer to the substrate; etching the photoresist layer to form a patterned photoresist layer comprising an interdigitated electrode pattern; applying a current collector layer over the patterned photoresist layer and exposed substrate; applying an Fe2O3 layer over the patterned photoresist layer and exposed substrate; contacting the Fe2O3 layer with a vapor mixture comprising HCI and 3,4-thylenedioxythiophene

(EDOT) monomer at a temperature of at least 140 °C to form a nanostructured PEDOT layer, the nanostructured PEDOT layer comprising a plurality of nanofibers, each nanofiber comprising a PEDOT outer layer and an underlying ferrous core; and removing the patterned photoresist layer from the substrate to form the m-SC, the m-SC comprising a pair of interdigitated PEDOT electrodes, each electrode comprising a portion of the current collector layer and the nanostructured PEDOT layer attached to the substrate, the nanostructured PEDOT layer comprising a plurality of nanofibers, each nanofiber comprising a PEDOT outer layer and an underlying ferrous core.

Yet another aspect of the present disclosure provides a polymer particle comprising a conducting polymer, wherein the polymer particle has an average diameter between about 1 nm and about 100 mm and/or a conductivity of at least about 300 S/cm.

Yet another aspect of the present disclosure provides for a method of aerosol vapor polymerization (AVP) for producing conducting polymer particles comprising: providing a monomer vapor (e.g., EDOT); providing an oxidant- carrying aqueous droplet (e.g., FeCI3 via an ultrasonic nebulizer); and contacting the monomer vapor and the oxidant-carrying aqueous droplet, resulting in a spherical or roughly spherical polymer particle comprising Cl- doped conjugated polymer chains.

Yet another aspect of the present disclosure provides for an aerosol vapor polymerization (AVP) system comprising: an ultrasonic nebulizer and a reaction bubbler operably connected to a reaction chamber; and a collection chamber. Yet another aspect of the present disclosure provides for a polymer layer comprising: conducting polymer nanofibers having an average diameter between about 75 nm and about 100 nm; an average aspect ratio greater than about 900; and/or a conductivity of at least about 300 S/cm.

Yet another aspect of the present disclosure provides for a method of vapor phase polymerization for producing a polymer layer, comprising: (i) providing a substrate selected from at least one of an iron(lll)-coated substrate, an iron(lll) containing substrate, a rust-containing substrate, or a rust-generating substrate; (ii) contacting a vapor phase Fe3+ liberating agent to the substrate; and (iii) contacting a vapor phase monomer to the substrate.

Yet another aspect of the present disclosure provides for a reaction system comprising a chamber, the chamber comprising: a first reservoir comprising a Fe3+ liberating agent; a second reservoir comprising a monomer in a volatile organic compound; and a substrate comprising rust.

Yet another aspect of the present disclosure provides for a supercapacitor or micro-supercapacitor device comprising a layer or a layer manufactured according to any one of the preceding claims.

Other objects and features will be in part apparent and in part pointed out hereinafter.

DESCRIPTION OF THE DRAWINGS

Those of skill in the art will understand that the drawings, described below, are for illustrative purposes only. The drawings are not intended to limit the scope of the present teachings in any way.

FIG. 1A-FIG. 1 H. Optical and electron microscopy analysis of a nanofibrillar PEDOT film produced from rusted steel. (A) The digital photograph shows a steel sheet coated by a yellow colored rust layer characterized by (B) a texturized porous architecture. (C) The scanning electron micrograph shows that the rust layer possesses a sea-urchin-like micro structure. When this rusted substrate is utilized for vapor-phase polymerization, (D) a blue PEDOT coating is deposited composed of (D) a nanofibrillar architecture possessing (F) a high packing density of high-aspect-ratio nanofibers. (G) Schematic illustration of rust- based vapor-phase polymerization (RVPP) carried out in a sealed vessel. (H) Scheme of the RVPP mechanism entailing dissolution of rust, liberation of Fe ³⁺ ions, and oxidative radical polymerization of EDOT monomer resulting in a doped form of PEDOT.

FIG. 2A-FIG. 2H. Scanning electron micrographs and schematic illustration of rust-based vapor-phase polymerization mechanism. (A) Rust microstructures collapse as HCI vapor dissolves a rust layer promoting the (B) liberation of Fe ³⁺ ions from bulk rust (a/g-FeOOH, FeSO ₄ ). (C) A PEDOT skin forms on the surface of rust as Fe ³⁺ ions contact EDOT vapor. During this process, water content and Fe ³⁺ concentration in the rust bulk increase leading to (D) hydrolysis and the precipitation of FeOOH nuclei. (E) One-dimensional microfibers pierce out of the PEDOT skin, and as the concentration of hydrolysis products increases, (F) FeOOH nuclei undergo Ostwald ripening forming large one-dimensional microstructures. Upon contact with HCI and EDOT vapor, these microstructures are dissolved and polymerized resulting in one-dimensional core-shell FeOOH-PEDOT microstructures. (G) A high packing density of nanofibers is obtained after a 6 h reaction that reduces ferric ions in the FeOOH core resulting in (H) one-dimensional core-shell FeCl ₂ -PEDOT nanofibers.

FIG. 3A-FIG. 3G. RVPP controls the thickness of freestanding PEDOT films. (A) A digital photograph shows a delaminated freestanding PEDOT film next to its original steel substrate. Cross sectional analyses carried out using scanning electron microscopy show rust layers with thicknesses of (B) 20 mm,

(C) 40 mm, and (D) 70 mm that lead to PEDOT films with thicknesses of (E) 6 mm, (F) 8 mm, and (G) 10 mm, respectively.

FIG. 4A-FIG. 4F. RVPP enables patterning of a freestanding PEDOT film as well deposition of a conformal coating on a three-dimensional substrate. (A) A digital photograph shows a rust layer masked by polyimide tape; this substrate is subsequently utilized for RVPP resulting in selective deposition of (B) a blue PEDOT film. (C) A patterned delaminated PEDOT film is robust retaining its “W- shaped” pattern and composed of nanofibers as shown by scanning electron microscopy. (D) Photographs show that a rusted screw also serves as a substrate for RVPP leading to (E) a conformal blue texturized PEDOT coating; upon delamination of the polymer coating, (F) the screw surface is cleaned and void of rust.

FIG. 5A-FIG. 5H. Spectroscopic characterization of RVPP-PEDOT. (A) Photograph shows that a PEDOT film retains the substrate’s original shape and delaminates by simple immersion in water. (B) High-angle annular dark-field (HAADF) scanning transmission electron micrograph for a PEDOT nanofiber. (C) A zoom in image shows its core-shell structure and an elemental composition composed of a sulfur-containing shell and an iron-containing core. (D) Powder X- ray diffractogram confirms that the core is composed of ferrous chloride— this salt is the product of reduced ferric ions. (E) An FT-IR spectrum shows a doped conjugated backbone. (F) A UV-vis-NIR spectrum of acid-treated PEDOT shows a bipolaronic state, and its (G) powder X-ray diffractogram indicates a polycrystalline structure. (H) Current-voltage curve shows that a freestanding PEDOT film is reversibly doped and de-doped by acid and base, respectively.

FIG. 6A-FIG. 6F. Three-electrode electrochemical characterization of a freestanding RVPP-PEDOT film serving as a working electrode. (A) Cyclic voltammograms, collected in the absence of a current collector using a degassed 1 M H ₂ SO ₄ aqueous electrolyte, show rectangular reversibility at a scan rate of 25 mV s ^-1 , and (B) as the scan rate increases from of 25 to 200 mV s ^-1 , the polymer remains capacitive. (C) Nyquist plots indicate a mixed pseudocapacitive and capacitive behavior remaining stable throughout 500 cycles. (D) Galvanostatic charge-discharge profiles at 0.7, 1.4, 3.5, 7, and 14 A g ^-1 show retention of triangular symmetry and demonstrate PEDOT’s high rate capability. (E) Cyclic voltammograms show similar rectangular curves in 1 and 6 M H ₂ SO ₄ electrolyte. (F) Electrochemical impedance spectroscopy Nyquist plots display more resistance in 6 M H ₂ SO ₄ from electrolyte ion diffusion.

FIG. 7A-FIG. 7I. Performance characteristics of two-electrode RVPP- PEDOT symmetric supercapacitors. (A) Flow process diagram of device fabrication. (B) Cyclic voltammograms collected in 1 M H ₂ SO ₄ at a scan rate of 25 mV s ^-1 using varied voltage windows show stable rectangular capacitive behavior. (C, D) A rectangular-shaped curve and stable capacitive behavior are exhibited as the scan rate increases from 25 to 4000 mV s ^-1 . (E) The device displays a high rate capability retaining gravimetric capacitance at fast scan rates. (F) Galvanostatic charge-discharge curves at current densities ranging from 0.35 to 14 A g ^-1 show stable profiles with (G) corresponding ohmic drops increasing from 1 to 5 mA as a function of current density. (H) The charge- discharge behavior remains stable as the voltage window is widened at a current density of 3.5 A g ^-1 , and (I) 80% of original capacitance is retained after 38,000 cycles.

FIG. 8 is a flow process diagram of rust formation.

FIG. 9 depicts a powder X-ray diffraction pattern showing the heterogeneous composition of the rust film.

FIG. 10 is a mechanistic scheme of the formation of PEDOT via step growth polymerization.

FIG. 11 depicts powder X-ray diffraction patterns of quenched experiments during rust-based vapor-phase polymerization.

FIG. 12A-FIG. 12D depicts scanning electron micrographs of quenched experiments during rust-based vapor-phase polymerization.

FIG. 13 depicts Fourier-transform infrared spectrograms of quenched experiments of PEDOT films during rust-based vapor-phase polymerization. The percent of IR transmittance from C=C decreases with respect to C-O-C as the reaction progresses; the conjugation length of the polymer backbone is maximized by 5 h.

FIG. 14 depicts scanning electron micrographs and powder X-ray diffraction patterns of product from rust-based vapor-phase polymerization without HCI vapor.

FIG. 15 is a scanning electron micrograph of product from rust-based vapor-phase polymerization without EDOT vapor.

FIG. 16 is a scanning electron micrograph of product from rust-based vapor-phase polymerization without chlorobenzene vapor. Nanofiber growth is stifled plausibly due to chlorobenzene’s polar organic structure providing a universal solvent for oligomers and polymer assembly during vapor phase polymerization. FIG. 17 depicts scanning electron micrographs of the underside of a rust- based vapor-phase polymerization PEDOT film.

FIG. 18 depicts sheet resistance and electrical conductivity of PEDOT film versus film thickness.

FIG. 19A-FIG. 19C. Scanning electron micrographs and profilometry measurements for (A) pristine steel, (B) rusted steel sheet cleaned by commercial solution and (C) rusted steel sheet cleaned during rust-based vapor- phase polymerization.

FIG. 20A-FIG. 20B. Microscopic analysis and X-ray diffraction measurements of (A) a pristine steel sheet and (B) a rusted-steel sheet cleaned during rust-based vapor-phase polymerization.

FIG. 21 depicts high angle annular dark field scanning transmission electron micrographs and energy-dispersive X-ray spectrograms of a PEDOT nanofiber; a polymer nanofiber contains an inorganic Fe core and a sulfur- containing PEDOT shell. Nanofibrillar growth occurs concomitantly as a polymer shell deposits on one-dimensional FeOOH iron(lll)-containing crystallites produced from dissolution of rust, hydrolysis, and Ostwald ripening.

FIG. 22 is an optical micrograph of a PEDOT film showing that the surface of the film is comprised a homogeneous distribution of one-dimensional micro structures.

FIG. 23 depicts high angle annular dark field scanning transmission electron micrographs and energy-dispersive X-ray spectrograms collected after a PEDOT film rinsed in 6 M HCI; the acid wash removes the inorganic core resulting in a hollow nanofiber.

FIG. 24 is a selected area diffraction pattern of purified PEDOT.

FIG. 25 is a profilometry measurement of a PEDOT film.

FIG. 26 is a charging and self-discharging curve of a two-electrode symmetric RVPP-PEDOT electrochemical capacitor.

FIG. 27 is a Nyquist plot of a two-electrode symmetric RVPP-PEDOT electrochemical capacitor. FIG. 28 is an image of PEDOT nanofibers generated by RVPP. The colored lines trace the length of individual nanofibers such that an aspect ratio of the nanofibers can be calculated.

FIG. 29 is a schematic depicting the use of a solid-state rust layer as an oxidant precursor rather than a corrosive substrate for vapor-phase polymerization.

FIG. 30 depicts images of PEDOT nanofibers generated from rust formed by exposing a steel substrate to H ₂ SO ₄ , HCI, sea water, or Dl water.

FIG. 31 is a flow process diagram for depositing PEDOT nanofibers onto an electrode.

FIG. 32 is a schematic of the micro-supercapacitor device configuration containing PEDOT generated from RVPP.

FIG. 33 is a graph depicting the potential window capability of the micro- supercapacitor device containing PEDOT generated from RVPP. FIG. 34 is a graph depicting the scan rate capability of the micro- supercapacitor device containing PEDOT generated from RVPP.

FIG. 35 is a graph depicting the capacitance retention of the micro- supercapacitor device containing PEDOT generated from RVPP.

FIG. 36 depicts fractal designs for the micro-supercapacitor device. FIG. 37 is an image of PEDOT / SiO ₂ (Fe ₂ O ₃ /glass).

FIG. 38 is a series of images of PEDOT / Fe ₂ O ₃ (red clay brick).

FIG. 39 is a series of images of PEDOT / b-FeOOH (core-shell fibers).

FIG. 40 is a series of images of PEDOT / TeO ₂ (core-shell fibers).

FIG. 41 is a series of images of PEDOT / SnO ₂ (core-shell fibers). FIG. 42 is a series of images of PEDOT / hard carbon fiber paper.

FIG. 43 is a series of images of PEDOT / carbon cloth.

FIG. 44 is a series of images of polypyrrole / hard carbon fiber paper.

FIG. 45 is a series of images of PEDOT nanofiber, poly(3- thiophenemethanol), and polypyrrole.

FIG. 46 is a series of images of a mat vs. a film.

FIG. 47 is a series of images showing high nucleation concentration leads to vertically directed fibers and low nucleation concentration leads to horizontally directed fibers.

FIG. 48 is a comparison of PEDOT to other conducting polymers in terms of crystallinity (XRD) and electrical conductivity.

FIG. 49. Mechanism and schematic flow process diagram of aerosol vapor polymerization (AVP). a, Oxidative radical polymerization occurs when monomer vapor contacts an oxidant-carrying aqueous droplet resulting in a spherical polymer particle comprised of Cl- doped conjugated polymer chains b, Reactor design uses a 510-cm long coiled tube, ultrasonic nebulizer and reactant bubbler (monomer reservoir). The inset shows suspended yellow oxidant droplets (carrying FeCI3) encountering a flow of monomer vapor; droplet color changes to blue as PEDOT forms. Particles are collected by pushing aerosol products through 3 ethanol-filled bubblers connected in series.

FIG. 50. PEDOT particle size, electrical conductivity and processing in water a left, Digital image of a blue powder of lyophilized PEDOT particles; a right, Dispersed PEDOT particles in water at pH 7 remain suspended for 1 hour b, Size distribution of particles exhibits a 750 nm mode and inset shows their spherical shape via scanning electron microscopy; scale bar is 2 mm. c, Current- voltage curve of a pelletized sample of PEDOT particles is characterized by ohmic behavior; top left inset shows the disc-shaped symmetry of the pellet. Bottom right inset is a scanning electron micrograph of the pellet architecture comprised of spherical particles; scale bar is 2 mm. d, In-line measurements demonstrate that the AVP reactor produces particles with diameter ranging between 250 nm and 2.5 mm; 350 nm is the mean diameter produced by the reactor (red color). The majority of the particles that bypass the ethanol-filled bubblers are 350 nm in diameter or less (blue color). Inset shows illustrations of reactor output and exhaust.

FIG. 51 . Stoichiometry versus structure and electrical conductivity of

PEDOT particles a, Theoretical (black) and experimental (red, blue) concentrations of EDOT vapor under different nitrogen flow rates at different monomer reservoir temperatures. The diffusion of EDOT vapor into the carrier gas is kinetically controlled and the experimental curve possesses a trend that follows the theoretical curve. Note that at higher flow rates, deviation from theory increases b, Electrical conductivity of PEDOT particles produced at different oxidant-to-monomer ratios while holding the total flow rate constant at 4000 seem; error bars are generated from 3 samples c, Fourier-transform infrared spectra show that lowering oxidant-to-monomer ratio results in PEDOT particles with shorter conjugation length as evidenced by the lower peak ratio between C=C and C-O-C. d, Powder X-ray diffraction patterns show PEDOT’s crystalline structure. The inset illustrates the directions assigned to each peak e, X-ray photoelectron spectra of Cl 2p reveal that a lower concentration of EDOT vapor leads to a higher doping level and to dopants with stronger affinity to the polymer backbone.

FIG. 52. Solid State NMR spectra of AVP PEDOT microparticles. 71-kHz matched Hartmann-Hahn proton-carbon cross-polarization 125-MHz 13C NMR spectra of PEDOTs (see the repeat-unit structure, black, and its nearest- neighbor green, panel c) with varying levels of doping (high, panels a and b; medium, panels c and d; and low, panels f and g). Time-domain data acquisition followed a two rotor-period Hahn echo (denoted as “cpecho”) (panels a, d and f). Spin editing involved 80-ms interrupted proton dipolar decoupling during the Hahn echo (panels b and g), or reduction of the cross-polarization contact from 1 ms to 100 ms (panel e). Each spectrum was the result of 80,000 scans with a 4-s repeat time and 6.25-kHz magic-angle spinning.

FIG 53. Submicron-sized AVP-PEDOT particles as universal additives for thermoplastic and cementitious composites a, Dispersibility of PEDOT particles (1 mg/10 mL) and PEDOT:PSS in common organic solvents shows the superior solution processing of our particles b, A polycaprolactone-PEDOT composite is produced using a 25 wt% loading of polymer particles and inset shows a scanning electron micrograph of the composite with a homogeneous particle distribution; scale bar is 1 mm. c, Current-voltage profile dependency on the doping of PEDOT particles is demonstrated as the composite is exposed to an acid (blue) or a base (red) d, PEDOT particles are processable in molten sulfur unlike PEDOT:PSS. e, Photographs show sulfur, sulfur concrete and PEDOT incorporated in sulfur concrete f, Scanning electron micrograph of the cross- section of a piece of sulfur concrete containing PEDOT particles distributed throughout the bulk; scale bar is 10 mm. g, Electrical resistance measurements on PEDOT-sulfur concrete composite are carried out by embedding metal electrodes connected to a multimeter. A 10 wt% addition of PEDOT particles changes sulfur concrete’s infinite resistance to 3.651 kW/square. h, The resistance change with respect to temperature is linear, reproducible and reversible. Error bars are produced using data from triplicate measurements collected every 30 min to allow for temperature homogenization and resistance stabilization i, Electrical resistances of a PEDOT-sulfur concrete composite cycled between room temperature and 80 °C. Measurements demonstrate a stable and reproducible response over 2 months, with standard deviation of 0.0400 (25 °C) and 0.0122 (80 °C).

FIG. 54. Photograph sequence shows the polymerization of EDOT at room temperature when oxidant and monomer are mixed in solution in a vial after a) 0 hour, b) 2 hours, c) 4 hours and d) 16 hours.

FIG. 55. Size comparison between ultrasonic nebulizer and collision nebulizer for droplets and particles a) In-line measurements of droplets produced from ultrasonic nebulizer and collision nebulizer show that a collision nebulizer produces droplets with a broader size distribution compared to its ultrasonic counterpart b) This leads to PEDOT particles with a broad size distribution. Inset shows their spherical shape via scanning electron microscopy; scale bar is 2 mm.

FIG. 56. Scanning electron micrographs and corresponding EDX for PEDOT particles a-d) Particles have a spherical symmetry; scale bar = 2 mm. e- h) Energy-dispersive X-ray spectroscopy images show the presence of carbon, oxygen, sulfur and chlorine elements; scale bars are 2 mm. i) X-ray photoelectron spectroscopy survey shows carbon, oxygen and sulfur ratio. Sulfur content is lower than the theoretical value due to contamination from environmental CO or CO2. j) Fourier-transform infrared spectroscopy shows characteristic bonding for PEDOT. FIG. 57. Single particle conductivity measurement carried out via atomic force microscopy. The l-V profile of single spheres shows ohmic behavior. Top left inset shows topographical information while bottom right shows the schematic of experimental setup; scale bar is 1 mm. The current limit of the probe tip is 5 nA, therefore, curves deviate from ideal linear behavior when current approaches the limit. Plots do not pass through the origin due to a possible systematic error stemming from contact interactions between the charged- particle surface and cantilever.

FIG. 58. Thin film preparation of a PEDOT particle coated-glass slide. A film is prepared via interfacial surface tension gradient resulting in directional fluid flow.

FIG. 59. Transparent films of AVP-particles and UV-vis-NIR characterization of AVP-particle aqueous dispersions a, Transparent PEDOT films supported on glass are produced at the water/oil interface via surface tension induced fluid flow. These photographs show color change in films when exposed to vapors via acid (doped) or base (dedoped). b, Cross-sectional scanning electron micrograph of PEDOT film indicates a thickness of 1.42 mm; scale bar = 1 mm. c, Base treatment results in partial conversion of bipolaronic state to neutral and polaronic PEDOT, thereby decreasing charge carrier density, transparency and conductivity.

FIG. 60. Reactors photos (a) A 510 cm coiled glass serving as tubular reactor; (b) Oil bath heating of coil during reaction. Side arms are heated by heating tapes and wrapped in aluminum foil; blue cables are thermocouples. The flow stream of blue colored PEDOT particles is visible during reaction; (c) A 51 cm straight glass reactor heated by a tube furnace.

FIG. 61 . Comparison between short and longer residence times. Top row images show PEDOT morphology after 20 seconds; discrete particles indicate complete polymerization inside a reactor. Bottom row shows images of particles collected after a 2 second residence time, the aggregated morphology indicates sintering of particles. All scale bars are 2 mm.

FIG. 62. In-line measurements at 600 seem total flow rate. The red bar shows that when using a 600 seem total flow rate, the main product are particles with a 350 nm diameter. All particles are collected by ethanol bubblers and none are exhausted i.e., bypass the ethanol collectors. The inset image shows collected PEDOT particles. Scale bar = 2 mm.

FIG. 63. Theoretical vapor pressure and concentration of EDOT versus temperature.

FIG. 64. Powder X-ray diffraction patterns of PEDOT particles synthesized with different oxidant to monomer ratio show almost identical peaks.

FIG. 65. Conductivity measurements of pelletized PEDOT particles: a) right after synthesis, purification and lyophilization. b) After 6 months in ambient condition.

FIG. 66. Pelletized PEDOT particles current-voltage curves and experimental setup a) Two-point configuration b) Four-point probe configuration.

FIG. 67. Deposition of a nanofibrillar PEDOT coating on brick. (A) Fired brick is coated by PEDOT (dark blue) in a one-step reaction. (B) A brick’s a- Fe2O3 microstructure is partially dissolved by acid vapor to liberate Fe3+, promote hydrolysis and precipitation of FeOOH spindles that control oxidative radical polymerization. As previously reported, monomer vapor reacts with partially dissolved FeOOH nuclei resulting in preferential directional growth of high aspect ratio PEDOT nanofibers (13). (C) A nanofibrillar PEDOT coating exhibits superior adhesion versus the commercial product PEDOT:poly(styrenesulfonate) during Scotch tape tests. (D) Synthesis is scalable to decimeter-sized bricks.

FIG. 68. Nanofibrillar PEDOT-coated bricks for electrochemical electrodes and supercapacitors. (A) Three-electrode cyclic voltammogram at 2 mV/s shows a quasi-rectangular shape stemming from PEDOT’s capacitive behavior with Fe3+/Fe2+ redox pair peaks at 0.37 V and 0.49 V (vs. Ag/AgCI); inset shows a schematic diagram of the working electrode. (B) Cyclic voltammograms for symmetric supercapacitor in 1 M H2SO4 and poly(vinyl alcohol)/ H2SO4 gel electrolyte. (C) Galvanostatic charge-discharge profiles for quasi-solid-state device at current densities ranging between 0.5 and 25 mA/cm2; curves at 1 , 5,

10 and 25 mA/cm2 are horizontally expanded 2x, 10x, 15x and 20x, respectively. Inset shows IR drop at current densities of 5, 10 and 25 mA/cm2. (D) Schematic illustration comparing active sites for charge storage in liquid (top) versus gel electrolytes (bottom). The black dash line in the liquid electrolyte represents a separator. (E) Quasi-solid-state supercapacitor charge-discharge curves after 10,000 cycles at 5 and 25 mA/cm2 exhibit 87% and 90% capacitance retention, respectively (coulombic efficiency is -100%). (F) Photograph of a supercapacitor module lighting up a green light-emitting diode. This tandem device contains three supercapacitors connected in series; the core-shell structure of an electrode is also shown.

FIG. 69. Deposition of PEDOT nanofibers on a-Fe2O3-impregnated carbon cloth for flexible supercapacitors. (A) Flow process diagram shows impregnation of carbon cloth with a-Fe2O3 particles, deposition of PEDOT nanofibers on carbon cloth surface and configuration of a symmetric flexible supercapacitor. (B) Low and high magnification scanning electron micrographs of bare carbon cloth (left column), a-Fe2O3-impregnated carbon cloth (middle column) and nanofibrillar PEDOT-coated carbon cloth (right column); insets show digital photographs of the carbon cloth substrate at each step of the coating protocol.

FIG. 70. Characterization of nanofibrillar PEDOT-coated carbon cloth electrodes and supercapacitors. (A) Thermogravimetric analysis compares PEDOT mass loading in a-Fe2O3-derived (30.2 mg/cm2) and FeCI3-derived (2.1 mg/cm2) electrodes. A high mass loading in a-Fe2O3-derived electrode leads to a high areal capacitance as shown by (B) a quasi-rectangular three-electrode cyclic voltammogram. (C) An electrode is mechanically flexible resulting in supercapacitors that readily bend from 0° to 180° while exhibiting stable cyclic voltammograms. (D) Devices retain -94% capacitance after 500 bending cycles. (E) Capacitive behavior is retained at scan rates up to 25 mV/s and current densities up to 50 mA/cm2 as shown by quasi-rectangular cyclic voltammograms and (F) triangular galvanostatic charge-discharge profiles, respectively.

FIG. 71 . Performance of a nanofibrillar PEDOT-coated carbon cloth flexible supercapacitor. (A) A device is stable and retains -89% of capacitance after 10,000 charge-discharge cycles (collected at 25 mA/cm2) while exhibiting -100% coulombic efficiency. (B) Triangularly symmetric galvanostatic charge- discharge profiles are obtained using voltage windows of 1 V and 1.2 V and current densities of 5, 10 and 25 mA/cm2. (C) Ragone plot shows that areal energy and power densities in our device exceed state-of-the-art metrics for flexible organic supercapacitors surpassing some inorganic pseudocapacitors and batteries.

FIG. 72. Synthesis of a nanofibrillar PEDOT coating on the surface of a brick. (A) Reaction set-up is comprised of two reactant reservoirs in a hydrothermal reactor. (B) The thickness of polymer coating is controlled by reaction time and stoichiometry, generating partially polymerized PEDOT-coated bricks (core/shell architecture) and fully polymerized bricks (monolithic PEDOT architecture). (C) Cross-sectional scanning electron micrographs of PEDOT- coated brick shows intimate grafting of PEDOT to porous brick microstructure; the polymer coating is comprised of a mat of nanofibers. (D) A diagram of reaction pathways shows both acid and Fe3+ serving as initiators for the polymerization of EDOT. Acid leads to acid-catalyzed polymerization producing oligomers whereas Fe3+ results in oxidative radical polymerization generating PEDOT. (E) Acid concentration ([HCI]) determines the reaction pathway as demonstrated by control of coating thickness and electrical resistance. These experiments are carried out using an EDOT vapor concentration of 5.2 mM and 6.8 mM. We calculate vapor concentration [HCI]vfrom the volume of concentrated HCI used (VHCI) and by assuming complete HCI evaporation. A lower [HCI]v increases PEDOT coating thickness because acid-catalyzed polymerization is minimized. Thicker coatings are produced by holding [HCI]v constant while increasing EDOT concentration. The lowest electrical resistance is obtained using 14 mM [HCI]v because this concentration promotes a both steady-state dissolution rate and oxidative radical polymerization. (F) A high concentration of HCI leads to acid-catalyzed polymerization and uncontrolled polymerization that results in oligomers of dark color. Photographs show coatings on monomer reservoirs that formed in situ during the reaction. (G) Plot of a PEDOT coating’s thickness versus electrical resistance collected at different EDOT concentrations ([EDOT]v) and using [HCI]v = 9.6 mM. A higher [EDOT]v increases the coating thickness however it also leads to higher resistance. In this plot, [EDOT]s = concentration of EDOT in chlorobenzene solution, and [EDOT]v = EDOT vapor concentration (calculated assuming all the EDOT evaporates). (H) Time-dependent plots of reactions for quenching experiments show thickness of a PEDOT coating versus electrical resistance.

FIG. 73. Electron microscopy, thermogravimetric analysis, energy dispersive spectroscopy and patterning of a PEDOT-coated brick. Scanning electron micrograph shows nanofibers and ImageJ software traces used for calculating (A) length and (B) width of nanofibers. (C) Histograms show the results of dimensional analysis. (D) Thermogravimetric analysis of brick and PEDOT-coated brick shows that the latter contains a 2.8 wt% mass loading of polymer. (E) Energy-dispersive X-ray spectra and maps of a purified PEDOT coating show C, O, S and Cl signals pertaining to a Cl- doped PEDOT structure; elemental ratios are non-stoichiometric due to brick’s impurities. (F) A patterned brick is produced using a tape mask during synthesis.

FIG. 74. Schematic illustration and photographs of flow process diagram for fabricating a sealed pipet reactor, nanofibrillar PEDOT coated hematite and photothermal characterization set-up. Step-by-step fabrication is shown in (A) schematic illustrations and (B) photographs. (C) Scanning electron micrograph shows a sample of hematite (a-Fe2O3) conformally coated by PEDOT nanofibers after synthesis. (D) Photothermal measurement involves an infrared camera that records a temperature map, a camera flash for injecting photons into a sample and a white paper background.

FIG. 75. PEDOT synthesis on natural minerals and rocks. PEDOT coatings on (A) hematite (a-Fe2O3), (B) rosy pink granite (contains a-Fe2O3), (C) pyrite (FeS2), (D) chalcopyrite (CuFeS2), (E) troilite (FeS) and (F) pyrrhotite (Fe1-xS). Synthesis is carried out by liberating Fe3+ from a mineral via dissolution using HCI vapor where Fe3+ serves as the oxidizing agent for controlling oxidative radical polymerization. A PEDOT coating deposits on a partially dissolved mineral surface thereby changing a mineral’s color to dark blue and interestingly, this PEDOT coating is characterized by facile photothermal excitation. A PEDOT coating heats up when exposed to a camera’s flash because it absorbs infrared energy. We use a thermal camera to compare the surface temperature between PEDOT-coated and pristine minerals after exposure to a flash of light (shown as temperature maps above each mineral). Sulfur-containing minerals (pyrite, chalcopyrite and troilite) decompose upon heating and leach acid that dopes a PEDOT coating. Doping is detected by a decrease in the electrical resistance of the polymer coating (histograms on the right show electrical resistance for each corresponding PEDOT-coated mineral) (see supplementary text). The surface two-point probe resistance of most minerals decreases after a PEDOT coating is deposited (except for pyrrhotite’s).

FIG. 76. Electrochemical characterization of PEDOT-coated brick electrodes and supercapacitors in liquid electrolyte. (A) Three-electrode cyclic voltammogram shows a smaller curve area for 1 M Na2SO4 electrolyte compared to 1 M H2SO4 , indicating lower capacitance. Inset shows digital photos of front and back of a PEDOT-coated brick electrode. (B) Three-electrode cyclic voltammograms in 1 M H2SO4 collected at scan rates ranging from 2 to 100 mV/s. (C) Electrode Nyquist plots are collected using different electrolytes (inset shows equivalent circuit diagram); fitted data is shown by solid lines and experimental data is shown by segregated points. (D) Step-by-step fabrication of a liquid-electrolyte based supercapacitor where 1) electrode is fabricated by attaching a Pt wire to a PEDOT-coated brick using polyimide tape. 2) A polymer separator is sandwiched between two electrodes held together using a binder clip and epoxy (the binder clip is removed once the epoxy cures). 3) Top view of a supercapacitor after epoxy cures. 4) Supercapacitor is placed in a petri dish containing 1 M H2SO4 electrolyte. (E) Nyquist plot of supercapacitor shows a Warburg region and internal resistance of 3 W. (F) Cyclic voltammograms for a supercapacitor collected at scan rates ranging from 2 to 25 mV/s. (G) Galvanostatic charge-discharge profiles for supercapacitor at current densities ranging between 0.5 and 25 mA/cm2 (inset shows IR drop at 0.5 and 1 mA/cm2). (H) Cyclic voltammograms and (I) galvanostatic charge-discharge profiles retain the shape of the curve at voltage windows of 1 V and 1.2 V.

FIG. 77. Electrochemical characterization of a tandem device comprised of three supercapacitors connected in series. (A) Cyclic voltammograms for a single device versus tandem device at 10 mV/s. (B) The supercapacitor lights a white light-emitting diode with (C) forward voltage (“turn-on” voltage) of 2.546 V. (D) Discharging profile of supercapacitor during lighting of light-emitting diode shows a decrease in voltage from 2.7 V to 2.4 V in 10 min with diminishing light intensity.

FIG. 78. Quasi-solid-state supercapacitor based on PEDOT-coated brick electrodes. (A) Supercapacitor is comprised of two electrodes (each 2.8 mm thick) and a poly(vinyl alcohol)/H2SO4 gel electrolyte layer (0.7 mm thick) and the entire device is sealed in epoxy. (B) The brick-gel-brick sandwich structure (518 mg) withstands a shearing force more than 1000 times its own weight - this is carried out by attaching an aluminum block (625 g) to the electrode and by allowing gravity to pull on it. (C) Nyquist plot shows an internal resistance of 2.5 W and the absence of semicircle associated with charge transfer resistance. (D) Cyclic voltammogram at scan rates ranging from 2 to 25 mV/s. (E) By cutting a brick electrode in half, one face of the PEDOT coating is removed resulting in a 50% decrease in the areal capacitance using a liquid electrolyte (calculated from cyclic voltammograms at a scan rate of 5 mV/s). This 50% lower magnitude in capacitance is similar to that of a quasi-solid-state device fabricated using fully coated bricks because gel permeation is limited to a single face. (F) Cyclic voltammogram of unsealed supercapacitor shows continuous capacitance degradation at room temperature due to solvent evaporation as indicated by a decreasing curve area. Histogram (right) shows capacitance recovery after wetting (blue line); note that curve area increases beyond original capacitance due to the creation of a liquid ionic percolation network after wetting. (G) Cyclic voltammograms and (H) Nyquist plots, collected during heating-cooling cycles between -20 °C and 60 °C, show that (I) capacitance is temperature-dependent and reversible. (J) Cyclic voltammograms for a single supercapacitor are collected at voltage windows of 1 V and 1 .2 V; a tandem device (comprised of three supercapacitors connected in series) exhibits a 3.6 V window. The tandem device is waterproof exhibiting a stable cyclic voltammogram while immersed in water. (K) Supercapacitor lights up a green light-emitting diode with (L) a forward voltage of 2.155 V.

FIG. 79. Scale-up of a quasi-solid-state supercapacitor based on PEDOT- coated brick electrodes. (A) Casting of poly(vinyl alcohol)/H2SO4 gel electrolyte on large PEDOT-coated brick electrodes (2 x 1 x 1 cm). Photographs show casting of electrolyte and permeation into brick resulting in a dry thin film (right top). A second casting of gel electrolyte, followed by drying, results in a thick gel electrolyte layer (right bottom). (B) Electric circuit diagram for three supercapacitors connected in series and used for lighting up a light-emitting diode; dark blue rectangles represent PEDOT-coated bricks and turquoise color represents the gel electrolyte. Digital photographs show the (C) front and the (D) back of a quasi-solid-state supercapacitor module. (E) Raw photographs (post- processing omitted) show 1) tandem device lighting up a green light-emitting diode and 2) the core-shell architecture of a nanofibrillar PEDOT-coated brick electrode.

FIG. 80. PEDOT coatings on carbon cloth and concrete: comparing a- Fe2O3-derived and FeCI3-derived PEDOT electrodes. (A) Thermogravimetric analysis shows ~18 wt% a-Fe2O3 content in carbon cloth after impregnation via drop-casting. (B) Cross-sectional scanning electron micrographs of carbon cloth, a-Fe2O3-impregnated carbon cloth and a-Fe2O3-derived PEDOT electrode. (C) Scanning electron micrographs for a FeCI3-derived PEDOT electrode show top- view (left top), cross-section (right) and close-up (left bottom); unfortunately, this synthesis leads to electrode areas with a heterogeneous distribution of nanofibers. (D) Optical micrographs of carbon cloth (top), a-Fe2O3-derived PEDOT electrode (middle) and FeCI3-derived PEDOT electrode (bottom) are collected with backlighting. Notice the diminished light transmittance for the a- Fe2O3-derived PEDOT electrode due to its dense nanofibrillar packing architecture. (E) A bar of concrete (left column) is coated with a-Fe2O3 particles (middle column) via dip coating in aqueous dispersion and used for synthesis resulting in PEDOT-coated concrete (right column). This composite exhibits a 1 kW two-point probe resistance and is comprised of a core/shell PEDOT/concrete structure. (F) Three-electrode cyclic voltammograms for a-Fe2O3-derived PEDOT-coated carbon cloth electrode collected in 1 M H2SO4 electrolyte at 2 and 25 mV/s. Note that the polymer is purified in concentrated acid and rinsed in methanol to remove all traces of iron. (G) Nyquist plots show internal resistance for carbon cloth (1.08 W), a-Fe2O3-derived PEDOT electrode (1.08 W) and FeCI3-derived PEDOT electrode (1.15 W).

FIG. 81 . Electrochemical characterization of nanofibrillar PEDOT-coated carbon cloth flexible supercapacitors. (A) The mechanical stability of a-Fe2O3- derived PEDOT electrode is enhanced by applying a coating of poly(vinyl alcohol) that increases adhesion between carbon and PEDOT. A supercapacitor fabricated without this pre-coating shows a distorted cyclic voltammogram after bending due to PEDOT delamination (inset photo shows a device encapsulated by polyimide tape). (B) Nyquist plots show lower internal resistance at higher bending angles thus indicating increased contact between cell components. (C) The voltage window is extended from 1 V to 1.2 V resulting in minimal change in a curve’s shape both at 2 mV/s and 10 mV/s. (D) Cyclic voltammograms for single and tandem (three supercapacitors connected in series) devices at 25 mV/s. (E) The tandem device is rolled into a cylinder and able to light up a white light-emitting diode for 5 min and its (F) discharging profile is plotted.

FIG. 82. Fabrication process and structural characterization of a 3D nanofibrillar PEDOT mSC. a) Si02 is deposited on Si wafer, photoresist is spin coated and the interdigitated pattern is produced via laser writer after exposure and develomment b) Adhesion layer (Or), current collector (Au) and oxidant precursor (Fe2O3) are deposited sequentially, the latter via sputtering. Chemical conversion is carried out by i) liberating Fe3+ from Fe2O3 via dissolution using HCI vapor and ii) oxidizing EDOT vapor with Fe3+ to produce PEDOT nanofibers c) A nanofibrillar PEDOT mSC is obtained after lift-off. d) Digital photograph shows the mSC configuration consisting of 2 electrodes with gold pads and each possessing 5 PEDOT-coated fingers e) A scanning electron micrograph shows a close-up of 200 mm wide polymer-coated fingers and demonstrates a gap void of polymer f) Cross-sectional electron micrograph captures device layers (augmented by color) and the active polymer coating comprised of a carpet of vertically directed PEDOT nanofibers.

FIG. 83. Direct characterization on intrinsic properties of a nanofibrillar PEDOT electrode a) Scanning electron micrograph shows bulk 1 D electrode morphology b) Transmission electron micrograph and c) high-angle annular dark-field STEM image of single fibers confirm a core-shell structure d) EDX maps for a nanofiber show an elemental composition consisting of Fe and S. e) Raman spectrum is characteristic of an oxidized conjugated backbone possessing high doping f) l-V curves collected throughout the entire polymer coating (inset) show ohmic behavior indicative of a homogenous percolation network g) PXRD confirms a polycrystalline structure with three characteristic peaks h) The electrode/gap interface, probed by i) line scan atomic force microscopy (green arrow), reveals j) a -250 nm thick polymer coating k) Four- point probe, carried out on a modified electrode (inset), aids in measuring electronic conductivity.

FIG. 84.. Electrochemical performance of nanofibrillar PEDOT mSCs in 1 M H2SO4 aqueous electrolyte a) Cyclic voltammograms, collected at a scan rate of 25 mV s-1 and under various voltage windows, show stable capacitive behavior and retain b) a rectangular shape under fast scan rates ranging from 1 to 50 V s-1. c) Nyquist plots show stable ESR and similar capacitive behavior using scan rates of 25 mV s-1 and 50 V s-1 after 500 cycles d) Galvanic charge- discharge curves, collected at a current density of 100 mA cm-2, for electrodes generated from 60 nm, 120 nm and 180 nm thick Fe2O3 layers show e) a 90% capacitance retention after 10,000 cycles f) Volumetrically normalized Ragone plot compares our mSCs with 2D lithium film battery, Al electrolytic capacitor, activated carbon commercial supercapacitor as well as 2D PEDOT, PANi/rGO, Co(OH)3/rGO and Mn02/rGO based micro-supercapacitors.

FIG. 85. Electrochemical performance of PEDOT mSCs, generated from a 60 nm Fe2O3 layer, and fabricated with various gap distances and fractal geometries using 1 M H2SO4 aqueous electrolyte a) Photograph and illustrations of micro-supercapacitors with gap distances of 500 mm and 200 mm. b) A Nyquist plot for the 200 mm gap device shows lower impedance and c) its cyclic voltammogram shows higher capacitance d) Schematic representations of interdigitated (L0) and fractal electrodes (L1) possessing a 200 mm gap. A fractal electrode augments ion diffusion pathways thereby lowering impedance in e) a Nyquist plot and increasing capacitance in f) a CV.

FIG. 86. Electrochemical performance of quasi-solid-state mSCs in 1 M H2SO4/PVA gel electrolyte and its temperature-dependent behavior a) Cyclic voltammograms compare aqueous and gel electrolytes at a scan rate of 1 V s-1. b) Nyquist plot for aqueous electrolyte exhibits lower impedance and more ideal capacitive behavior versus gel electrolyte; this is also confirmed via Bode plots (inset) c) After 10,000 cycles with a gel electrolyte, 94% of original capacitance is retained and rectangular-shaped cyclic voltammograms are obtained at scan rates of 10 V s-1 , 20 V s-1 and 50 V s-1. d) Schematic illustration of experimental setup for studying temperature e) Cyclic voltammograms of a quasi-solid-state mSC show increasing capacitance as temperature rises from 25 °C to 60 °C; note that at 70 °C, capacitance is restricted f) Nyquist plots exhibit similar ESR between 25 °C - 60 °C, however, as temperature increases, capacitive behavior diminishes resulting in resistive charge transfer (semi-circle) and Warburg impedance as temperature approaches 70 °C.

FIG. 87. Cross-sectional SEM image shows thickness for each layer in our device including iron oxide, gold/chromium, and silicon dioxide layer. These have been painted with Photoshop to delineate boundaries.

FIG. 88. Schematic illustration of our rust-based vapor-phase polymerization setup. A glass reactor is loaded with individual containers carrying concentrated hydrochloric acid,

EDOT/chlorobenzene, and a micro-fabricated electrode (sample). This glass reactor is then sealed and heat to 150 °C in an oven for 1 .5 h.

FIG. 89. Mechanistic scheme for the formation of PEDOT via step-growth polymerization.

FIG. 90. Left SEM image shows photoresist deformation during rust- based vapor-phase polymerization. Right SEM image shows the formation of thin PEDOT film over the photoresist at the electrode/gap interface.

FIG. 91 . Left SEM image show a morphological change on nanofibers after a 5 min oxygen plasma treatment that fused fibers together. Right SEM images shows a rupture in the nanofibrillar polymer film after sonication using a common bath sonicator.

FIG. 92. SEM image shows a line mark created by an undercut at the electrode/gap interface that led to successful lift-off.

FIG. 93. Photograph image shows Scotch tape test being performed on a micro-supercapacitor. Notice the absence of polymer on tape after pulling it off.

FIG. 94. Close-up SEM image of a nanofibrillar PEDOT coating on a current collector.

FIG. 95. SEM images of PEDOT nanofibers synthesized on current collectors possessing aspect ratios of 10 nm (left), 30 nm (middle) and 100 nm (right).

FIG. 96. Images of EDX maps for a micro-supercapacitor collected after synthesis. Note that purification of polymer was omitted as indicated by iron signal.

FIG. 97. Autocad file (left) and optical photograph (right) of electrode configuration utilized for carrying out l-V and conductivity measurements.

FIG. 98. Schematic illustrations of two micro-supercapacitor configurations. Left diagram shows copper wires soldered on Au pads for connecting external circuit. Right diagram shows platinum leads stuck on Au pads via adhesive Kapton tape.

FIG. 99. Profilometry data for 250 mm (left), 600 mm (middle), and 900 mm (right) thick nanofibrillar PEDOT films generated from 60 nm, 120 nm, and 180 nm thick Fe2O3 layers, respectively.

FIG. 100. Cyclic voltammograms and Nyquist plots for polymer coatings generated from 60 nm, 120 nm and 180 nm thick Fe2O3 layers.

FIG. 101. Galvanostatic charge and discharge curves for a polymer coating generated using a 60 nm Fe2O3 layer - data is collected under different current densities ranging from 20 pA cm-2 to 1 ,000 pA cm-2.

FIG. 102. Ragone Plot for a PEDOT mSC normalized by area.

FIG. 103. Optical images for devices generated from a 60 nm thick Fe2O3 layer and with gap distances of 500 mm (a), 200 mm (b), and 100 mm (c). These same devices are then coated with PEDOT nanofibers (d-f) and their corresponding clean gaps (void of polymer) are shown in SEM images (g-i).

FIG. 104. Comparison of cyclic voltammograms (left) and Nyquist plots (right) for liquid and quasi-solid-state devices both possessing a 500 mm gap distance.

FIG. 105. Comparison of cyclic voltammograms (left) and Nyquist plots (right) for devices possessing fractal electrodes; polymer coatings were generated from 60 nm and 120 nm thick Fe2O3 layers.

FIG. 106. Schematic diagram shows comparison between solution-loaded versus particle-impregnated carbon cloth for synthesis a) Structure of carbon cloth b) Loading carbon cloth with FeCl ₃ solution causes solvent evaporation upon heating and forms a thin layer of FeCl ₃ on carbon fibers that limits polymerization and results in low PEDOT mass loading in PEDOT-coated carbon cloth c) a-Fe2O ₃ particles efficiently impregnate carbon cloth and liberate Fe ³⁺ using HCI vapor during synthesis, producing thick PEDOT-coated carbon cloth exhibiting high mass loading and a hierarchical sandwiched structure.

FIG. 107. Microscopic characterization of carbon cloth substrate at each step of the electrode synthesis a) Pristine carbon cloth (grey color, inset) is comprised of interwoven carbon fiber bundles. b) Cross-section profile of carbon cloth shows the space between fibers; the inset is an optical micrograph with background light showing a ~ 200 mm mesh size c) After a-Fe ₂ O ₃ particle impregnation, the carbon cloth changes color from grey to red (inset) d) The a- Fe203 particles pack the space between carbon fibers with a high filling factor that decreases mesh size to ~ 100 mm (inset) e) Vapor-phase polymerization homogeneously deposits PEDOT nanofibers on the carbon cloth surface f) Profile of thick PEDOT-coated carbon cloth shows a hierarchical structure where PEDOT coats the surface leaving the cloth center full of void space. The mesh size is ~ 100 mm after PEDOT coating (inset) g) PEDOT-coated carbon cloth synthesized from FeCl ₃ solution loading exhibits sparse PEDOT nanofibers h) The absence of PEDOT between carbon fibers leads to a low filling factor showing mesh size ~ 200 mm similar to that of the pristine carbon cloth.

FIG. 108. Three-electrode characterization of nanofibrillar PEDOT-coated carbon cloth electrodes a) Thermogravimetric analysis compares PEDOT mass loading in a-Fe ₂ O ₃ -derived (30.2 mg/cm ² ) and FeCl ₃ -derived (2.1 mg/cm ² ) electrodes b) The setup and electrode configuration of three-electrode measurements c) Cyclic voltammograms of a-Fe ₂ O ₃ -derived PEDOT-coated carbon cloth electrode show quasi-rectangular shape at scan rates below 25 mV/s and change to fusiform at 100 mV/s. d) Nyquist plots show internal resistance for carbon cloth (1.08 W), a-Fe ₂ O ₃ -derived (1.08 W) and FeCl ₃ -derived (1 .15 W) PEDOT electrodes, as well as a Warburg region in the a-Fe2O ₃ -derived electrode e) A high mass loading in a-Fe2O ₃ -dehved electrode leads to a high areal capacitance compared to carbon cloth and FeCl ₃ -derived electrodes f) Comparison of areal capacitance (top), total electrode’s gravimetric capacitance (middle) and PEDOT’s gravimetric capacitance (bottom) between a-Fe2O ₃ - derived and FeCl ₃ -derived PEDOT electrodes.

FIG. 109. Electrochemical performance of a nanofibrillar PEDOT-coated carbon cloth flexible supercapacitor a) Schematic illustration of the supercapacitor comprises separator, electrodes, 1 M H2SO4 electrolyte, current leads and polyimide tape sealing b) Quasi-rectangular cyclic voltammograms and c) triangular galvanostatic charge-discharge profiles show capacitive behaviors at scan rates up to 25 mV/s and current densities up to 50 mA/cm ² , respectively d) The device is stable and retains -89% of capacitance after 10,000 charge-discharge cycles (collected at 25 mA/cm ² ). e) Galvanostatic charge-discharge profiles and f) cyclic voltammograms retain the shape after extending the voltage windows from 1 V to 1.2 V albeit with an increased IR drop (e inset) g) Areal capacitance of the supercapacitor obtained from galvanostatic charge-discharge measurements at 1 V and 1.2 V voltage windows and between 0.5 - 50 mA/cm ² current densities h) Ragone plot shows areal energy and power densities of our device exceeding state-of-the-art metrics for flexible organic supercapacitors and surpassing some inorganic pseudocapacitors and batteries.

FIG. 110. Bending tests and a tandem supercapacitor a) An electrode is mechanically flexible resulting in supercapacitors that readily bend from 0° to 180° while exhibiting stable cyclic voltammograms. b) Nyquist plots show lower internal resistance at higher bending angles indicating increased contact between cell components c) Devices retain -94% capacitance after 500 bending cycles between 0° and 180°. d) Cyclic voltammograms for single and tandem (three supercapacitors connected in series) devices at 25 mV/s. e) The tandem device is rolled into a cylinder and lights up a white light-emitting diode for 100 s with f) discharging profile plotted. FIG. 111. Schematic illustration of electrode synthesis procedures. The a- Fe2O ₃ -impregnated carbon cloth is placed on a reservoir inside an autoclave reactor. The reservoir is loaded with 200 mL 0.85 M EDOT / chlorobenzene solution and 20 mL 11 M HCI solution. After reacting at 160 °C for 14 h, the generated nanofibrillar PEDOT-coated carbon cloth is washed in methanol and hydrochloric acid then dried in air before characterization.

FIG. 112. Powder X-ray diffraction of the red pigment shows patterns of a-

Fe203.

FIG. 113. Thermogravimetric analysis for determining mass loading a) Temperature-dependent plots of a-Fe ₂ O ₃ (directly obtained from red pigment) and a-Fe ₂ O ₃ -impregnated carbon cloth show that carbon cloth starts to degrade at 600 °C and a-Fe ₂ O ₃ remains stable, leading to ~18 wt% a-Fe ₂ O ₃ content b) Time-dependent plots show a complete carbon cloth degradation in air at 800 °C after 1 h while a-Fe ₂ O ₃ remains stable c) Time-dependent plots of PEDOT- coated carbon cloth electrodes derived from a-Fe ₂ O ₃ and FeCl ₃ respectively show a two-step degradation for determining PEDOT mass loading. The first step completely degrades PEDOT at 500 °C in air for 1 h to calculate PEDOT mass loading, and the second step completely degrades carbon cloth at 800 °C in air for 1 h to confirm the complete removal of inorganic species after electrode purification.

FIG. 114 contains a series of images illustrating a PEDOT coating for use in water purification in accordance with one aspect of the disclosure.

FIG. 115. contains a pair of images illustrating the inhibition of bacterial growth near a positive electrode with a PEDOT coating in accordance with one aspect of the disclosure.

FIG. 116. is a schematic illustration showing an arrangement of heating elements used to produce a PEDOT-based device in situ within a brick cavity in accordance with one aspect of the disclosure.

FIG. 117. is a schematic diagram illustrating the PEDOT-based device of FIG. 116 with additional caps containing inlet and outlet openings in accordance with one aspect of the disclosure. FIG. 118 is a cross-sectional view of the PEDOT-based device of FIG.

117.

FIG. 119 is a flowchart illustrating a method of forming a PEDOT-based electrical device within a brick cavity opening in accordance with one aspect of the disclosure.

FIG. 120 is a schematic illustration of a method of forming a PEDOT- coated Fe203/Fe3/04 nanoparticle in accordance with one aspect of the disclosure.

FIG. 121 is a graph summarizing water uptake by various substrates with varying PEDOT coating levels at a range of relative humidities.

DETAILED DESCRIPTION OF THE INVENTION

The present disclosure is based, at least in part, on the discovery that rust, when treated with an acid, dissolves affording an ideal source of ferric (Fe ³⁺ ) ions for vapor phase polymerization and synthesis of conductive nanofibrillar layers. As shown herein, novel compositions and methods were developed for the conversion of rust into conductive poly(3,4- ethylenedioxythiophene) (PEDOT) nanofibrillar layers, polypyrrole, thiophene derivatives, as well as other polymers using rust-based vapor phase polymerization (RVPP) and/or aerosol vapor polymerization (AVP).

As described herein, RVPP and AVP can produce stable, highly conductive PEDOT nanofibrillar layers that are suitable for use in supercapacitors. RVPP and AVP do not require the use of the corrosive, expensive, and hygroscopic iron-containing salts that are typically used in vapor phase polymerization.

The presently disclosed compositions and methods are different from previously disclosed compositions and methods in that the presently disclosed layers are covalently bound to the substrate as a single structure.

The presently disclosed fiber layers have electronic properties that differ from conventional materials and so do their functionality. Furthermore, the synthetic procedure can be performed in one step. But conventional methods would require a two-step process if the fiber layer were covalently bound, not one step as described herein.

SUBSTRATES

As described herein, a substrate can be of any material that comprises Fe ³⁺ or any material from which Fe ³⁺ can be liberated. For example, the substrate can be rust, steel, or any material that can provide Fe ³⁺ . For example, the substrate can comprise solid-state rust (or a rust layer). As another example the substrate can be coated with a material comprising Fe ³⁺ or any material from which Fe ³⁺ can be liberated.

Rust

As described herein, the substrate can be rust on a material, such as steel or rust in or coated on glass, plastics, clay, ceramics, carbon, cement, sulfur, concrete, apple juice, cereal, or brick, among other materials and substances, where the rust is not chemically coordinated to steel.

Rust can be a product of iron corrosion, which can result from a redox reaction between iron and oxygen in the presence of water (or any other oxidizer). Rust can also be a product of a reaction between iron and an anion (e.g., Cl-, CO ₃ ^2- , SO ₄ ^2- ) in a low-oxygen environment, such as “green rust”.

Iron corrosion species such as hematite (a-Fe2O ₃ ), maghemite (g-Fe2O ₃ ), goethite (a-FeOOH), and lepidocrocite (g-FeOOH), first documented ca. 800 BCE, make up the solid-state chemical family composed of iron oxides, oxyhydroxides, and hydroxides that are known as rust. Natural sources of these species can be found all over the world. Rust is a byproduct of redox reactions between iron and oxygen in the presence of water. Rust is thermodynamically stable, ubiquitous, and inexpensive. Because rust contains ferric ions (Fe ³⁺ ), it can serve as an attractive candidate for developing oxidative chemical reactions. The ferric ion, with an oxidation potential of 0.77 V, is a nominal oxidizing agent used in the syntheses of conducting polymers such as poly(3,4- ethylenedioxythiophene) (PEDOT), polythiophene, and polypyrrole resulting in doped states and long conjugation lengths (wherein the conjugation length is considered long if the length is sufficient to result in bulk conductivity higher than 100 S/cm. Rust can be a heterogeneous solid-state material comprised of multiple species, such as iron oxide, iron hydroxide, or iron oxyhydroxide.

Rust can be a homogeneous material comprised of a single phase (e.g., Fe2O ₃ ), such as in construction red bricks.

Irond(III)-containinq substrate: Rust-generating substrate or rust-containing substrate

Because rust is inert and lacks chemical coordination to steel, the inventors discovered it would be an ideal substrate for PEDOT synthesis.

As described herein, the iron(lll)-containing substrate can be a rust- containing substrate or a rust-generating substrate. A rust-generating substrate or rust-containing substrate can be any substrate that is capable of producing ferric ions (Fe ³⁺ ) when contacted with an iron oxidizing agent. For example, the rust-generating or rust-containing substrate can be a recycled material, such as recycled bricks, stone, rocks, concretes, scrap metal, iron-containing materials found in the ocean, or any source of rust.

The iron(lll)-containing substrate can be any material that is coated with an iron(lll) containing material or comprises an iron(lll) containing material or can be oxidized to comprise an iron(lll) containing material. The iron(lll)-containing substrate can be a substrate comprising iron(lll), such as steel, stainless steel, silicon steel, tool steel, bulat steel, chromoly, crucible steel, Damascus steel, high-strength low-alloy (HSLA) steel, high-speed steel, maraging steel, Reynolds 531®, Wootz steel, cast iron, wrought iron, anthracite iron, pig iron, nickel-iron alloy, nickel-cobalt-iron alloy, ferroboron, ferrochrome, ferromagnesium, ferromanganese, ferromolybdenum, ferronickel, ferrophosphorus, ferrotitanium, ferrovanadium, orferrosilicon.

As another example, the iron(lll)-containing substrate can be a substrate comprising iron(lll) or a substrate coated with an iron(lll) containing material, such as glass (Fe ₂ O ₃ /SiO ₂ ), brick (e.g., red clay brick), b-FeOOH (core-shell fibers), TeO ₂ (core-shell fibers), SnO ₂ (core-shell fibers), hard carbon fiber paper, carbon cloth, plastic, clay, ceramic, carbon, cement, sulfur, concrete, tiles, apple juice, or cereal, among other materials and substances. As another example, the iron(lll)-containing substrate can be a mineral or comprise a mineral selected from one or more of: hematite (Fe ₂ O ₃ ), rosy granite, pyrite (FeS ₂ ), chalcopyrite (CuFeS ₂ ), troilite (FeS), or pyrrhotite (Fe1-xS).

As another example, a substrate can be a substrate engineered to incorporate Fe(lll) (e.g., rust) into a: thermoplastic such as poly(lactic acid), polycaprolactone, polyvinyl chloride, teflo, polystyrene, clay, ceramics, carbon fibers, carbon nanotubes, graphene, graphene oxides, reduced graphene oxides, cement, sulfur, or concrete.

As another example, a substrate can be a substrate comprising iron oxide. For example, an iron oxide can comprise any iron oxide that contains a Fe(lll), such as Fe ₂ O ₃ , a-Fe ₂ O ₃ (hematite), b-Fe ₂ O ₃ , g-Fe ₂ O ₃ (maghemite), or s- Fe ₂ O ₃ or a mixture of Fe(ll) and Fe(lll), such as Fe ₄ O ₄ (magnetite), Fe ₄ O ₄ , Fe ₅ O ₆ , Fe5O ₇ , Fe ₂₅ O ₃₂ , or Fe ₁₃ O ₁₉ . The presence of Fe(II) in combination with Fe(III) does not appear to interfere with the synthesis of PEDOT fibers.

As another example, a substrate can be a substrate comprising iron hydroxide. For example, an iron hydroxide can be Fe(OH) ₃ (bernalite).

As another example, a substrate can be a substrate comprising iron oxyhydroxide. For example, an iron oxyhydroxide can be a-FeOOH (goethite), b- FeOOH (akaganeite), g-FeOOH (lepidocrocite), d-FeOOH (feroxyhyte), (Fe ³⁺ ) ₂ O ₃ ·0.5H ₂ O (ferrihydrite), high-pressure FeOOH, Fe ₈ O ₈ (OH) ₆ (SO nH ₂ O (Schwertmannite), GR(CO ^2- 3): [Fe ²⁺ 4Fe ³⁺ 2(HO-) ₁₂ ] ²⁺ · [CO ^2- 3·2H ₂ O] ^2- (carbonate green rust), GR(CI-): [Fe ²⁺ 3Fe ^3* (HO-) ₈ ] ⁺ · [Cl--nH ₂ O] (chioride green rust), or GR(SO ^2- ₄ ): [Fe ²⁺ ₄ Fe ³⁺ O-) ₁₂ ] ² · [SO ² - ₄ -2H ₂ O] ² - (sulfate green rust).

IRON OXIDIZING AGENT

As described herein, the rust can be generated from a material comprising iron, such as steel. For example, the rust can be generated by exposing the material comprising iron to an iron oxidizing agent.

The iron oxidizing agent can comprise oxygen (O ₂ ), ozone (O ₃ ), hydrogen peroxide (H ₂ O ₂ ), fluorine (F ₂ ), chlorine (CI ₂ ), halogens, nitric acid (HNO ₃ ), nitrate compounds, sulfuric acid (H ₂ SO ₄ ), peroxydisulfuric acid (H ₂ S ₂ O ₈ ), peroxymonosulfuric acid (H ₂ SO ₅ ), chlorite, chlorate, perchlorate, halogen compounds, hypochlorite, bleach (NaCIO), hexavalent chromium compounds, chromic acid, dichromic acid, chromium trioxide, pyridinium chlorochromate, chromate compounds, dichromate compounds, permanganate compounds, potassium permanganate, sodium perborate, nitrous oxide (N ₂ O), nitrogen dioxide (NO ₂ ), dinitrogen textroxide (N ₂ O ₄ ), potassium nitrate (KNO3), sodium bismuthate, ammonium peroxydisulfate (NH ₄ ) ₂ S ₂ O ₈ , or electrochemical oxidation, optionally in the presence of water.

FERRIC ION (FE ³⁺ ) LIBERATING AGENT

As described herein, dissolution of rust and liberation of ferric ions (Fe ³⁺ ) from rust-containing or rust-generating substrate was performed using a ferric ion (Fe ³⁺ ) liberating agent.

As shown herein, rust, when treated with a ferric ion (Fe ³⁺ ) liberating agent (e.g., an acid), can be an ideal source of Fe ³⁺ ions affording an oxidation potential of 0.77 V for oxidizing thiophene-based moieties and producing conducting polymers.

As described herein, the Fe ³⁺ ions can be liberated from the rust- containing or rust-generating substrate by placing the substrate in contact with a Fe ³⁺ liberating agent (e.g., an acid solution, such as an HCI solution). For example, the Fe ³⁺ liberating agent can be hydrochloric acid (HCI), hydrofluoric acid (HF), hydrobromic acid (HBr), hydroiodic acid (HI), hypochlorous acid (HCIO), chlorous acid (HCIO ₂ ), perchloric acid (HCIO ₄ ), halogen oxoacids, hypofluorous acid (HFO), sulfuric acid (H ₂ SO ₄ ), fluorosulfuric acid (HSO ₃ F), nitric acid (HNO ₃ ), phosphoric acid (H ₃ PO ₄ ), fluoroantimonic acid (HSbFe), fluoroboric acid (HBF ₄ ), hexafluorophosphoric acid (HPFe), chromic acid (^CrCU), boric acid (H ₃ BO ₃ ), mesylic acid (CH ₃ SO ₃ H), esylic acid (CH ₃ CH ₂ SO ₃ H), besylic acid (C ₆ H ₅ SO ₃ H), tosylic acid (CH ₃ C ₆ H ₄ SO ₃ H), triflic acid (CF ₃ SO ₃ H), sulfonate polystyrene ([CH ₂ CH(C ₆ H ₄ )SO ₃ H] _n ), acetic acid (CH ₃ COOH), citric acid (C ₆ H ₈ 0 ₇ ), formic acid (HCOOH), gluconic acid (HQCH ₂ -(CHOH) ₄ -COQH), lactic acid (CH ₃ -CHOH-COOH), oxalic acid (HOOC-COOH), tartaric acid (HOOC- CHOH-CHGH-CGOH), fluoroacetic acid, trifluoroacetic acid, chloroacetic acid, dichloroacetic acid, trichloroacetic acid, or ascorbic acid.

Dissolution can be a process where a solute dissolves or substantially or partially dissolves in a solvent to form a solution, whereas solubility is the outcome of dissolution. As described herein, rust undergoes dissolution when placed into contact with a Fe ³⁺ liberating agent (e.g., an acid). Dissolution of rust liberates Fe ³⁺ ions from the rust. The Fe ³⁺ ions are then available to oxidize thiophene-based moieties and produce conducting polymers.

CONDUCTING POLYMER

The universal RVPP and AVP methods, as described herein, can be used to synthesize a variety of polymer nanofibers. The polymer nanofibers can comprise PEDOT, thiophene derivatives, or pyrrole derivatives. As an example, the polymer can be a conducting polymer. The conducting polymer can comprise poly(3,4-ethylenedioxythiophene) (PEDOT). Poly(3,4-ethylenedioxythiophene), “PEDOT”, is a heteroaromatic organic electronic polymer that is chemically and physically stable and has high electrical conductivity compared to the current state of the art materials. Similar to PEDOT, polypyrrole or poly(3-thiophenemethanol) are also suitable polymers that can be used in the disclosed RVPP and AVP methods.

Other examples of polymers (or monomers that can be polymerized into polymers) that can be used in the disclosed RVPP and AVP methods can be selected from one or more of the following: 1. Polythiophene and derivatives:

2. Polypyrrole and derivatives with backbone of:

H , synthesized from a monomer including, but not limited to: polymer comprising one or more of the following monomer:

4. Polyaniline and derivatives:

5. Polyphenylene and derivatives: polyanthracene polydopamine

As another example, the polymer can be a high conducting polymer. For example, a high conductivity polymer can have an electrical conductivity of greater than about 100 S cm ^-1 or about 300 S cm ^-1 or more. As another example, the polymer can have high electrochemical stability. For example, the polymer can have an electrochemical stability of 38,000 cycles or more or about 300,000 cycles to about 500,000 cycles or more.

Conducting polymers, such as PEDOT, can be used in OLEDs, solar cells, and as an electrostatic discharge coating for film. PEDOT is also used as an electrode material for supercapacitors. A supercapacitor is a high-capacity capacitor which can be used to supplement or replace a battery system.

PEDOT is a particularly useful heteroaromatic organic electronic polymer, possessing excellent chemical and physical stability as well as high electrical conductivity. This polymer is typically synthesized via a solution phase reaction, electrochemical oxidation, or vapor-phase polymerization. Among these, vapor- phase polymerization is a particularly promising strategy that results in conformal coatings of low electrical resistance in a single step. Polymerization from the vapor phase previously generally required a ferric-ion-containing salt serving as the oxidizing agent. Some ferric-ion-containing salts are quite expensive, and all ferric-ion containing salts are corrosive requiring safety precautions during handling. Moreover, Fe ³⁺ salts are hygroscopic and chemically unstable undergoing hydrolysis over time which challenges the reproducibility of experiments.

PEDOT can be synthesized via a solution phase reaction, electrochemical oxidation, or vapor-phase polymerization. Vapor-phase polymerization is a single step method that provides conformal coatings with low electrical resistance. Vapor-phase polymerization usually required a ferric-ion-containing salt to serve as the oxidizing agent. Ferric-ion-containing salts are corrosive, requiring safety precautions during handling, and may also be prohibitively expensive. Moreover, these salts are hygroscopic and chemically unstable.

The present invention provides for compositions and methods for vapor- phase polymerization of PEDOT nanofibers utilizing solid-state rust (or a rust layer) as an oxidizing agent, avoiding the need for expensive or corrosive salts.

Nanofibers

As described herein, PEDOT nanofibers were synthesized using vapor phase polymerization and rust as an oxidizing agent. This universal RVPP and AVP methods can also be used to synthesize a variety of polymer nanofibers including thiophene derivatives and pyrrole derivatives. A nanofiber can be a solid fiber with a diameter of less than 100 nm and a length exceeding the diameter by orders of magnitude.

As described herein, PEDOT nanofibers with a diameter of 70 nm and lengths of up to 100 mm were synthesized.

As described herein, similar nanofibrillar architecture was produced using different polymers, but the aspect ratio differs between polymers (see e.g., Example 3).

The present disclosure provides for longer fiber lengths (long conjugation lengths) and differences in aspect ratios than previous fibers. Known PEDOT materials comprise two structures, the fiber layer (e.g., mat or film) and a substrate. Generally, a mat can be a random dispersion of fibers or fibrous structure, high porosity, large active area and a film can be thinner than a fiber mat with less voids than a fiber mat. Generally, the thickness of film is below 20 mm, mat is above 20 mm (see e.g., FIG. 46).

Here is shown the production of fiber layers comprising fibers that are vertically directed vs. horizontally directed fibers (see e.g., Example 5, FIG. 47).

Aspect ratio

As described herein, high-aspect-ratio PEDOT nanofibers were synthesized using vapor phase polymerization and rust as an oxidizing agent. Aspect ratio is the ratio between nanofiber width and height. It is well known in the art that a high aspect ratio (e.g., aspect ratio of at least 100) is a defining structural characteristic of nanofibers and is important for their functionality.

As described herein, the synthesized PEDOT nanofibers were confirmed to have an aspect ratio of at least 1000 with high-angle annular dark-field images collected via scanning transmission microcopy.

Freestanding nanofibrillar layers

As described herein, high-aspect ratio PEDOT nanofibers can be deposited onto a substrate with high packing density to form a nanofibrillar PEDOT layer or film.

Described herein is a method that enables formation of freestanding nanofibrillar PEDOT layers that readily delaminate from a substrate. For example, the substrate can be simply immersed in water to delaminate the nanofibrillar PEDOT layer. The nanofibrillar PEDOT layer can be covalently bound to the substrate to allow for ease of delamination.

As described herein, the thickness of the nanofibrillar PEDOT layer can have a tunable thickness or can be varied by varying the thickness of the rust layer on the substrate. For example, a rust layer of 20 mm can be used to generate a nanofibrillar PEDOT film of 6 mm. As another example, a rust layer of 40 mm can be used to generate a nanofibrillar PEDOT film of 8 mm. As another example, a rust layer of 70 mm can be used to generate a nanofibrillar PEDOT film of 10 mm.

As described herein, freestanding nanofibrillar PEDOT layers can be used to develop energy storage applications.

For example, the freestanding nanofibrillar PEDOT layer can be used in a supercapacitor. As described here, a nanofibrillar PEDOT layer can have an electrical conductivity of at least 323 S cm ^-1 . As described herein, a nanofibrillar PEDOT-based supercapacitor can have a gravimetric capacitance of at least 181 F g ^-1 at a current density of 3.5 A g ^-1 and ability to retain 80% of capacitance after 38,000 charge-discharge cycles.

Oxidation potential of rust

As described herein, the rust substrate, optionally treated with an acid, can have an oxidation potential sufficient for oxidizing thiophene-based moieties and producing conducting polymers characterized by a long conjugation length, wherein the length is sufficient to result in bulk conductivity higher than 100 S/cm. Here, it was possible to qualitatively analyze conjugation via FT-IR to determine the presence of peaks and to analyze the conjugation length, indirectly, by correlating it to electrical conductivity.

As an example, rust, when treated with an acid, is an ideal source of Fe ³⁺ ions affording an oxidation potential between about 0.5 V and about 3 V. In a preferred embodiment, the oxidation potential is between about 0.5 V and 1 V or about 0.77 V.

ONE-POT, SINGLE STEP METHOD OF PRODUCING PEDOT

As disclosed herein, conductive poly(3,4-ethylenedioxythiophene) (PEDOT) nanofibrillar layers can be synthesized using a single step method inside a single sealed hydrothermal reactor (e.g., one-pot, single step method).

Rust-based vapor-phase polymerization

Described herein is a method that enables the deposition of freestanding nanofibrillar layers of a conducting polymer (e.g., poly(3,4- ethylenedioxythiophene) (PEDOT)) via rust-based vapor-phase polymerization (RVPP).

RVPP can take place in a single step inside a sealed hydrothermal reactor. RVPP can comprise the use of EDOT monomer vapor placed in contact with a solid-state rust coating (or rust layer) undergoing dissolution. The approaches described herein can be scalable using only a substrate (e.g., rusted steel surface), an iron liberating agent (e.g., acid vapor), and EDOT monomer vapor. For example, a rusted steel surface can be introduced into a reactor with about 20 mL concentrated hydrochloric acid (or any other acid capable of liberating Fe ³⁺ from the rust) and about 200 mL of a 0.0034 M EDOT solution in chlorobenzene (or any composition capable of facilitating a vapor phase monomer, such as any volatile organic compound (VOC)). The reactor can be sealed and heated to about 150 °C for about 6 hours, then cooled in an ice bath for about 15 minutes.

RVPP can comprise the use of an iron liberating agent (e.g., hydrochloric acid vapor) to dissolve rust and liberate aquated Fe ³⁺ ions. The liberated Fe ³⁺ ions can oxidize EDOT monomer vapor and initiate polymerization when placed in contact with EDOT monomer vapor (oxidative radical polymerization). The liberation of Fe ³⁺ ions can occur in unison with polymerization, resulting in a conformal coating of inorganic-organic core-shell FeCl ₃ -PEDOT nanofibers. The polymerization can consume rust, which can result in nanofibrillar PEDOT layers that delaminate from a substrate (e.g., steel).

RVPP can enable synthesis of freestanding nanofibrillar PEDOT layers that delaminate from a substrate (e.g., steel) and have high electrochemical stability and conductivity that can be used for energy storage applications. For example, PEDOT layers synthesized by RVPP can be engineered into supercapacitors resulting in devices with capacitance of at least 181 F g- ¹ , current density of at least 3.5 A g ^-1 , and ability to retain 80% of capacitance after 38,000 charge-discharge cycles.

RVPP can enable control of nanofibrillar PEDOT layer patterning and thickness. For example, a rusted surface can be masked with polyimide tape to selectively deposit a PEDOT nanofibrillar layer into a pattern.

RVPP can enable growth of nanofibrillar PEDOT layers on a rust- generating or rust-containing substrate of any dimension. For example, a rusted screw can be used as a substrate for RVPP.

RVPP can be used to rid a surface of unwanted rust. For example, a rusted screw that undergoes RVPP is clean and devoid of rust upon delamination of the nanofibrillar PEDOT layer.

Oxidative radical polymerization

As described herein, the RVPP method of producing PEDOT can comprise an oxidative radical polymerization step. The oxidative radical polymerization step can comprise an Fe ³⁺ liberating agent (e.g., an acid) coming into contact with rust, dissolving the rust, and liberating Fe ³⁺ ions. The Fe ³⁺ ions can then reduce EDOT monomer vapor to EDOT ^+‘ radical cations. A conjugated PEDOT backbone can be assembled via step growth as radical coupling is concomitant with deprotonation and oligomer formation. The conjugated PEDOT backbone can be doped by ions from the Fe ³⁺ liberating agent to form an electronically conductive form of PEDOT.

Definitions and methods described herein are provided to better define the present disclosure and to guide those of ordinary skill in the art in the practice of the present disclosure. Unless otherwise noted, terms are to be understood according to conventional usage by those of ordinary skill in the relevant art.

In some embodiments, numbers expressing quantities of ingredients, properties such as molecular weight, reaction conditions, and so forth, used to describe and claim certain embodiments of the present disclosure are to be understood as being modified in some instances by the term “about.” In some embodiments, the term “about” is used to indicate that a value includes the standard deviation of the mean for the device or method being employed to determine the value. In some embodiments, the numerical parameters set forth in the written description and attached claims are approximations that can vary depending upon the desired properties sought to be obtained by a particular embodiment. In some embodiments, the numerical parameters should be construed in light of the number of reported significant digits and by applying ordinary rounding techniques. Notwithstanding that the numerical ranges and parameters setting forth the broad scope of some embodiments of the present disclosure are approximations, the numerical values set forth in the specific examples are reported as precisely as practicable. The numerical values presented in some embodiments of the present disclosure may contain certain errors necessarily resulting from the standard deviation found in their respective testing measurements. The recitation of ranges of values herein is merely intended to serve as a shorthand method of referring individually to each separate value falling within the range. Unless otherwise indicated herein, each individual value is incorporated into the specification as if it were individually recited herein. The recitation of discrete values is understood to include ranges between each value.

In some embodiments, the terms “a” and “an” and “the” and similar references used in the context of describing a particular embodiment (especially in the context of certain of the following claims) can be construed to cover both the singular and the plural, unless specifically noted otherwise. In some embodiments, the term “or” as used herein, including the claims, is used to mean “and/or” unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive.

The terms “comprise,” “have” and “include” are open-ended linking verbs. Any forms or tenses of one or more of these verbs, such as “comprises,” “comprising,” “has,” “having,” “includes” and “including,” are also open-ended.

For example, any method that “comprises,” “has” or “includes” one or more steps is not limited to possessing only those one or more steps and can also cover other unlisted steps. Similarly, any composition or device that “comprises,” “has” or “includes” one or more features is not limited to possessing only those one or more features and can cover other unlisted features.

All methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The use of any and all examples, or exemplary language (e.g., “such as”) provided with respect to certain embodiments herein is intended merely to better illuminate the present disclosure and does not pose a limitation on the scope of the present disclosure otherwise claimed. No language in the specification should be construed as indicating any non-claimed element essential to the practice of the present disclosure.

Groupings of alternative elements or embodiments of the present disclosure disclosed herein are not to be construed as limitations. Each group member can be referred to and claimed individually or in any combination with other members of the group or other elements found herein. One or more members of a group can be included in, or deleted from, a group for reasons of convenience or patentability. When any such inclusion or deletion occurs, the specification is herein deemed to contain the group as modified thus fulfilling the written description of all Markush groups used in the appended claims.

All publications, patents, patent applications, and other references cited in this application are incorporated herein by reference in their entirety for all purposes to the same extent as if each individual publication, patent, patent application or other reference was specifically and individually indicated to be incorporated by reference in its entirety for all purposes. Citation of a reference herein shall not be construed as an admission that such is prior art to the present disclosure.

Having described the present disclosure in detail, it will be apparent that modifications, variations, and equivalent embodiments are possible without departing the scope of the present disclosure defined in the appended claims. Furthermore, it should be appreciated that all examples in the present disclosure are provided as non-limiting examples. EXAMPLES

The following non-limiting examples are provided to further illustrate the present disclosure. It should be appreciated by those of skill in the art that the techniques disclosed in the examples that follow represent approaches the inventors have found function well in the practice of the present disclosure, and thus can be considered to constitute examples of modes for its practice.

However, those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments that are disclosed and still obtain a like or similar result without departing from the spirit and scope of the present disclosure.

EXAMPLE 1: CONVERTING RUST TO PEDOT NANOFIBERS FOR

SUPERCAPA CITORS

The following example describes the conversion of rust to PEDOT nanofibers for use in supercapacitors.

Iron corrosion, a product from the chemical reaction between iron and oxygen in the presence of water and commonly referred to as rust, is a heterogeneous solid-state material composed of multiple phases that represent an abundant source of chemical waste. Here is introduced a strategy that advances the state-of-the-art in chemical synthesis by demonstrating the usefulness of this ubiquitous inexpensive inorganic material for developing oxidative radical polymerizations. Rust, when treated with an acid, is an ideal source of Fe ³⁺ ions affording an oxidation potential of, ideally between about 0.5 V and 3 V or about 0.77 V for oxidizing thiophene-based moieties and producing conducting polymers characterized by long conjugation lengths, wherein a long conjugation length is sufficient to result in bulk conductivity higher than 100 S/cm. As show herein, the inventors developed a novel mechanism that enables the deposition of freestanding nanofibrillar films of the conducting polymer poly(3,4-ethylenedioxythiophene) (PEDOT) via rust-based vapor-phase polymerization (RVPP). The process takes place in a single step inside a sealed hydrothermal reactor when monomer vapor makes contact with a solid-state rust coating (or rust layer) undergoing dissolution. This approach is scalable using only a rusted steel surface, acid vapor, and monomer vapor. Freestanding nanofibrillar PEDOT films delaminate from a steel substrate characterized by an electronic conductivity of 323 S cm ^-1 and high electrochemical stability; RVPP enables patterning of a film in situ during synthesis. RVPP-PEDOT films are engineered into supercapacitors resulting in devices that exhibit a state-of-the-art capacitance of 181 F g- ¹ at a current density of 3.5 A g ^-1 and retain 80% of their original capacitance after 38,000 cycles.

Presented herein is a robust platform for synthesizing PEDOT nanofibers utilizing solid-state rust or rust layer as a reactant and rust-based vapor-phase polymerization (RVPP). This universal RVPP method can also be used to synthesize a variety of polymer nanofibers such as thiophene derivatives and pyrrole derivatives. RVPP, introduced here, is a versatile alternative to traditional vapor-phase polymerization. This approach obviates the need of a corrosive salt, affords a nonabsorbent solid-state oxidant source for engineering facile reactions, and provides a sustainable approach for the synthesis of organic electronics using what is typically considered chemical waste. The protocol for producing a nanofibrillar PEDOT film uses hydrochloric acid vapor to dissolve rust and liberate aquated Fe ³⁺ ions that initiate polymerization upon contact with monomer vapor. Hydrolysis of Fe ³⁺ ions occurs in unison with polymerization resulting in the precipitation of one-dimensional inorganic nanostructured colloidal iron species such as FeOOH. The approach to synthesis herein is scalable, producing nanostructured polymer film growth on any rusted surface of any dimension using only a sealed vessel for reactant vapors to interact with solid-state rust or rust layer. Dissolution of rust and hydrolysis of Fe ³⁺ ions occur in situ resulting in a conformal coating of inorganic-organic core-shell FeCl ₃ - PEDOT nanofibers. A rust layer on a steel surface plays the role of a versatile solid-state 2D reactant that enables control of polymer patterning and film thickness. Polymerization consumes rust resulting in nanofibrillar PEDOT films that delaminate from the steel. This synthetic strategy results in freestanding nanofibrillar PEDOT films that are ideal for developing energy storage applications possessing an electronic conductivity of 323 S cm ^-1 . The nanofibrillar PEDOT -based supercapacitors exhibit a state-of-the-art performance with gravimetric capacitance of 181 F g ^-1 at a current density of 3.5 A g ^-1 , retaining 80% of the original capacitance after 38,000 charge-discharge cycles.

Results and Discussion

Rust-based vapor-phase polymerization (RVPP) is a synthetic strategy that utilizes rust (see e.g., FIG. 1A-FIG. 1C) to produce nanofibrillar freestanding films of the conducting polymer PEDOT (see e.g., FIG. 1 D-FIG. 1 F). Rusted steel serves as substrate and is produced by immersing a steel sheet in a 0.01 M sulfuric acid aqueous solution for 48 h (see e.g., FIG. 8); the rust layer possesses a microstructure composed of two-dimensional platelets and acicular one-dimensional spherulites (see e.g., FIG. 1C). Powder X-ray diffraction patterns indicate that these microstructures are a-FeOOH, g-FeOOH, and FeSO ₄ phases (see e.g., FIG. 9). A nanofibrillar PEDOT film (see e.g., FIG. 1 F) is vapor-phase-deposited using a 7 mm x 7 mm rusted steel sheet that is introduced into a sealed vessel along with 20 mL of concentrated hydrochloric acid and 200 mL of a 0.0034 M EDOT solution in chlorobenzene (see e.g., FIG. 1 G). The reactor is sealed, placed in an oven (150 °C) for 6 h, and subsequently cooled in an ice bath for 15 min resulting in a blue nanofibrillar PEDOT film that readily delaminates from the steel sheet. Initially, as temperature is ramped, hydrochloric acid vapor diffuses (see e.g., FIG. 1 H) dissolving rust and liberating Fe ³⁺ ions that oxidize monomer vapor upon contact. During this oxidative radical polymerization, ferric ions are reduced to ferrous ions (Fe ²⁺ ), and the monomer 3,4-ethylenedioxythiophene (EDOT) is oxidized to EDOT+* radical cations (see e.g., FIG. 10). The conjugated PEDOT backbone assembles via step growth as radical coupling is concomitant with deprotonation and oligomer formation; protons produced lower the pH of a reaction. Finally, the conjugated polymer chain is doped by Cl- ions present from hydrochloric acid resulting in an electronically conductive form of PEDOT.

Mechanistic studies of nanofiber evolution are carried out using microscopy and time-lapse spectroscopy by quenching reactions. Initially, HCI vapor condensation and rust dissolution lead to the collapse of microstructured rust (see e.g., FIG. 2A) as proposed in the schematic diagram (see e.g., FIG.

2B). After 2 h of reaction, a porous PEDOT skin (see e.g., FIG. 2C) forms when aquated Fe ³⁺ ions oxidize EDOT monomer vapor; this reaction reduces Fe ³⁺ ions and produces the Fe ²⁺ -ion-containing salt FeCl ₃ detected via powder X-ray diffraction (PXRD) (see e.g., FIG. 11). As the reaction progresses, HCI vapor permeates and diffuses throughout bulk rust increasing the concentration of Fe ³⁺ , Cl-, and FT aquated ions; dissolution of rust with HCI vapor promotes hydrolysis-driven precipitation of one-dimensional FeOOH nuclei detected via PXRD (see e.g., FIG. 11). Based on previous work, these nuclei are responsible for nanofibrillar formation; however, at this early stage of polymerization, nuclei redissolve under the low-pH environment (see e.g., FIG. 2D). Upon reaching 3 h of synthesis, the entire rusted surface is coated with a dense PEDOT film visible via scanning electron microscopy (see e.g., FIG. 12A). As the reaction continues, HCI vapor is consumed, and pH increases thus facilitating hydrolysis and the formation of FeOOH nuclei (see e.g., FIG. 11). These inorganic nuclei are unstable and undergo Ostwald ripening by coalescing into more energetically favored larger one-dimensional structures. A nanofibrillar morphology is initially observed after 4 h of synthesis and is characterized by low-aspect-ratio PEDOT microfibers that pierce out of the polymer skin (see e.g., FIG. 2E, FIG. 12B). A one-dimensional PEDOT structure possesses a FeOOH inner scaffold identified by PXRD (see e.g., FIG. 11), and as polymerization continues, diffusion of HCI vapor liberates Fe ³⁺ ions from this FeOOH inner scaffold promoting polymerization radially inward (see e.g., FIG. 2F). Vertical growth of a PEDOT- coated FeOOH nanofiber is driven by hydrolysis, Ostwald ripening, and polymerization. Moreover, one-dimensional FeOOH ribbons undergo crystal splitting along the crystal lattice (see e.g., FIG. 12C) resulting in extended growth of one-dimensional nanofibers (see e.g., FIG. 12D). Ferric ions in FeOOH scaffolds are reduced to ferrous ions during oxidative radical polymerization forming a FeCl2 core. This reaction comes to completion after 6 h, resulting in a freestanding polymer film characterized by a high packing density of nanofibers (see e.g., FIG. 2E) with a core-shell FeCl ₃ -PEDOT architecture (see e.g., FIG. 2H).

The conjugation length of the polymer is studied via Fourier-transform infrared spectroscopy (see e.g., FIG. 13) and shows the characteristic C=C stretch from the doped quinoid PEDOT structure at 1510 cm ^-1 and the C — O — C vibration peak at 965 cm ^-1 from the ethylenedioxy group. Oxidative doping converts benzoid to quinoid leaving the ethylenedioxy group unaltered; this enables characterization of conjugation length by their relative ratios. The quinoid to ethylenedioxy group ratio is maximized in the 6 h synthesis, resulting in PEDOT nanofibers of long conjugation length.

The mechanism is tested by obviating HCI vapor from synthesis; this results in suppression of polymerization as indicated by powder X-ray diffraction and electron microscopy (see e.g., FIG. 14). HCI is necessary for rust dissolution and liberation of Fe ³⁺ ions and provides a source of water for hydrolysis. A synthesis without EDOT leads to the collapse of the rust layer and a product void of polymer (see e.g., FIG. 15). Nanofibrillar growth uses HCI vapor, EDOT monomer, and chlorobenzene as a carrier gas; this organic polar solvent affords a universal surfactant-like solvent for the assembly of oligomers and facile polymer formation (see e.g., FIG. 16). Benzene, dichloromethane, nitromethane, and 1 -butanol also generate nanofibers albeit of low electronic conductivity. FIG. 17 shows the granular morphology on the underside of a PEDOT film while a rip demonstrates that nanofibers make up the bulk of a cross-section of the film.

Rust-based vapor-phase polymerization is scalable due to rust’s chemical stability and the ease with which an iron-containing substrate corrodes; the thickness of the rust layer on a substrate is controlled to produce a freestanding PEDOT film that readily delaminates. FIG. 3A shows a digital picture of a 10 mm x 50 mm delaminated PEDOT film next to its original supporting steel sheet; the thickness of this polymer film is controlled by the thickness of the pre-deposited rust layer. The thickness of a rust layer is optimized by immersing a steel substrate in a 0.01 M H ₂ SO ₄ corroding aqueous solution for 24, 40, and 48 h leading to 20, 40, and 70 mm rust-layer thickness, respectively (see e.g., FIG. 3B-FIG. 3D). These corroded substrates are utilized for RVPP resulting in PEDOT films with thicknesses of 6, 8, and 10 mm, respectively (see e.g., FIG. 3E-FIG. 3G). A 10 mm thick PEDOT film, produced using a heavily corroded rust layer, possesses a sheet resistance of 2.85 W sq ^-1 and a conductivity of 323 S cm ^-1 . When a substrate is immersed in a corroding solution for less than 48 h, the rust layer is composed of a porous architecture that leads to higher sheet resistance (see e.g., FIG. 18). Patterning of a PEDOT film is carried out by masking a rusted surface with polyimide tape resulting in selective deposition of an electronically conductive blue nanofibrillar PEDOT film (see e.g., FIG. 4A-FIG. 4C). In RVPP, a rusted three-dimensional substrate is also an ideal source of Fe ³⁺ ions; for example, a rusted screw was utilized as a substrate (see e.g., FIG. 4D and FIG. 4E). Note that the rust layer on this screw is from natural exposure to environmental moisture and air. A PEDOT coating deposits conformally on a screw via RVPP and delaminates completely by immersion in water leaving the screw surface clean, lustrous, and void of rust (see e.g., FIG. 4F). To ascertain the extent of cleaning attained using this protocol, another rusted screw is cleaned using a commercial rust-removing solution, and the surface roughness is characterized before and after via profilometry and scanning electron microscopy. Surface roughness is then compared to that of a cleaned screw after PEDOT delamination (see e.g., FIG. 19). The surface roughness of a 2D steel substrate treated with a rust-removing solution exhibits a surface roughness with a 600 nm amplitude in variance after cleaning. When a rusted steel substrate is utilized for RVPP, and rust is removed, its surface roughness exhibits a 400 nm amplitude variance; a steel surface cleaned via RVPP is smoother than steel cleaned by a commercial rust-removing solution. A steel substrate utilized in the synthesis is pre-polished, and after synthesis, the surface roughness is composed of aggregates of iron as determined by powder X-ray diffraction and scanning electron microscopy (see e.g., FIG. 20).

A PEDOT film readily delaminates from its substrate by simple immersion in water resulting in a freestanding polymer film; this is because a rust layer lacks chemical coordination to steel (see e.g., FIG. 5A). A freestanding RVPP- PEDOT film is characterized by a high packing density of high-aspect-ratio core- shell nanofibers as confirmed by high-angle annular darkfield (HAADF) images collected via scanning transmission electron microscopy (see e.g., FIG. 5B). The contrast in these images is roughly proportional to the square of the atomic number enabling characterization of elemental composition. These nanofibers have a 20 nm diameter core, 50 nm thick shell (see e.g., FIG. 21), and lengths up to 100 mm reaching an aspect ratio of 1000 (see e.g., FIG. 22). Energy- dispersive X-ray spectroscopy maps show that a PEDOT film possesses an elemental distribution composed of S and Fe (see e.g., FIG. 5C) pertaining to a polymer shell and an inorganic core, respectively. Powder X-ray diffraction confirms the existence of a FeCl ₃ core (see e.g., FIG. 5D); purification in 6 M HCI removes this inorganic core resulting in a hollow nanofiber (see e.g., FIG. 23).

The molecular structure of purified RVPP-PEDOT, characterized by Fourier-transform infrared spectroscopy (see e.g., FIG. 5E), shows an oxidized doped form of PEDOT. The bands at 1510 and 1312 cm ^-1 correspond to aromatic C=C asymmetric stretches in the polythiophene ring and inter-ring C — C stretches, respectively. Bands at 1180, 1135, and 1085 cm ^-1 are assigned to C — O — C stretches in the ethylenedioxy ring whereas the C — S — C stretches in the thiophene ring appear at 962, 903, and 743 cm ^-1 .

Ultraviolet-visible-near-infrared absorption spectroscopy aids in probing doping levels, and typically, neutral PEDOT produces a broad absorption peak in the visible region (400-600 nm) corresponding to the p-p * transition. The broad absorption above 800 nm corresponds to free charge carriers and a conductive PEDOT state. Post-synthetic doping of RVPP-PEDOT using HCI vapor introduces Cl- ions that increase the doping level and conductivity of the polymer (see e.g., FIG. 5F). This post-synthetic doping is detected as a p -p * transition that stifles absorption between 400 and 600 nm and leads to a broad absorption hump between 800 and 1000 nm due to the generation of polaronic and bipolaronic states. To understand charge transport as a function of a material’s structure, a polymer film is examined via powder X-ray diffraction revealing three sharp peaks at 2Q = 6.5°, 13.0°, and 26.5° (see e.g., FIG. 5G) with peak widths at half height of 0.8, 1.1 , and 2.7 characteristic of crystalline PEDOT. The broad diffraction peak at 2Q = 26.5° due to p -p stacking is assigned to the (020) reflection with an interchain spacing distance of -0.4 nm; a sharp peak for (100) at 2Q = 6.5° and its second-order reflection (200) at 2Q = 13.0° correspond to lateral chain packing. Lamellar stacking (100) exhibits a distance of 1.3 nm. The selective area electron diffraction pattern of PEDOT shows discrete diffraction spots indicating a preferred polycrystalline orientation (see e.g., FIG. 24). Charge transport studied via two-point probe l-V measurements shows linear ohmic behavior for HCI-treated PEDOT and a low electrical resistance (see e.g., FIG. 5H); treating a PEDOT film with ammonium hydroxide de-dopes the polymer resulting in a more negative l-V curve slope. The four-point probe sheet resistance of PEDOT (2.85 W sq ^-1 ) indicates that its electrical conductivity is approximately 323 S cm ^-1 (see e.g., FIG. 26) calculated from sheet resistivity. This high electrical conductivity is a function of an ordered crystal structure and a high charge carrier concentration.

The electrochemical properties of a freestanding PEDOT film are studied utilizing a current-collector-less three-electrode configuration, platinum lead, and a polyimide mask that exposes the polymer to the electrolyte (see e.g., FIG. 6A, inset). Cyclic voltammograms demonstrate a high electrochemical stability with nearly rectangular shapes after 500 cycles in 1 M H2SO4 at a scan rate of 25 mV s ^-1 (see e.g., FIG. 6A, inset). Notably, cyclic voltammograms retain their rectangular shape as the scan rate is increased from 25 to 200 mV s ^-1 ; this is due to a low electrode resistance and the polymer’s high electronic conductivity (see e.g., FIG. 6B). Nyquist plots (see e.g., FIG. 6C), collected via electrochemical impedance spectroscopy, are carried out at the open circuit potential with a 10 mV sinusoidal perturbation frequency ranging from 0.1 Hz to 100 kHz and show real impedance Z" versus imaginary impedance Z'' A 1.23 W internal resistance for the cell is obtained from the x-axis intercept at high frequency; this represents the aggregated resistances from the electrode material, electrode/electrolyte interface, and electrolyte.42 Low charge-transfer resistance is determined from the nearly 90° angled curve in the low-frequency region of the plot and by curve shape retention after 500 cycles. The low internal resistance enables reversible facile doping during charging and discharging. PEDOT-RVPP exhibits a state-of-the-art capacitance of 181 F g ^-1 calculated from a discharge curve (see e.g., FIG. 6D) at a current density of 3.5 A g ^-1 ; moreover, triangular-shaped curves at various current densities indicate facile charge transfer. Cyclic voltammetry (see e.g., FIG. 6E) and electrochemical impedance spectroscopy (see e.g., FIG. 6F) experiments are carried out using 1 and 6 M H2SO4 electrolytes to test the doping effect of the SO4 ^2- counteranion — a commonly used dopant for increasing the conductivity of PEDOT. The Nyquist plots in FIG. 6F show a large semicircle diameter for 6 versus 1 M electrolyte stemming from interfacial resistance, poor charge propagation, and a low ionic mobility. Supercapacitors are fabricated using RVPP-PEDOT as shown in the flow process diagram of FIG. 7A; these are developed utilizing a 1 M H2SO4 aqueous electrolyte and a hard carbon paper current collector for stable device performance and rapid charging-discharging. A device is encased in polyimide tape to mitigate electrolyte evaporation and retains rectangular-shaped cyclic voltammograms at 25 mV s ^-1 in voltage windows of 0.6, 0.8, 1 , and 1 .2 V (see e.g., FIG. 7B). After 1000 cycles at a specific scan rate, ranging from 25 to 4000 mV s ^-1 (see e.g., FIG. 7C and FIG. 7D), electrochemical performance and capacitive behavior remain stable due to a high rate capability. FIG. 7E shows that, as the scan rate increases, the capacitance decreases from 185 F g- ¹ (25 mV s ^-1 ) to 148 F g ^-1 (4000 mV s ^-1 ) because of the limited time for electrolyte ions to diffuse to the electrode surface. Galvanostatic charge-discharge curves show nearly linear and triangular symmetry (see e.g., FIG. 7F) stemming from a low internal resistance and high Coulombic efficiency. The highest capacitance attained is 181 F g ^-1 at a current density of 3.5 A g ^-1 with 98.3% Coulombic efficiency. FIG. 7G shows a plot of the ohmic drop versus Coulombic efficiency at different current densities; this ohmic drop (DII) = IR stems from stifled ion diffusion due to a time constraint. Throughout this test, resistance (R) is held constant, and DII is proportional to current density (I). The ohmic drop increases with current density as ion diffusion becomes more limited; however, a small initial ohmic drop present at a low current density is due to self-discharge (see e.g., FIG. 26). A supercapacitor possesses a self-discharge rate and a charging rate (current density); the effective charging rate is the difference between the set current density and the intrinsic self-discharging rate of a device. Galvanostatic charge-discharge curves show Coulombic efficiencies of 99.1 %, 98.3%, and 96.5% as a device is charged to 0.6, 0.8, and 1 V, respectively, at a current density of 3.5 A g ^-1 (see e.g., FIG. 7H). The Nyquist plot in FIG. 27 shows an equivalent series resistance of 0.43 W and an approximate 90° angle at low frequencies. This low aggregated device resistance is characteristic of an electronically conducting nanostructured electrode possessing a large electrode/electrolyte contact area that shortens ion diffusion pathways. A supercapacitor exhibits 80% retention of the original capacitance after 38,000 cycles (see e.g., FIG. 7I) buttressed by a nanofibrillar architecture that mitigates the stresses associated with polymer swelling and shrinking during long-term charge and discharge.

Conclusions

Rust is a ubiquitous byproduct from the corrosion of iron, and when treated with acid, this solid-state material serves as an ideal source of Fe ³⁺ ions for carrying out chemical synthesis. This work demonstrates the advantages of utilizing rust for developing oxidative radical polymerizations from the vapor phase thereby overcoming the hygroscopic and corrosive nature of iron(lll) salts that serve as nominal oxidants for synthesizing conducting polymers.

Discussions and results presented here address mechanistic steps associated with dissolution of rust during synthesis and introduce a fundamental understanding of a synergistic chemical process that combines polymer growth, hydrolysis, and crystallization. Rust-based vapor-phase polymerization is a novel approach for producing high-performing conducting polymers as proven by the attained high conductivity and high electrochemical stability of the robust freestanding nanofibrillar PEDOT films. In this approach, conducting polymer synthesis, vapor-phase chemical deposition, and energy storage is advanced through a facile, scalable, and patternable process that enables deposition of electroactive high-surface-area electrodes for fabricating state-of-the-art supercapacitors.

Materials and Methods

Materials

3,4-Ethylenedioxythiophene (EDOT, 97%), chlorobenzene (99%), hydrochloric acid (37%), sulfuric acid (98%), and methanol (>99.8%) were purchased from Sigma-Aldrich and used as received.

Corroding Protocol a-FeOOH, g-FeOOH, and FeSO ₄ rust layers were produced as per ASTM A109 protocol by immersing a 2.5 cm x 4 cm low-carbon steel sheet in 20 mL of a 0.01 M H ₂ SO ₄ solution for 48 h at 25 °C. Corroded sheets were then rinsed and dried under ambient conditions; the thickness of a rust layer is controlled by immersion time ranging between 24 h and 40 h. Rust-Based Vapor-Phase Polymerization of PEDOT

The Teflon liner of a hydrothermal reactor is loaded with a 7 mm x 7 mm rusted steel substrate, 20 mL of concentrated hydrochloric acid, and 200 mL of a 0.0674 M EDOT solution (3.37 x 10 ^-5 mol) in chlorobenzene (4.93 x 10 ^-3 mol). Each of these components is contained in a glass vial. This reactor is sealed, heated for 6 h in an oven at 150 °C, and then cooled in an ice bath for 15 min. A PEDOT-coated substrate is immersed in water to delaminate the polymer film.

Supercapacitor Fabrication

A wet freestanding RVPP-PEDOT film is placed on a hard carbon fiber paper current collector and vacuum-dried overnight to enhance adhesion. A platinum foil lead contacts the backside of the current collector, and a 25 mm thick Celgard 3501 separator is prewetted with a 1 M H2SO4 aqueous solution during assembly.

Morphology and Structure Characterization

Scanning electron micrographs and energy-dispersive X-ray spectrograms were collected using a JEOL 7001 LVF FE-SEM instrument. Transmission electron micrographs were obtained in a JEOL 2100 instrument by squeezing a PEDOT film in a folding double TEM grid. A Bruker d8 Advanced X- ray diffractometer was utilized to collect powder X-ray diffractograms of pulverized samples at room temperature, with Cu Ka radiation source (l =

1.5406 Å) and LynxEye XE detector, operating at 40 kV and 40 mA; the sample holder was rotated at 30 rmm with a scan step of 0.02°. Ultraviolet-visible-near- infrared absorption spectra were obtained on a Cary 5000 UV-vis-NIR spectrophotometer using pulverized samples dispersed in 6 M H ₂ SO ₄ . Samples were de-doped using concentrated NH4OH. Fourier-transform infrared spectra were collected on a Bruker ALPHA Platinum- ATR instrument. Current-voltage (I- V) curves were obtained using a built-in-house 3D printed probe station using two gold needles with 1 mm of separation.

Electrochemical Measurements

Cyclic voltammetry and electrochemical impedance spectroscopy were performed on a BioLogic VMP3 multipotentiostat. Three-electrode experiments utilized a platinum mesh lead, affixed to the back side of the working electrode that was fully covered with polyimide tape. Platinum mesh connected with a platinum wire served as counter electrode. The reference electrode (BASi Ag/AgCI RE-5B) is pointed directly at the working electrode to compensate and minimize solution resistance. A 5 mm x 5 mm PEDOT film serving as the working electrode was washed in 6 M HCI and methanol prior to a measurement. A 1 M H2SO4 aqueous electrolyte was employed using milli-Q water (18 MW) that was degassed for 15 min. Electrochemical impedance spectroscopy was carried out at the electrode’s open circuit potential after obtaining a reversible cyclic voltammogram. Impedance values were recorded using a 10 mV sinusoidal disturbance at frequencies ranging from 100 kHz to 100 mHz. EXAMPLE 2: RVPP GENERA TION OF PEDOT NANOFIBRILLAR FILMS AND THEIR USE IN

MICRO-SUPERCAPA CITORS

The following example describes how PEDOT nanofibers are generated from solid-state rust and their use in micro-supercapacitors.

Rust-based vapor-phase polymerization (RVPP) can be used to generate PEDOT nanofibers with an aspect ratio of at least 900 (see e.g., FIG. 28 and TABLE 1).

TABLE 1 : Diameter, length, and aspect ratio of the individually traced nanofibers depicted in FIG. 28.

Evaporative vapor-phase polymerization is conventionally carried out by placing a droplet of an oxidant aqueous solution (e.g., FeCl ₃ ) on a current collector and by heating to 130 °C. It has been previously shown that these corrosive conditions oxidize current collectors, such as nickel foam, aluminum foil, as well as steel mesh, and result in non-nanofibrillar bulk PEDOT. Therefore, it was previously thought that these substrates are not compatible with vapor- phase polymerization of PEDOT, whereas inorganic carbons such as amorphous particles, graphite, and carbon fibers would be considered more desirable substrates. As such, it was surprising that steel can serve as a substrate and rust can serve as an oxidant precursor for the generation of nanofibrillar PEDOT films (see e.g., FIG. 29).

A steel substrate alone subjected to vapor phase polymerization is not sufficient to generate PEDOT nanofibrillar films. However, subjecting a solid- state rust or a rust layer on the steel substrate to vapor phase polymerization is sufficient to generate PEDOT nanofibrillar films. The rust layer can be formed by exposing the steel substrate to an iron oxidizing agent, such as H ₂ SO ₄ , HCI, sea water, or Dl water. Rust formed by exposing a steel substrate to these agents are capable of being used as substrates for the generation of PEDOT nanofibers (see e.g., FIG. 30).

The PEDOT nanofibrillar films generated from RVPP can be used to build high energy-density micro-supercapacitors (see e.g., FIG. 31 and FIG. 32). The micro-supercapacitors are low-cost, as a 3D morphology electrode can be fabricated in situ without a pre-made scaffold. The PEDOT generated from RVPP is higher quality than that generated from electrochemical deposition. The PEDOT generated from RVPP has higher crystallinity and conductivity.

Higher crystallinity was determined by sharp XRD peaks, higher conductivity, and lower sheet resistance. Polycrystalline domains are prevalent when using electron microscopy to image lattice fringes associated with periodicity in the polymer structure (crystallinity). The XRD pattern for the disclosed PEDOT materials is unique (see e.g., Example 6, FIG. 48). The RVPP PEDOT shows higher conductivity and lower sheet resistance than many other nanostructured conducting polymers (see e.g., Example 6, TABLE 2).

The micro-supercapacitors exhibit state-of-the-art performance, including potential window capability (see e.g., FIG. 33), scan rate capability (see e.g.,

FIG. 34), and capacitance retention (see e.g., FIG. 35). The micro-supercapacitor exhibits an energy density of 4300 mWh/cm ³ , a power density of 48.7 W/cm ³ , an areal capacitance of 42.8-45.6 mF/cm ² , an areal energy density of 3.9-4.3 mJ/cm ² , a volumetric capacitance of 548.3 F/cm ³ , and a volumetric energy density of 48.7 mWh/cm ³ . A number of fractal designs can be incorporated into the micro-supercapacitor (see e.g., FIG. 36).

The PEDOT nanofibrillar films generated by RVPP can be used in transparent electrochromic windows.

EXAMPLE 3: PEDOT AND POLYPYRROLE SYNTHESIS ON IRON(III)-CONTAINING

SUBSTRATES

The following example provides data for polymer/oxide composites and polymer/carbon composites.

PEDOT has been successfully synthesized on SiO ₂ (glass); Fe ₂ O ₃ (red clay brick); b-FeOOH (core-shell fibers); TeO ₂ (core-shell fibers); SnO ₂ (core- shell fibers); hard carbon fiber paper; carbon cloth (see e.g., FIG. 38-FIG. 43).

Polypyrrole has been successfully synthesized on hard carbon fiber paper (see e.g., FIG. 44).

EXAMPLE 4: NANOFIBER ASPECT-RATIO DIFFERENCE BETWEEN POLYMERS

The following example describes the measurement of aspect ratios of different polymers. The aspect-ratio of PEDOT nanofiber was determined to be bout 1000, poly(3-thiophenemethanol) is about 6 and polypyrrole is about 150 (see e.g., FIG. 45).

EXAMPLE 5: SYNTHESIS OF BOTH HORIZONTALLY DIRECTED AND VERTICALLY

DIRECTED FIBER

The following example describes both horizontal and vertical directed fibers synthesized using the RVPP methods.

Conventionally, a horizontally directed nanostructure is deposited using an electrospun mat template that adds extra weight and volume as well as lowers the energy density and power density. Template-less electrospinning requires additive polymers that lead to a low electrical conductivity of less than 100 S/cm. The methods described herein synthesize both horizontally directed and vertically directed fibers by controlling nucleation concentration.

High nucleation concentration leads to vertically directed fibers and low nucleation concentration leads to horizontally directed fibers (see e.g., FIG. 47). EXAMPLE 6: CRYSTALLINITY p(RD) AND ELECTRICAL CONDUCTIVITY

The following example describes the data for the XRD diffraction patterns and electronic conductivity for other conducting polymers with a comparison to our peaks, enabling a more quantitative definition of the improved crystallinity.

Crystallinity The RVPP PEDOT shows similar peak sharpness compared to single crystal PEDOT nanowires, sharper than solution synthesized PEDOT, oxidative chemical vapor deposition (oCVD) synthesized PEDOT and PEDOT:PSS, suggesting an enhanced crystallinity (see e.g., FIG. 48).

Conductivity Our RVPP PEDOT shows higher conductivity and lower sheet resistance than many other nanostructured conducting polymers.

TABLE 2. Comparison of the nanofibers generated by the RVPP method vs. conventional materials.

EXAMPLE 7; SYNTHESIS OF SUBMICRON PEDOT PARTICLES OF HIGH

ELECTRICAL CONDUCTIVITY VIA CONTINUOUS AEROSOL VAPOR

POLYMERIZATION

This example shows the synthesis of conducting polymer particles using aerosol and vapor flow reaction. Using these methods, submicron particles can be produced continuously and in large quantity (100 mg/hour). Furthermore, the methods described here can produce conducting polymers with tunable particles size distribution and electrical conductivity.

A platform for synthesizing scalable submicron-sized particles of the conducting polymer poly(3,4-ethylenedioxythiophene) (PEDOT) is shown herein.

The synthesis is based on a hybrid approach utilizing an aerosol of aqueous oxidant droplets and monomer vapor to engineer a scalable synthetic scheme. This aerosol vapor polymerization (AVP) technology results in bulk quantities of discrete solid-state submicron particles (750 nm diameter) with so far the highest reported particle conductivity (330±70 S/cm). Moreover, particles are dispersible in organics and water, obviating the need for surfactants, and remain electrically conductive and doped over a period of months. This enhanced processability and environmental stability enables incorporation in thermoplastic and cementitious composites for engineering chemoresistive pH and temperature sensors.

EXAMPLE 8: DEPOSITION OF PEDOT NANOFIBRILLAR COATINGS ON C1-FE2O3-

CONTAINING BRICK SUBSTRATES

The following example provides data for methods of coating α-Fe ₂ O ₃ - containing brick substrates with a nanofibrillar conducting coating of poly(3,4- ethylenedioxythiophene) (PEDOT), and devices formed using the disclosed coating method.

A scalable, cost-effective and versatile chemical synthesis using a fired brick to control oxidative radical polymerization and deposition of a nanofibrillar coating of the conducting polymer poly(3,4-ethylenedioxythiophene) (PEDOT) is disclosed. A fired brick’s open microstructure, mechanical robustness and ~8 wt% a-Fe203 content afford an ideal substrate for developing electrochemical PEDOT electrodes and stationary supercapacitors that readily stack into modules. Five-minute epoxy serves as a waterproof case enabling the operation of PEDOT-coated brick supercapacitors while submerged underwater and a gel electrolyte extends cycling stability to 10,000 cycles with -87% capacitance retention. Using a-Fe ₂ O ₃ -impregnated carbon cloth as a substrate, we also produce state-of-the-art flexible nanofibrillar PEDOT supercapacitors characterized by high areal capacitance (2243 mF/cm ² for two-electrode vs.

6194 mF/cm ² for three-electrode) and high areal energy density (412 mWh/cm ² ).

Fired brick, typically used for construction and architectural aesthetics, is one of the most durable materials with a 5,000-year history dating back to Neolithic China. This masonry building block is commonly found in various red tones and mostly comprised of fused particles of silica (SiO ₂ ), alumina (AI ₂ O ₃ ) and hematite (a-Fe ₂ O ₃ ). The red color of a brick originates from hematite, a pigment first utilized by humans 73,000 years ago and serving today as a low- cost naturally abundant inorganic precursor for catalysts, magnets and alloys. State-of-the-art energy storage materials are also produced from hematite. For example, FeN _x , FeP and Li ₅ FeO ₄ are synthesized via anionic or cationic exchange for potassium-ion batteries, Zn-air batteries, pseudocapacitors and lithium-ion batteries; electrochemical transformation of hematite leads to FeOOH supercapacitor anodes.

Chemistries enabled by hematite provide an opportunity for developing cutting-edge functionalities on a fired brick where 8 wt% a-Fe ₂ O ₃ content and a 3D porous micro-structure afford an ideal substrate for engineering a mechanically robust electrode. A supercapacitor is developed using a brick’s hematite microstructure as reactant to vapor deposit a nanofibrillar coating of the conducting polymer poly(3,4-ethylenedioxythiophene) (PEDOT). Vapor-phase synthesis leads to PEDOT coatings exhibiting a high electronic conductivity and facile charge transfer making it an ideal route for producing electrodes. This synthesis utilizes a brick’s open microstructure and thermal stability by permeating acid and monomer vapor through its pores at 160 °C to control a- Fe ₂ O ₃ dissolution and Fe ³⁺ hydrolysis with concomitant oxidative radical polymerization.

Conversion of a fired brick’s a-Fe203 microstructure to a nanofibrillar PEDOT coating

Deposition of PEDOT nanofibers is initiated by dissolving a-Fe2O3 at 160 °C with HCI vapor, this process liberates Fe3+ ions, promotes hydrolysis and initiates precipitation of colloidal 1 D FeOOH nuclei. As previously reported, partially dissolved FeOOH nuclei serving as templates oxidize 3,4- ethylenedioxythiophene (EDOT) monomer vapor and control oxidative radical polymerization (FIGS. 67A, 67B, 72A, 72B, and 72C). We advance previous findings by demonstrating here that the synthesis utilizes two potential polymerization initiators, i.e., oxidant (Fe3+) and acid (HCI) where the former leads to oxidative radical polymerization and the latter, to acid-catalyzed polymerization (fig. S1 D). We readily control the polymerization mechanism as an acid-catalyzed polymerization typically produces nonconductive oligomers stemming from active chain termination. Only PEDOT synthesized via oxidative radical polymerization exhibits long conjugation length, ordered chain packing, low electrical resistance, as well as high chemical and physical stability.

In our polymerization mechanism, the acid concentration determines both dissolution rate and synthetic pathway. Using a HCI vapor concentration less than 4.8 mM leads to an incomplete reaction because both oxidative radical polymerization and acid-catalyzed polymerization are impeded (FIG. 72E). Increasing concentration to 14 mM liberates Fe3+ and promotes oxidative radical polymerization resulting in PEDOT of low electrical resistance, whereas concentrations above 14 mM activate the acid-catalyzed polymerization pathway resulting in uncontrolled reactions (FIG. 72F). When oxidative radical polymerization dominates, the EDOT vapor concentration determines the thickness of a polymer coating and its electrical resistance (FIG. 72G).

A blue PEDOT coating is visible on a brick 4 h after initiating a reaction and its thickness increases inversely proportional with electrical resistance until the end of reaction at 14 h. An extended polymerization time increases the polymer coating’s two-point probe electrical resistance (FIG. 72H) because PEDOT loses dopant during heating; fortunately, post-synthetic doping lowers the electrical resistance. Our synthesis produces a 400 mm thick nanofibrillar PEDOT coating (2.8 wt%) exhibiting 2 W two-point probe electrical resistance and nanofibers characterized by a ~30 mm length and -190 nm diameter (FIGS. 73A, 73B, 73C, 73D). Polymer, purified by repeated rinses in methanol, is comprised of S, C, O and doped by Cl- in situ during polymerization as shown in energy-dispersive X-ray spectra (FIG. 73E). Polymer deposits as an embedded network throughout pores (FIG. 72C) because a brick’s a-Fe2O3 microstructure enables nanofibrillar grafting in-situ resulting in strong adhesion. A nanofibrillar PEDOT coating exhibits minimal delamination via a Scotch tape test whereas a coating of the commercial product PEDOT:poly(styrenesulfonate) peels off completely (FIG. 67C). Notably, this deposition technology is scalable (FIG. 67D) and patternable (FIG. 73F).

Other iron-containing substrates such as natural minerals (FIG. 74 and FIG. 75) are readily coated as well. For example, PEDOT coatings on hematite (a-Fe2O3) and a-Fe2O3-containing rosy pink granite exhibit two-point probe electrical resistances of 10 W and 170 W, respectively. Nanofibers form on hematite uniformly because its high Fe3+ content promotes hydrolysis and oxidative radical polymerization simultaneously. Iron sulfide minerals such as pyrite (FeS2), chalcopyrite (CuFeS2) and troilite (FeS), once coated by polymer, exhibit photothermal heating due to PEDOT’s absorption of infrared energy.

Nanofibrillar PEDOT-coated brick electrochemical electrode

A nanofibrillar PEDOT-coated brick electrode exhibits a quasi-rectangular shaped three-electrode cyclic voltammogram and a capacitance of 2612 mF/cm2 (187 F/g based on PEDOT’s mass) at 2 mV/s in 1 M H2SO4. The Fe3+/Fe2+ redox pair at 0.37 V and 0.49 V (versus Ag/AgCI) arise due to iron species in brick and disappears as scan rate increases to 25 mV/s because double-layer capacitance outcompetes faradic pseudocapacitance at faster rates (FIG. 68A and FIG. 76A). Rate performance tests demonstrate capacitive behavior as scan rate increases to 100 mV/s (FIG. 76B); however, due to limited charge transport the curve changes to fusiform shape. Capacitance is also dependent on the aqueous electrolyte and sulfuric acid leads to greater capacitance (1635 mF/cm2) than sodium sulfate (878 mF/cm2) at 25 mV/s (FIG. 76A). This drastic difference is plausibly due to higher ionic mobility in H+ (36.23 x 10-8 m2 V-1 s-1) versus Na+ (5.19 x 10-8 m2 V-1 s-1) or a lower electrical resistance caused by doping at low pH. We probed this behavior further using electrochemical impedance spectroscopy. Nyquist plots and an equivalent circuit diagram demonstrate a significantly lower ion diffusion resistance for H2SO4 (1 .7 W) versus Na2SO4 (4.6 W) and minimal change in electrode material electrical resistance at low pH (FIG. 76C).

Symmetric nanofibrillar PEDOT-coated brick supercapacitor

Two nanofibrillar PEDOT-coated bricks serve as electrodes in a symmetric supercapacitor using 1 M H2SO4 liquid electrolyte (FIG. 76D).

Nyquist plot shows an aggregated internal resistance of 3 W and a line with a -45° slope between semicircle and low-frequency domain (Warburg region)

(FIG. 76E). This line is characteristic of thick electrodes where a tortuous path stifles ion diffusion. Cyclic voltammogram shows a quasi-rectangular shape between 0 and 1 V (collected at 2 mV/s) leading to an areal capacitance of 1 ,591 mF/cm2 (FIG. 68B, black curve and FIG. 76F). Our supercapacitor possesses low internal resistance resulting in a low IR drop (0.01V) during galvanostatic charge-discharge experiments at 0.5 mA/cm2 current density in a 1 V window. These curves demonstrate an areal capacitance of 1 ,597 mF/cm2 as well as areal energy and power densities of 222 mWh/cm2 and 0.25 mW/cm2, respectively (FIG. 76G). High power density (12.5 mW/cm2) is obtained at a current density of 25 mA/cm2 albeit with lowered capacitance (706 mF/cm2) and energy density (98 mWh/cm2) because ion transport in our thick electrode is limited (shown by high IR drop of 0.4 V). Our device works in an extended voltage window (1.2 V) resulting in cyclic voltammograms and galvanostatic charge-discharge curves that retain shape (FIG. 76H and FIG. 76I). Connecting three devices in series increases the voltage window to 3.6 V, this also triples the internal resistance and reduces output current to one-third (FIG. 77A). A tandem device reaches an output voltage of 2.685 V (charged at 4.5 V for 15 s) and lights a white light-emitting diode for 11 min. The output voltage decreases to the same level as the light-emitting diode’s turn-on voltage (2.546 V) after discharging for 214 s (FIGS. 77B, 77C, and 77D).

Quasi-solid-state nanofibrillar PEDOT-coated brick supercapacitor

To minimize electrolyte leakage, we develop a symmetric supercapacitor using a poly(vinyl alcohol) / 1 M H2SO4 gel that binds PEDOT-coated bricks and serves as electrolyte and separator (FIG. 78A). The gel electrolyte layer (0.7 mm thick) prevents bricks (2.8 mm thick) from short-circuiting and leads to enhanced adhesion between electrodes. In a tensile test, our electrode-gel-electrode structure withstands a shearing force equal to 1000 times the device’s weight (FIG. 78B). Intimate contact between gel and PEDOT nanofibers enhances charge transfer resulting in low internal resistance (2.5 W) and a linear Nyquist plot (FIG. 78C). Areal capacitance (868 mF/cm2) and areal energy density (121 mWh/cm2) are calculated from galvanostatic charge-discharge curves (collected at 0.5 mA/cm2); cyclic voltammograms also show capacitive behavior (FIG. 68B, 68C, and 78D). A gel electrolyte leads to lower figures of merit than a liquid electrolyte because gel permeation throughout the electrode is effectively 50% less (FIG. 68D and FIG. 78E).

Outdoor exposure is inevitable for a stationary supercapacitor and epoxy encapsulation affords a cost-effective, mechanically robust and waterproof housing. An epoxy-coated supercapacitor retains -90% of original capacitance and exhibits -100% coulombic efficiency after 10,000 charge-discharge cycles (collected at 25 mA/cm2) (FIG. 68E, red curves). This 5-minute epoxy coating prevents water evaporation from the gel’s hydrated ionic percolation network (FIG. 78F) enabling 10,000 charge-discharge cycles at 5 mA/cm2 (640 h of continuous operation) with -87% capacitance retention (fig. 2E, black curves). Gel electrolyte and encapsulation enables operation at temperatures ranging between -20 °C and 60 °C as shown by cyclic voltammograms and Nyquist plots (FIGS. 78G and 78H) where capacitance increases proportionally with temperature due to enhanced ionic transport; PEDOT remains capacitive after reversible heating-cooling cycles (FIG. 78I).

Epoxy renders a device waterproof resulting in a stationary supercapacitor module that charges to 3 V in 10 s while immersed in water and lights up a green light-emitting diode (2.155 V forward voltage) for ~10 min (FIGS. 78J, 78K, and 78L). A gel electrolyte and our deposition technology enable scale up as demonstrated by connecting six large nanofibrillar PEDOT- coated brick electrodes (2 cm x 1 cm x 1 cm) in series resulting in a supercapacitor module that charges to 3 V in 5 s readily lighting up a green light- emitting diode (FIG. 68F and FIG. 79).

Materials

Chlorobenzene (99%), 3,4-ethylenedioxythiophene (97%), poly(vinyl alcohol) (Mw 89, 000-98,000, 99+% hydrolyzed), methanol (>99.8%), ammonium hydroxide solution (28.0-30.0%) and hydrochloric acid (37%) are purchased from Sigma-Aldrich; sulfuric acid (AR) is purchased from Macron. The PEDOT:PSS solution (Clevios PH 1000) is purchased from Heraeus company.

All chemicals are used without further purification. Platinum foil (0.025 mm thick, 99.9%) is purchased from Alfa Aesar and utilized for engineering electrode leads and Celgard 3501 membrane is used as a separator. The ELAT hydrophilic plain carbon cloth is purchased from FuelCellStore (College Station, Texas) and fired brick (20.32 cm x 10.16 cm x 5.72 cm) is purchased from The Home Depot Inc. and cut using a diamond saw for developing electrodes. Hematite particles (□- Fe2O3) produced by NewLook Inc. are purchased at The Home Depot Inc. Materials for making concrete (purchased from The Home Depot Inc.) include commercial-grade Quikrete Portland cement (Type I/ll), Quikrete all-purpose sand and Pavestone multi-purpose patio/paver base.

Characterization methods

Scanning electron micrographs and energy-dispersive X-ray spectra are collected with a JEOL 7001 LVF FE-SEM. An infrared camera (FLIR ETS320) is used for temperature mapping and a Canon EOS 70D digital camera for photography. Two-point probe resistance measurements are carried out using a Fluke 177 True RMS digital multimeter with 3 mm distance between two probes. Thermogravimetric analysis is conducted on a Discovery TGA (TA Instruments). Cyclic voltammetry, galvanostatic charge-discharge measurements and electrochemical impedance spectroscopy are performed in a BioLogic VMP3 multi-potentiostat. For electrochemical impedance spectroscopy, the sinusoidal disturbance is 10 mV with frequencies scanned between 100 kHz and 0.1 Hz. A Nyquist plot shows real impedance Z versus imaginary impedance -Z” under a sinusoidal disturbance at the open circuit potential. Fitting of Nyquist plot using an equivalent circuit diagram contains solution resistance (Rs), electrode material resistance (Rm), material capacitance (Cm), double layer capacitance (Cdl) and constant phase element (CPE). Here, Rs reflects the electrolyte ionic mobility and Rm represents the electrical resistance of the electrode.

Preparation of poly (vinyl alcohol)/H2SQ4 gel electrolyte

The gel electrolyte is formulated using 1 g of poly(vinyl alcohol) powder dissolved in 10 mL deionized water under vigorous stirring at 90 °C and cooled to around 50 °C. Dropwise addition of 1 g of concentrated H2SO4 (1 M) is then carried out by pipetting acid on the inner wall under vigorous stirring to prevent carbonization of poly(vinyl alcohol). Stirring minimizes localized heating and is carried out for 1 h resulting in a homogeneous, translucent and colorless solution.

Direct deposition of a nanofibrillar PEDOT coating on minerals and rocks

FIG. 74A shows a step-by-step fabrication of the glass reactor that we use for obtaining polymer coatings on minerals, rocks and brick. This is carried out in a sealed pipet affording an ideal chamber where stoichiometries are readily controlled. First, a mineral sample is loaded into a glass pipet with a sealed bottom; a Bunsen burner is utilized for sealing the pipet. A narrow neck is created by pressing on the heated and soft pipet glass wall enabling minerals to be kept apart from the acid and monomer solution during the reaction. The pipet tip is cut to widen the hole diameter and serves as a funnel enabling the introduction of liquid reactants. This polymerization requires 1) EDOT in chlorobenzene solution and 2) hydrochloric acid; these two liquid reactants are mixed in 10:1 ratio using a bath sonicatorfor 1 min until a homogeneous emulsion forms. Then, 30 mL of the emulsion is added in the pipet reactor using a funnel; note that the mineral and the emulsion do not meet each other prior to sealing this reactor. The funnel is carefully withdrawn to avoid touching the mineral, then the pipet tip is flame sealed and the reactor is introduced into an oven.

Synthesis is carried out at 140 °C for 14 h using stoichiometries that are sample dependent, for example, pyrite and troilite require 0.85 M EDOT and 10 M HCI, chalcopyrite and pyrrhotite require 0.85 M EDOT and 12 M HCI and rosy pink granite, fired brick and hematite require 0.35 M EDOT and 12 M HCI. Polymer is purified by rinsing sample in methanol three times prior to characterization. Samples are dedoped by soaking in 250 mL concentrated ammonium hydroxide solution (28.0 -30.0%) for 2 h and subsequently thrice washed with methanol and air dried at 25 oC for 0.5 h.

Synthesis of nanofibrillar PEDOT coating on a brick

Brick is cut using a diamond saw (± 0.03 cm error) into the following four different sizes: 1.00 cm x 0.50 cm x 0.28 cm (for studying synthesis and electrochemistry), 1.27cm x 1.27cm x 0.20 cm (for patterning), 2.00 cm x 1.00 cm x 1.00 cm (for supercapacitor) and 10.16 cm x 6.77 cm x 5.72 cm (for scaling demonstration). A brick is thrice washed with deionized water to remove surface dust then dried at 160 °C for 1 h and cooled to room temperature.

The synthesis of a 1 cm x 0.5 cm x 0.28 cm brick is performed in a 25 ml_ Teflon-lined stainless-steel autoclave as shown in fig. S1A. A brick is placed on a glass reservoir then 200 mL of a 0.70 M EDOT/chlorobenzene solution are loaded in a separate glass reservoir and 25 mL of 12 M HCI are directly injected in the Teflon liner. The reactor is closed and introduced into an oven at 160 °C for 14 h. The product is washed thrice with excess methanol and dried at room temperature before carrying out tests. Similarly, the synthesis of a 2 cm x 1 cm x 1 cm brick is performed in a 125 ml_ Teflon-lined stainless-steel autoclave using 600 mL of a 0.70 M EDOT in chlorobenzene solution and 75 mL of 12 M HCI.

A brick (1.27cm x 1.27cm x 0.20 cm) is patterned using a polyimide tape mask. Synthesis is carried out at 150 °C for 14 h in a 125 ml_ Teflon-lined stainless-steel autoclave containing 1 ml_ of a 0.85 M EDOT solution in chlorobenzene and 0.6 mL of 12 M HCI (fig. S2F). Scaling is carried in a glass reactor (12.30 cm x 8.55 cm x 11.30 cm) with a large brick (10.16 cm x 6.77 cm x 5.72 cm). The reaction is carried out using 15 ml_ of 12 M HCI and 15 mL of a 0.85 M EDOT in chlorobenzene solution at 150 °C for6 h (fig. 1C).

We produce a PE DOT coating on concrete by applying D-Fe2O3 particles to a concrete surface (fig. S9E). This composite is produced by is mixing sand, stone, Portland cement and water in a weight ratio of 3 : 1.5 : 1 :

0.7. An uncured concrete slurry is then injected into a 1 27cm x 1 27cm x 2.54 cm mold, stirred to remove gas bubbles and cured for 3 days in ambient conditions. A partially cured concrete bar is dipped in an aqueous dispersion of a-Fe2O3 (0.25 g/mL) for 3 s, then dried in air. Synthesis is performed at 150 °C for 14 h in a 125 mL Teflon-lined stainless-steel autoclave loaded with 1 mL of a 0.45 M EDOT in chlorobenzene solution and 0.1 mL of 12 M HCI.

Fabrication of a tandem nanofibrillar PEDOT-coated brick supercapacitor comprised of 3 cells connected in series

The gel electrolyte is a mixture of 0.1 g of poly(vinyl alcohol) in 1 mL of 1 M H2SO4 and 200 mL are added to a PEDOT-coated brick face (1 cm x 1 cm). This electrolyte is allowed to stabilize for 12 h at ambient conditions; this process is repeated for all brick electrodes resulting in a thick gel layer (fig. S8A). An additional 50 mL of poly(vinyl alcohol)/H2SO4 solution are added serving as a binder between bricks. Electrodes are dried in an oven at 50 °C for 2 h and platinum foil, serving as a lead, is attached to bricks using polyimide tape.

The results of these experiments demonstrated methods of energy storage on the surface of a common brick by using a-Fe ₂ O ₃ as an oxidant precursor to control oxidative radical polymerization and conformal deposition of a capacitive nanofibrillar PEDOT coating from the vapor phase. A brick’s mechanical stability and open structure result in mechanically robust PEDOT- coated brick electrodes that when connected in series and coated with epoxy, produce a stable stationary waterproof supercapacitor module.

EXAMPLE 9: DEPOSITION OF PEDOT NANOFIBRILLAR COATINGS ON C1-FE2O3-

IMPREGNATED CARBON CLOTH

The following example provides data for methods of coating a-Fe ₂ O ₃ - impregnated carbon cloth substrates with a nanofibrillar conducting coating of poly(3,4-ethylenedioxythiophene) (PEDOT), and devices formed using the disclosed coating method.

Using a-Fe2O3-impregnated carbon cloth as a substrate, the chemical synthesis methods described in Example 8 were used to produce state-of-the-art flexible nanofibrillar PEDOT supercapacitors characterized by high areal capacitance (2243 mF/cm ² for two-electrode vs. 6194 mF/cm ² for three- electrode) and high areal energy density (412 mWh/cm ² ).

The chemical synthesis methods described in Example 8 are versatile and enable the processing of a-Fe203 particles for use in developing state-of- the-art flexible nanofibrillar PEDOT supercapacitors supported on carbon cloth substrates. A nanofibrillar PEDOT flexible electrode is characterized by a mass loading (30.2 mg/cm ² ) three times greater than existing metal oxide or carbon- based electrodes (10 mg/cm ² ) affording the potential for storing large amounts of charge. Retaining electrochemical performance at increased mass loading is fundamentally challenging because scaling introduces charge transport limitations and mechanical instability in the electrode structure. Our PEDOT- coated carbon cloth electrode overcomes these limitations because its 3D hierarchical architecture combines PEDOT nanofibers and carbon cloth micropores that increase surface area and provide a percolation network for conducting ions. A core-shell carbon cloth-PEDOT structure mitigates mechanical deformation as the carbon cloth core is void of polymer and characterized by large free volume. A flexible electrode exhibits a high areal three-electrode capacitance (6194 mF/cm ² ) whereas a two-electrode bendable supercapacitor possesses state-of-the-art areal capacitance (2,243 mF/cm ² ) and areal energy density (412 mWh/cm ² ) in 1 M H ₂ SO ₄ electrolyte. Notably, the areal capacitance and energy density reported here are the highest among flexible organic supercapacitors exceeding some inorganic pseudocapacitors.

Impregnation of carbon cloth with a-Fe2Q3 particles for developing flexible nanofibrillar-PEDOT supercapacitors

Carbon cloth (1.27 cm x 0.635 cm) is treated with O2 plasma for 30 s and impregnated with 50 mL of a 0.25 g/mL aqueous dispersion of a-Fe2O3 particles; two glass plates enable pressing particles into carbon cloth with ~50 N of force. This process is repeated thrice to achieve a high loading, excess liquid is removed in between pressings using Kimwipe and a coated substrate is dried at 50 °C for 0.5 h.

Deposition of a nanofibrillar PEDOT coating is carried out at 160 °C for 14 h and requires 200 mL of a 0.85 M EDOT in chlorobenzene solution and 20 mL of 11 M HCI. A polymer-coated substrate is washed by sequential immersion in methanol (1 h) and concentrated hydrochloric acid (1 h) and repeated thrice at room temperature. A dry electrode is soaked with gel electrolyte by immersion in poly(vinyl alcohol)/H2SO4 solution at 70 °C for 12 h followed by ambient drying for 3 h. An electrode is then swelled in 1 M H2SO4 solution at room temperature for 12 h prior to device testing.

Flexible nanofibrillar PEDOT-coated carbon cloth supercapacitor

A flexible electrode was developed by impregnating a-Fe2O3 particles in carbon cloth (450 mm thick) affording a high particle loading (18 wt%, FIG. 80A) and a porous substrate for carrying out vapor polymerization. The electrode is characterized by a core-shell carbon-PEDOT architecture where PEDOT nanofibers deposit as a 100 mm thick coating on both sides of carbon cloth (FIG. 69A). The carbon cloth core (250 mm thick) is void of polymer affording large free volume for mitigating mechanical deformation and developing flexible electronics (FIG. 69B and FIG. 80B). Mass loading in our PEDOT electrode (30.2 mg/cm ² , FIG. 70A) is three times greater than in a commercial supercapacitor electrode (10 mg/cm ² ). For comparison, a PEDOT electrode produced from FeCI3 solution and EDOT monomer vapor possesses a low mass loading (2.1 mg/cm ² , FIG. 70A) (FIGS. 80C and 80D). Impregnation with a-Fe2O3 particles leads to a high packing density of PEDOT nanofibers on carbon cloth and enables formation of PEDOT composites using Portland-based concrete (FIG. 80E).

An electrode exhibits an areal capacitance of 6,194 mF/cm ² (205 F/g based on PEDOT’s mass) and a quasi-rectangular cyclic voltammogram at 2 mV/s after removal of iron species (FIG. 80F). This exceedingly high areal capacitance originates from a high mass loading, nanofibrillar morphology and high surface area. Note that capacitance is negligible for carbon cloth (20 mF/cm ² ) while FeCI3-derived PEDOT electrode exhibits low capacitance (302 mF/cm ² and 144 F/g) at 2 mV/s (FIG. 70B). Scanning electron micrographs for FeCI3-derived electrode show intermittent contact between PEDOT and carbon support whereas a-Fe2O3-derived electrode features intimate contact resulting in lower resistance and higher gravimetric capacitance (FIGS. 80C and 80G).

A flexible symmetric supercapacitor is developed using two a-Fe2O3- derived PEDOT-coated carbon cloth electrodes and 1 M H2SO4 electrolyte; the electrode is coated with poly(vinyl alcohol) to enhance PEDOT adhesion (FIG.

81 A). This device retains a cyclic voltammogram’s shape while bending from Oo to 180° and exhibits -94% capacitance retention after 500 bending cycles (FIGS. 70C and 70D). Internal resistance in Nyquist plot decreases from 1.9 to 1.6 W as bending angle changes from 0° to 180° because device components are compacted during bending thereby enhancing electrical contact and ion transport (FIGS. 81 B). Our device exhibits areal energy and power densities (collected at 0.5 mA/cm ² ) of 312 mWh/cm ² and 0.25 mW/cm ² , respectively. A flexible PEDOT supercapacitor exhibits state-of-the-art areal capacitance (2,243 mF/cm ² ) determined from a galvanostatic charge-discharge curve collected at 0.5 mA/cm ² ; quasi-rectangular cyclic voltammogram leads to 2107 mF/cm ² at 2 mV/s (FIGS. 70E and 70F).

The core-shell microstructure of our electrode is characterized by an open architecture with free volume affording a path of low tortuosity for ions.

Moreover, the polymer’s nanofibrillar structure mitigates expansion-contraction of electrode during charge-discharge cycles resulting in a device with a high-rate capability. Our supercapacitor retains -89% capacitance after 10,000 cycles with a constant coulombic efficiency above 99% (collected at 25 mA/cm2) (FIG. 71 A). Capacitance increases by 10% after the first 1 ,000 cycles because of improved wetting and ion diffusion and slowly decreases thereafter due to electrolyte evaporation and loss of dopant. Expanding the voltage window to 1.2 V maximizes energy (412 mWh/cm2 at 1 mA/cm2) and power (30 mW/cm2 at 50 mA/cm2) densities (FIG. 71 B and FIG. 81 C). Flexible devices connected in series (3.6 V voltage window) are readily rolled into a cylinder (2.25 mm radius) and sealed with polyimide tape (FIGS. 81 D, 81 E, and 81 F). A Ragone plot shows the superior areal energy density of our flexible supercapacitor versus state-of-the-art devices based on materials such as carbon allotropes, metal oxides as well as a metal-organic framework, MXene and transition metal dichalcogenide (FIG. 71 C).

The results of these experiments demonstrated a strategy for fabricating flexible nanofibrillar PEDOT supercapacitors by impregnating carbon cloth with a-Fe2O3 particles. The flexible supercapacitors feature the highest areal capacitance (2243 mF/cm2) and areal energy density (412 mWh/cm2) among existing flexible organic-based supercapacitors.

EXAMPLE 10: PHOTOTHERMAL ENHANCEMENT OF PEDOT COATINGS

The following example provides data for photothermal enhancement of substrates with PEDOT coatings formed using the methods described in Example 8.

PEDOT coatings on granite and iron sulfide minerals are featureless because granite contains low Fe3+ concentration and sulfides generate SO ₄ ^2- upon heating, both stifle formation of 1 D b-FeOOH templates. A one-dimensional ferric oxyhydroxide inorganic seed serves as preferential nucleation site for polymerization resulting in nanofibrillar growth. Resistances of semiconducting pyrite (FeS2), chalcopyrite (CuFeS2) and troilite (FeS) decrease when coated by PEDOT from approximately 8,000, 260 and 20 W to 460, 20 and 13 W, respectively. Pyrrhotite (Fei- _x S) is an exception, here electrical resistance increases from 0.4 W to 47 W after deposition of a PEDOT coating.

After photon injection, a PEDOT coating exhibits an enhanced photothermal effect (FIG. 75, insets). A PEDOT coating on a mineral absorbs light and generates heat via non-radiative dissipation leading to a surface temperature 5 °C higher than non-coated minerals. This photothermal effect is not exhibited by a PEDOT coating on hematite because pristine a-Fe2O3 also converts light to heat.

EXAMPLE 11: 3D NANOFIBRILLAR PEDOT MICRO-SUPERCAPACITORS

The following example provides data for the fabrication and characterization of 3D nanofibrillar PEDOT micro-supercapacitors formed direct conversion of Fe2O3 using methods related to the methods described in Example 8.

Micro-supercapacitors (mSCs) are attractive electrochemical energy storage devices serving as alternatives to batteries in miniaturized portable electronics owing to high-power density and extended cycling stability. Current state-of-the-art micro-fabrication strategies are limited by costly steps producing materials with structural defects that lead to low energy density. Here, we introduce an electrode engineering platform that combines conventional micro- fabrication and polymerization from the vapor phase producing 3D mSCs of the conducting polymer poly(3,4-ethylenedioxythiophene) (PEDOT). A sputtered Fe2O3 precursor layer enables deposition of a 250 nm thick polymer coating comprised of a high packing density of vertically aligned PEDOT nanofibers possessing exceptional electrical conductivity (3580 S cm-1). Our 3D mSCs exhibit state-of-the-art volumetric energy density (16.1 mWh cm-3) as well as areal (21.3 mF cm-2) and volumetric (400 F cm-3) capacitances in 1 M H2SO4 aqueous electrolyte. These figures of merit represent the highest values among conducting polymer-based mSCs. Electrochemical performance is controlled by investigating coating thickness, gap distance, fractal geometry and gel electrolyte (1 M H2SO4/polyvinyl alcohol (PVA)). Our quasi-solid-state mSCs exhibit extended rate capability (50 V s-1), retain 94% of original capacitance after 10,000 cycles and remain thermally stable up to 60 °C.

The rapidly growing market for autonomous devices, micro- electromechanical systems, micro-robots and radio frequency identification tags creates a demand for high-performing miniaturized electrochemical energy storage. Currently, high energy density micro-batteries fulfill the market’s needs albeit suffering from diffusion-limited low power density and poor cycling stability that shortens lifespan. Micro-supercapacitors (mSCs) are an alternative to micro- batteries affording high charge/discharge rates and high cycling stability for developing power supplies for portable electronics. These devices are produced via inkjet printing, laser scribing or electrochemistry with a 3D electrode configuration comprised of nanostructures that facilitate ion diffusion and enhance energy density.

Interdigitated electrodes for mSCs are commonly coated by state-of-the- art materials such as capacitive onion-like carbon or carbide-derived carbons as well as by pseudocapacitive Ru02 or Ni(OH)2. Unfortunately, these carbon allotropes possess low energy density, and metal oxides/hydroxides exhibit low cycling stability as well as poor rate capability due to phase transition and limited ion migration. Alternatively, mSCs based on conducting polymers such as polyaniline (PANi), polypyrrole (PPy) and poly(3,4-ethylenedioxythiophene) (PEDOT) show excellent rate capability due to facile charge transport. Moreover, faradic near-surface redox reactions in conducting polymers also afford a pseudocapacitive mechanism for increasing energy density. Among conducting polymers, PEDOT is an ideal electrode material for mSCs due to its high electrical conductivity (reaching 7000 S cm-1) and its ability to produce nanofibers, nanowires and nanosheets that increase an electrode’s active surface area.

However, integrating 3D PEDOT nanostructures in mSCs is challenging. Current fabrication strategies require pre-fabricated templates that unfortunately possess low packing density and low aspect ratio scaffolds resulting in low surface area. Moreover, polymer is typically produced via electrochemical syntheses, that due to an inhomogeneous distribution of reactants, produce conjugated backbones with structural defects. These limitations highlight the urgency for advancing current electrode fabrication protocols.

Here, we present a superior low-cost approach for engineering 3D mSCs compatible with conventional micro-fabrication and based on our previously reported rust-assisted vapor-phase polymerization. We produce homogeneous coatings of vertically directed PEDOT nanofibers characterized by an electronic conductivity of 3580 S cm-1 ; deposition occurs in one step resulting in a high packing density of 1 D nanostructures with aspect-ratio of 100. Our electrode engineering protocol utilizes sputtered a-Fe2O3 as a ferric ion-containing solid- state oxidant-precursor to induce dissolution, liberation of ferric ions, and Fe3+ hydrolysis concomitant with oxidative radical polymerization. The electrode thickness, gap distance and fractal geometry of PEDOT mSCs is controlled resulting in state-of-the-art areal (21 .3 mF cm-2) and volumetric (400 F cm-3) capacitances as well as energy density (16.1 mWh cm-3). Such magnitudes exemplify the highest figures of merit for conducting polymer-based mSCs. This work also develops quasi-solid-state mSCs exhibiting extended rate capability (50 V s-1) and high cycling stability (94 % capacitance retention after 10,000 cycles) as well as able to operate efficiently at a temperature of 60 oC.

Fabrication of a nanofibrillar PEDOT micro-supercapacitor

Micro-fabrication requires spin coating photoresist on Si02 and photolithography to produce an interdigitated pattern. Chromium, gold and iron oxide (Fe2O3) are sequentially deposited serving as an adhesion layer, current collector and oxidant precursor, respectively (FIG. 82A, FIG. 87). An Fe2O3- coated electrode is introduced into a reaction vessel, heated, and exposed to vapors of 1) concentrated hydrochloric acid (HCI) and 2) 3,4- ethylenedioxythiophene (EDOT) monomer in chlorobenzene to produce PEDOT nanofibers (FIG. 82B, FIG. 88). Nanofibers, as previously reported35, form via dissolution of an iron(lll)-containing solid and hydrolysis of ferric ions concomitant with preferential nucleation of monomer vapor on 1 D FeOOH spindles. Polymer grows on partially dissolved spindles via oxidative radical polymerization in a step-growth fashion resulting in the deposition of PEDOT nanofibers on a current collector (FIG. 82B, FIG. 84). After lift-off, PEDOT nanostructures remain intact as interdigitated electrodes adhered to the current collector FIG. 82C). This production of this electrode structure is aided by a double-layered photoresist undercut that facilitates lift-off (FIGS. 90, 91 , and 92).

A mSC is comprised of two five-finger electrodes (FIG. 82D) homogeneously coated by PEDOT nanofibers; a finger is 200 mm wide and the device’s interdigitated gap is 200 mm (FIG. 82E). Polymer and metal electrodes are strongly adhered as demonstrated after Scotch tape delamination tests and repeated sonication cycles (FIG. 93). A high packing density of vertically aligned nanofibers is produced, as shown in cross-sectional electron micrograph (FIG. 82F), plausibly due to confined polymerization within the photoresist trenches.

Direct characterization of the intrinsic properties of a nanofibrillar PEDOT-coated electrode

Understanding intrinsic chemical properties of the electrode material is necessary for advancing mSC engineering and yet few discussions on characterization are present in the literature. Here, we apply microscopy, spectroscopy, current-voltage (l-V) profiles and four-point probe resistivity measurements to probe molecular and solid-state structure. A finger electrode is coated by a high packing density of nanofibers possessing a 10 mm length and 100 nm diameter (aspect ratio -100) (FIGS. 83A, 83B, 94, and 95). Each nanofiber possesses a core-shell structure confirmed via high-angle annular darkfield (HAADF) imaging in scanning transmission electron microscopy (FIG. 83C) where image contrast is roughly proportional to the square of the atomic number. Elemental maps via energy-dispersive X-ray spectroscopy show an iron-containing core and a sulfur-containing shell for a nanofiber (FIGS. 83D and 96). This iron core is dissolved in 6 M HCI to produce a hollow 1 D polymer nanostructure with an extended surface area and a highly oxidized conjugated backbone; chemical bond resonance signals in Raman spectroscopy (FIG. 83E) indicate a high doping state.

Electronic charge transport, plotted by applying a potential and measuring current (FIG. 83G inset, and FIG. 97), shows that a PEDOT coating is characterized by ohmic behavior (straight line), low resistance (large slope) (FIG. 83G) and a small Schottky barrier that facilitates charge transport. The line slope decreases linearly and inversely proportional with probe distance because the polymer coating is a homogeneous and continuous percolation network. To probe charge transport as a function of PEDOT’s polycrystalline structure, powder X-ray diffraction (FIG. 83F) is carried out revealing three characteristic peaks at 6.4°, 12.1 ° and 25.8° corresponding to lateral chain packing of (100), (200) and p- p stacking of (020), respectively. Peaks’ widths at half-heights are 0.43, 0.79 and 1 .38; these small values indicate a crystalline polymer structure that enhances charge transport. A four-point probe conductivity measurement (FIG. 83K), carried out using a collinear configuration (FIG. 83K inset), demonstrates an exceedingly high conductivity of 3580 S cm-1 stemming both from PEDOT’s ordered structure and homogeneous deposition.

Testing micro-supercapacitor performance

A PEDOT mSC (FIG. 84A inset) is tested using 1 M H2SO4 electrolyte (FIG. 98) by collecting cyclic voltammograms (CVs) with potential windows of 0.6, 0.8 and 1 V (FIG. 84A). These curves exhibit a symmetric rectangular shape due to reversible doping and dedoping during charging and discharging. Free movement of electrolyte ions is facilitated by the ordered nanofibrillar electrode architecture resulting in rectangular CVs under scan rates spanning 1 V/s to 50 V/s throughout 500 cycles (FIG. 84B). To identify charge transfer and ion diffusion processes, a mSC is cycled 500 times at scan rates of 25 mV s-1 and 50 V s-1 enabling electrochemical impedance spectroscopy (EIS) studies. Nyquist plots of real impedance Z (x-axis) versus imaginary impedance -Z” (y- axis) show similar equivalent series resistances (ESR) (x-axis intercept) and capacitive behavior (vertical line), regardless of scan rate (FIG. 84C). The high rate capability of our mSCs stems from strong mechanical adhesion between PEDOT and current collector enabling facile electron transfer. The absence of a semicircle at high frequency in FIG. 84C indicates a low charge transfer resistance as electrolyte readily accesses a nanofibrillar electrode.

Electrode thickness versus electrochemical performance is investigated by developing various PEDOT coatings. Nanofibrillar PEDOT coatings of 250 nm, 500 nm and 900 nm are synthesized from sputtered Fe2O3 layers of 60 mm, 120 mm and 180 mm, respectively (FIGS. 99 and 100). These electrodes lead to areal capacitances of 10.08 mF cm-2, 14.6 mF cm-2 and 21.3 mF cm-2, respectively, calculated from galvanostatic charge-discharge (GOD) measurements at a current density of 100 pA cm-2 (FIG. 101). Areal capacitance increases with polymer coating thickness, whereas volumetric capacitance decreases because of limited contact between electrolyte and active electrode surface area. This is shown by an increasing electrolyte resistance (IR drop) and decreasing coulombic efficiency of 99.5%, 97.6% and 94.8%. Notably, a 900 nm thick electrode is well adhered to the current collector and characterized by an areal capacitance of 21.3 mF cm-2, while a thick polymer coating (1 mm) exhibits poor adhesion and peels-off during sonication. An electrode thickness of 250 nm leads to a volumetric capacitance of 400 F cm-3. These figures of merit are the highest values among all reported conducting polymer-based mSCs. Cycling stability is evaluated via continuous charging and discharging cycles, at a current density of 100 pA cm-2, demonstrating 90% capacitance retention after 10,000 cycles for all devices regardless of thickness (FIG. 84E).

A Ragone plot representing the relationship between power density and energy density enables evaluation of energy storage metrics after normalizing figures of merit by volume or area. A 200 nm gap device coated by polymer (250 nm) exhibits a volumetric energy density of 16.1 mWh cm-3, and by increasing coating thickness (900 nm), an areal energy density of 1 .9 mWh cm-2 is achieved. These values represent the highest performance for conducting polymer-based mSCs surpassing Li-thin film battery metrics (FIGS. 84F and 102).

Engineering gap distance and fractal electrode geometry

Typically, decreasing the gap distance between electrodes results in lower ion diffusion resistance and higher capacitance due to shorter ion diffusion pathways. To evaluate the effect of gap distance we fabricate devices from a 60 nm Fe2O3 coating using 200 mm or 500 mm gaps (FIG. 85A). A device with shorter gap exhibits lower ESR and a more capacitive behavior as demonstrated by a vertical line in Nyquist plots (FIG. 85B) resulting in a 30% capacitance increase as calculated from CVs (FIG. 85C). Note that engineering a 100 mm gap is challenging because polymer coating at the gap/electrode interface limits contact between photoresist and lift-off solution (FIGS. 90, 103, and 104).

Fractal electrodes are predicted to enhance electrochemical performance of devices because they maximize 2D active surface area and ion diffusion pathways. Here, we demonstrate, for the first time, the effect of a fractal geometry in a 3D mSC architecture by designing interdigitated (L0) and fractal (L1) nanofibrillar PEDOT electrodes (FIG. 85D). These devices, possessing a 200 mm gap, and generated from a 60 nm Fe2O3 layer are characterized by ESRs of ~17 W and ~13 W, respectively (FIG. 86E). Unlike 2D electrodes, where ESR increases with complexity in electrode geometry, L1 possesses lower ESR than L0 due to its vertically directed nanofibrillar structure providing accessible ion pathways and resulting in a 10% capacitance increase (FIGS. 86F and 105). Our results highlight the synergistic convergence between fractal geometry and a 3D nanofibrillar electrode architecture, demonstrating low impedance, high capacitance and a cost-effective fabrication strategy.

Gel electrolyte and thermostability

Minimization of electrolyte leakage is of paramount importance for developing stable mSCs and is achieved using a gel electrolyte comprised of a viscous hydrated polymer matrix carrying free ions. Micro-supercapacitors, characterized by a 200 mm gap and 1 M H2SO4/polyvinyl alcohol (PVA) gel electrolyte (FIG. 86A inset), exhibit symmetric rectangular CVs due to facile charge transfer processes (FIG. 86A). Nyquist plots, however, show higher ESR and deviation from capacitive behavior (smaller slope inclination) (FIG. 86B) due to the electrolyte’s high viscosity that stifles ion diffusion. To evaluate the power performance of our quasi-solid-state devices, their frequency response is represented by a Bode plot (FIG. 86B inset) and analyzed via phase angle and relaxation time studies. Our device exhibits 2-4 times faster relaxation time (11 ms) than current state-of-the-art mSCs with a phase angle of 76°. This excellent power performance stems from an extended ordered nanofibrillar surface area as well as low electronic and ionic resistances resulting in a scan rate capability of 50 V s-1. Quasi-solid-state mSCs retain 94% of original capacitance after 10,000 cycles whereas liquid water-based devices retain 90% (FIG. 86C); a gel electrolyte prevents leakage thus enhancing cycling stability.

High-temperature stability is crucial for micro-devices since heat, inevitably generated during energy consumption, imposes thermal stress that leads to current leakage and decreases a device’s lifespan. Therefore, the effect of high temperature on quasi-solid-state mSCs is studied and FIG. 86D shows the experimental setup consisting of a heated aluminum block. The curve area for cyclic voltammograms expands proportionally as temperature rises from 25 °C to 60 °C and contracts drastically at 70 °C. At the latter temperature, charge storage decline stems from rapid electrolyte evaporation significantly increasing charge transfer resistance as detected by the appearance of a semi-circle and a 45° line (Warburg impedance) in Nyquist plots. Fortunately, chemically and physically stable nanofibrillar PEDOT coatings result in enhanced ion mobilities that boost capacitance in our devices by ~8% at 40 °C, -10% at 50 °C and -12% at 60 °C (FIG. 86E).

Methods

Materials. 3,4-ethylenedioxythiophene (EDOT, 97%), chlorobenzene (99%), hydrochloric acid (37%), sulfuric acid (98%), methanol (>99.8%) and polyvinyl alcohol were purchased from Sigma Aldrich and used as received. Highly n-doped 4-inch Si (100) polished wafers with a 10 W sheet resistance were purchased.

Fabrication of an interdiqitated PEDOT nanofibrillar electrode. 1) Micro- fabrication process. An insulating layer of Si02 (1.5 mm thick) was deposited on a 4-inch silicon wafer via plasma-enhanced chemical vapor phase deposition (PECVD). To generate a 4 mm thick double layer photoresist on Si02, photoresist LOR 10B (Mico Chem) was spun coated at 1500 rmm for 60 s followed by 10 min soft-bake at 195 °C. Subsequently, photoresist AZP 4620 (AZ Electronic Materials) was spun coated at 3000 rmm for 60 s followed by 2 min soft-bake at 115 °C. The photoresist etching was carried out using a Heidelberg Laser Lithography system. After etching, samples were developed in AZ 400 developer solution (AZ Electronic Materials) for 3 min resulting in an interdigitated electrode pattern. Metal layers comprised of 10 nm Cr (adhesion layer) and 50 nm Au (current collector) were sequentially deposited over the patterned photoresist layer via thermal evaporation (Edwards 306 Vacuum Coater). A solid-oxidant precursor, Fe2O3, was sputtered over the gold current collector via physical vapor deposition (Kurt J. Lesker PVD 75 RF and DC). 2) Rust-based vapor-phase polymerization. A glass reactor is loaded the Fe2O3- coated micro-electrode, and using glass vials, 40 mL of concentrated hydrochloric acid and 200 mL of a 0.674 M EDOT solution in chlorobenzene are also loaded. This rector is sealed and heated in an oven 150 °C / 1.5 hr. A PEDOT-coated substrate is immersed in Remover-PG (Mico Chem) solution and agitated on a shaker for 2 h to induce lift-off.

Spectroscopic characterization of electrode material. Scanning electron micrographs and energy-dispersive X-ray spectra were collected using a JEOL 7001 LVF FE-SEM. Transmission electron micrographs were obtained using a JEOL 2100 TEM by drop-casting a dispersion of PEDOT nanofibers on a TEM grid. Raman spectra were obtained using a Renishaw inVia confocal Raman spectrometer mounted on a Leica microscope with a 20* objective and 785 nm wavelength diode laser serving as an illumination source. A low power was necessary to mitigate heating of PEDOT sample. A Bruker d8 Advanced X-ray diffractometer was utilized to collect powder X-ray diffractograms of pulverized samples at room temperature using a Cu Ka radiation source (l = 1 .5406 A) and LynxEye XE detector (operating at 40 kV and 40 mA); the sample holder was rotated at 30 rmm with a scan step of 0.02°. Current-voltage (l-V) curves were obtained with a built-in-house 3D printed probe station using two gold needles 1.24 mm apart. A Bruker Multimode 8 Atomic Force Microscope was used to measure the films’ surface morphology and thickness. Four-point probe sheet resistance measurements were carried out using a Keithley 2450 SourceMeter with a Signatone SP4 four-point probe head.

Preparation of gel electrolyte. A 1 M H2SO4 aqueous electrolyte required milli-Q water (18 MW) degassed for 15 min. A 1 M H2SO4/polyvinyl alcohol gel electrolyte was prepared by adding 1 g of concentrated H2SO4 to 10 mL of deionized water, followed by addition of 1 g of polyvinyl alcohol powder. The whole mixture was heated to 85 °C while stirring until a clear solution was obtained.

Fabrication of PEDOT micro-supercapacitor. Platinum foil leads were connected to two gold pads using polyimide sticky tape, and 100 mL of a 1 M H2SO4 aqueous electrolyte were added to polymer-coated interdigitated electrodes by drop-casting. For a gel electrolyte, 100 mL of a 1 M H2SO4/polyvinyl alcohol mixture were added to interdigitated electrodes by drop-casting followed by drying at 55 °C / 2 h.

Electrochemical characterization of micro-supercapacitor. Cyclic voltammetry and electrochemical impedance spectroscopy were performed on a BioLogicVMP3 multi-potentiostat. Cyclic voltammetry was carried out from 25 mV s-1 to 50 mV s-1 between 0 V and 1 V. Electrochemical impedance spectroscopy was carried out at the electrode’s open circuit potential after obtaining a reversible cyclic voltammogram. Impedance values were recorded using a 10 mV sinusoidal disturbance at frequencies ranging from 100 kHz to 100 mHz.

Conclusion

The results of these experiments demonstrated the integration of chemical synthesis with conventional micro-fabrication producing a streamlined, superior and low-cost electrode engineering strategy for 3D nanofibrillar PEDOT mSCs. This technology produced 250-nm thick coatings of 3D vertically directed capacitive conducting polymer nanofibers characterized by an electronic conductivity of 3580 S cm -1. Our nanofibrillar PEDOT mSCs exhibit areal and volumetric capacitances of 21 .3 mF cm-2 and 400 F cm-3, respectively, and possess a state-of-the-art energy density of 16.1 mWh cm-3 in 1 M H2SO4 aqueous electrolyte. Quasi-solid-state mSCs retain 94% of their original capacitance after 10,000 cycles exhibiting high scan rate capability at 50 V s-1 and thermostability at 60 °C in 1 M H2SO4/PVA gel electrolyte. This work advances electrode characterization in situ and explores electrode thickness, gap distance and fractal geometry to control electrochemical performance. Nanofibrillar PEDOT mSCs are produced using a cost-effective and superior electrode engineering platform enabling in situ deposition of a 3D architecture bridging the energy density gap between micro-batteries and conventional micro- supercapacitors for miniaturized portable electronics.

EXAMPLE 12: SYNTHESIS OF SUBMICRON PEDOT PARTICLES OF HIGH

ELECTRICAL CONDUCTIVITY VIA CONTINUOUS AEROSOL VAPOR

POLYMERIZATION

The following example describes the data for methods of synthesizing scalable submicron-sized particles of the conducting polymer poly(3, 4- ethylenedioxythiophene) (PEDOT).

Current state-of-the-art synthetic strategies produce conducting polymers suffering from low processability as well as unstable chemical and/or physical properties stifling research and develomment. Here, we introduce a platform for synthesizing scalable submicron-sized particles of the conducting polymer poly(3,4-ethylenedioxythiophene) (PEDOT). The synthesis is based on a hybrid approach utilizing an aerosol of aqueous oxidant droplets and monomer vapor to engineer a scalable synthetic scheme. This aerosol vapor polymerization (AVP) technology results in bulk quantities of discrete solid-state submicron particles (750 nm diameter) with so far the highest reported particle conductivity (330±70 S/cm). Moreover, particles are dispersible in organics and water, obviating the need for surfactants, and remain electrically conductive and doped over a period of months. This enhanced processability and environmental stability enables incorporation in thermoplastic and cementitious composites for engineering chemoresistive pH and temperature sensors.

The fast-growing organic electronics industry will have an estimated worth of more than $75 billion by 2020. Manufacturing organic electronic materials that possess enhanced charge transport, as well as both high chemical and high physical stability, is paramount for developing commercial applications. Among coveted manufacturing technologies, production of solution processable conducting polymers ranks high given that these functional organic materials are characterized by light weight, flexibility, transparency, electronic conductivity, unique electronic band properties and reversible redox doping mechanisms. Currently, poly(3,4-ethylenedioxythiophene) (PEDOT) with polystyrene sulfonate (PSS) is the most common conducting polymer formulation used for aqueous- based applications due to surface-activating properties by PSS and PEDOT’s high intrinsic conductivity, polycrystallinity and environmental stability. Unfortunately, PEDOT:PSS suffers from both an unstable bulky polystyrene sulfonate dopant that phase separates in non-aqueous solutions and limits conductivity, as well as from an energy intensive multistep synthetic process.

Conducting polymers are typically synthesized using solutions, electrochemistry or vapors via synthetic oxidative strategies. Developing a continuous platform that produces functional materials is important to enable commercialization of these polymers for energy harvesting, 14 electrochemical energy storage, flexible electronics and biomedical applications. However, there is a pervasive lack of manufacturing technologies in the field with current synthetic protocols characterized by low yields and materials that are insoluble, poorly dispersible in water, and inherently chemically and/or physically unstable. The poor solubility of conducting polymers stems from a high surface energy and high lattice energy due to their conjugated backbone and ordered chain packing in the solid state.

A solid-state dispersion of polymer particles serving as a stable colloid is an attractive alternative for developing applications at liquid/solid interfaces. Among colloids, submicron- sized spherical particles are highly processable because their shape minimizes particle-particle interaction thus reducing aggregation. This provides building blocks for printing 2D and 3D hierarchical organic electronic architectures. Typically, spherical submicron-sized polymer particles are produced using templates such as hard particles or soft micelles. Removal of these templates is destructive to the polymer structure and is therefore impractical. In a few cases, “template-free” syntheses lead to particles via self-assembly. However, these approaches suffer from fixed stoichiometries that result in minimal opportunities for optimizing chemical and physical polymer properties. In addition, these synthetic methods cannot continuously produce PEDOT particles and result in particles of low electrical conductivity. Scalable technologies such as chemical aerosol flow synthesis, spray pyrolysis/drying and aerotaxy readily produce particles for inorganic materials. Among these, a modified version of the former, known as ultrasonic spray polymerization, is able to produce PEDOT particles; unfortunately, this process involves an uncontrolled solution oxidative polymerization that leads to particles of low conductivity (FIG. 54). An ideal approach to overcoming current limitations to produce particles of high conductivity would combine aerosols and vapor phase polymerization.

Here, we introduce aerosol vapor polymerization (AVP), a scalable continuous batch-processing technique that produces spherical submicron-sized PEDOT particles collected as a powder of high electrical conductivity. Particles exhibit extended chemical and physical stability characterized by a high doping level and surface charge that enables solution processing without the need for surfactants. AVP is based on a flow-stream reaction, between an aerosol of suspended water droplets carrying oxidizing ions and monomer vapor, where each reactant is delivered independently. We control stoichiometry, mass transport and aerosol residence time during vapor phase polymerization leading to a kinetically facile approach for producing PEDOT particles of high crystallinity, long conjugation length, and high doping stability. In AVP, polymer particles are synthesized in a tubular reactor, collected in ethanol-filled bubblers, purified in acid, and lyophilized from water resulting in a blue-colored powder comprised of discreet particles that remain doped for months. AVP-PEDOT particles, unlike PEDOT:PSS particles, are readily processed in non-aqueous systems to develop thermoplastic and cementitious solid-state composites as well as to engineer pH and temperature sensors.

Results and Discussion Polymerization mechanism and particle generation

In traditional vapor phase polymerization, a conducting polymer is produced by flowing a monomer vapor over a solid-state oxidant bed. In AVP, monomer vapor reacts with an aerosol of aqueous droplets carrying a soluble oxidant (FeCI3). Each of these two reactants is introduced independently into a reactor enabling in situ continuous replenishment. During polymerization, monomer vapor is oxidized upon contact with ferric ions dissolved in suspended water droplets resulting in the formation of monomer radicals at the droplet surface (FIG. 49A). Radical coupling promotes the assembly of oligomers and formation of the conjugated backbone via kinetically controlled deprotonation where water serves as the proton scavenger. Doping occurs in situ increasing charge carrier concentration and imparting a positive charge on the polymer chain that leads to the colloidal stability of particles. This charge in the polymer backbone is balanced by the oxidant’s chloride counter anion. In AVP, an oxidant-laden aerosol water droplet plays the role of template, proton scavenger, polymerization initiator and dopant (FIG. 49A inset). Each spherical particle is produced by preferential polymer nucleation at the droplet surface, and by radially-inward driven step-growth polymerization as a droplet evaporates.

An aerosol of oxidant droplets carried by nitrogen gas is generated by nebulizing a 0.266 M iron(lll) chloride aqueous solution using a 1.7 MHz ultrasonic transducer adapted from a household humidifier (FIG. 49B). A syringe pump replenishes the oxidant aqueous solution in situ as the reaction occurs. Monomer vapor is produced by bubbling nitrogen gas into a heated reservoir containing the liquid monomer 3, 4-ethylenedioxythiophene. Changing the size of this reservoir enables scale up. Mass transport of monomer vapor is controlled by attenuating the speed of the carrier gas using electronic mass flow controllers and by adjusting the monomer reservoir temperature. Synthesis is carried out in a coiled tubular glass reactor at 130 °C under laminar flow conditions that minimize droplet-droplet interactions and produce discrete polymer particles. PEDOT particles are collected as they exit the reactor in ethanol-filled bubblers; using an impactor or a filter paper for collection, leads to particle aggregation due to particle-particle collision. This synthetic approach results in a high production rate of discrete submicron-sized polymer particles. Elemental analysis and electrical properties of particles

Purified particles are lyophilized and collected as a blue powder (FIG. 50A left). Statistical analysis of particle diameter, aided by scanning electron microscopy, shows a size distribution range between 200 nm and 2 mm with a 750 nm mode (FIG. 50B). Particle size is governed by droplet size, and compared to an ultrasonic nebulizer, a collision nebulizer produces droplets with a broader size distribution (FIG. 55A) resulting in PEDOT particles with a broad distribution in size as well (FIG. 55B). A small number of particles exhibit distorted spherical symmetry possibly due to collapse of the polymer shell after removing iron residues in the core during purification (FIG. 50B inset and FIGS. 56A, 56B, 56C, and 56D). Elemental mapping via energy-dispersive X-ray spectroscopy shows carbon, oxygen, sulfur and chlorine present in doped PEDOT particles (FIGS. 56E, 56F, 56G, and 56H). X-ray photoelectron spectroscopy measurements (FIG. 56I) indicate atomic ratios of 7.17 and 2.87 for C:S and 0:S, respectively. These ratios are higher than the expected values of 6 and 2, possibly due to adventitious carbon and oxygen contamination. Fourier-transform infrared spectra (FIGS. 56J) display characteristic chemical bonds for the doped PEDOT conjugated backbone. A current-voltage profile for a pelletized sample demonstrates the polymer’s ohmic behavior (FIG. 50C) A disc-shaped pellet is formed (FIG. 50C top left inset) using 10 mg of particles by applying a hydrostatic pressure of 3,500 kg/cm2 for 2 min resulting in a compressed particle morphology as shown by scanning electron microscopy in (FIG. 50C bottom right inset). A four-point probe measurement for a pellet demonstrates a conductivity of 330±70 S/cm after HCI vapor doping. This high electrical conductivity stems from PEDOT’s polycrystalline structure, high doping level and plausible long conjugation length. To our knowledge, this is the highest reported electrical conductivity for a solid-state conducting polymer powder, shown summarized below in Table 3.

TABLE 3: Comparison of Existing Pelletized PEDOT Particle Conductivities

Note that a pellet’s conductivity differs from that of a continuous thin film owing to the many boundaries present that add resistance to conduction pathways. A single particle also exhibits ohmic behavior as demonstrated by the current- voltage plot collected using atomic force microscopy (FIG. 57). Particles, washed in acid and lyophilized, exhibit a zeta potential of +27 mV that provides colloidal stability and enables particles to remain suspended for 1 h by sonication in ultrapure water at pH 7 (FIG. 68A right). Solution processing of particles at the liquid/oil interface enables deposition of transparent coatings via surface tension driven flow. A particle-laden PEDOT film switches between a dark blue

(dedoped) and light blue color (doped) by exposure to vapors of base (dedoping) or acid (doping) (FIG. 59A); color change is reversible and controlled by charge carrier concentration as evidenced by UV-vis-NIR spectroscopy (FIG. 59B). A doped film is more transparent than a dedoped film because the latter exhibits a higher absorption in the visible range (380 nm - 740 nm).

Control of particle manufacturing

The step growth polymerization of PEDOT is systematically studied via aerosol residence time to control conjugation length, crystallinity and electrical conductivity. Our reaction is carried out in a 510-cm long coiled glass reactor (FIGS. 60A and 60B) at 130 °C using a total flow rate of 4000 seem resulting in a residence time of 20 s and a particle production rate of 100 mg/h. This total flow rate, a combination of the individual flow rates of monomer vapor (2000 seem) and oxidant aerosol (2000 seem), promotes interaction between reactants and results in a pelletized sample of high electrical conductivity (330±70 S/cm). Adjusting the total flow rate to 150 seem leads to the longest attainable residence time of 10 min that unfortunately, significantly decreases the particle production rate to less than 10 mg/h. Utilizing a straight short glass reactor (51 cm) (FIG. 60C) with a total flow rate of 4000 seem leads to a 2 s residence time resulting in a pelletized conductivity of -300 S/cm. However, this ultra-rapid synthesis leads to incomplete polymerization in the tubular reactor, sinters particles leading to poor solution processability (FIG. 61), and stifles interaction between reactants resulting in a low particle production rate (50 mg/h). In-line particle diameter measurements, carried out as particles exit the coiled glass reactor, show that the majority of particles have a diameter ranging between 350 nm and 750 nm; this reactor output stems from discreet nebulized aerosol droplets that polymerize inside the reactor (FIG. 50D red column); this reactor output stems from discreet nebulized aerosol droplets that polymerize inside the reactor (FIG. 50D red column). Measurements also show that particles of 350 nm diameter or less, bypass the ethanol-filled bubblers (FIG. 50D blue column). Decreasing the total flow rate to 600 seem shifts particle distribution to smaller sizes, centered at 350 nm in diameter (FIG. 62 red column), plausibly due to a greater influx of smaller nebulized droplets. Moreover, a smaller and narrower particle distribution is observed (FIG. 62 inset) indicating that ethanol-filled bubblers are effective at capturing particles of lower kinetic energies. Unfortunately, this lower flow rate, reduces particle production to -10 mg/h.

The mass transport of monomer vapor provides a chemical handle for controlling stoichiometry effecting a change in particle color that is directly related to conjugation length and electrical conductivity. To test mass transport, oxidant solution concentration, nebulization chamber liquid level, and oxidant delivery rate (88.6±5 mmol/min) are kept constant. The monomer vapor delivery rate is controlled by nitrogen carrier gas flow rate and by the temperature of the liquid monomer reservoir. Diffusion of EDOT monomer vapor in the carrier gas is thermodynamically activated by its enthalpy of evaporation (44.35 kJ/mol). To predict monomer concentration, a theoretical monomer vapor delivery rate curve is constructed assuming that EDOT’s partial vapor pressure equilibrates immediately upon contact with bubbling nitrogen gas (FIG. 51 A black curve). Experimental measurements show however, that the delivery rate is kinetically constrained by the contact time between nitrogen bubbles and liquid monomer. The experimental monomer delivery rate is lower than theoretically predicted both when using a slow (75 seem) (FIG. 51 A red points) or a fast (2000 seem) (FIG. 51 A blue points) nitrogen flow rate. This deviation from theory is proportional to flow rate given that at higher values, minimal diffusion of EDOT vapor in nitrogen bubbles occurs. A calibration curve enables our theoretical monomer vapor delivery-rate curve to predict accurate monomer concentration at any given flow rate. Concentration of EDOT vapor is also controlled by the monomer reservoir temperature between 25 °C and 90 °C under a constant total flow rate of 4000 seem. This results in the oxidant-to-monomer ratio decreasing from 7.5 to 0.12 and electrical conductivity decreasing from 330±70 S/cm to 1.22±0.22x10-3 S/cm (FIG. 51 B). In AVP, a 7.5 oxidant-to-monomer ratio produces a doped polymer of long conjugation length and blue particles whereas a ratio of 0.1 leads to purple particles comprised of short chained oligomers. Particle color is discernable during the first several minutes of collection enabling facile quality control of product. Extended heating of the monomer reservoir induces a darkening of the liquid monomer due to oxidation, with negligible effect on polymerization.

Chemical characterization of PEDOT

Polymer conjugation length, probed using Fourier-transform infrared spectroscopy (FIG. 51 C), shows the characteristic C=C stretch from the doped quinoid PEDOT structure at 1500 cm-1 and the C-O-C vibration peak at 1000 cm-1 from the ethylenedioxy group. Oxidative doping converts PEDOT’s structure from benzoid to quinoid leaving the ethylenedioxy group unaltered thus enabling characterization of the conjugation length by their relative ratios. As the monomer vapor concentration increases, the peak intensity of quinoid gradually drops with respect to ethylenedioxy due to the formation of oligomers of short conjugation length. Solid-state packing of particles, synthesized under various monomer vapor concentrations, is characterized by crystallinity determined via powder X-ray diffraction (FIG. 51 D). Three major peaks are present, the first two are assigned to (100) and (200) directions indicating edge-on packing; the third peak, assigned to (020), suggests face-to-face packing (FIG. 51 D, inset). Different degrees in conjugation length exhibit similar levels of crystallinity (FIG. 64) given that PEDOT’s tendency to crystallize is driven by its semi-rigid backbone and solid-state packing. Crystallinity therefore is a non-determinant contributor to conductivity among various samples. The chloride counter anion dopant in PEDOT’s backbone is probed using X-ray photoelectron spectroscopy (FIG. 51 E) by comparing CI:S ratio and using the Cl 2p peak normalized against S 2p. The area underneath the curve indicates doping level and PEDOT synthesized from a lower monomer vapor concentration exhibits a chloride to sulfur ratio of 30%, close to the theoretical limit of 33%. This high doping level on the conjugated backbone leads to a conductivity of 330±70 S/cm. X-ray photoelectron spectra for all samples show three doublets from three chloride environments with binding energies of 198.6 eV, 196.9 eV and 194.5 eV. These values are lower than that of covalent Cl-C bonding at 200 eV47 and are thus assigned to ionic doping Cl- moieties. Lower energies are assigned to mobile chloride ions and higher energies to less mobile chloride ions. The stifling of chloride ion mobility is caused by PEDOT’s high lattice energy and charge carrier interactions that prevent ion diffusion, this is also observed on conducting polymer films that exhibit minimal exchange of counterions. The low mobility for chloride ions indicates a deep and stable incorporation into the polymer backbone, resulting in high chemical and physical stability; PEDOT particles remain doped for months even after undergoing multiple lyophilization cycles (FIG. 65).

Solid state NMR of doped PEDOT

Solid state NMR is an ideal quantitative characterization technique for conducting polymers providing exquisite molecular and structural information; therefore, we conducted solid state 13C NMR to understand PEDOT’s backbone architecture and doping levels. NMR spectra are obtained with a 4-frequency spectrometer described in detail earlier. Each 100-mg sample consists of one part PEDOT and four parts sulfur ground and mixed uniformly to suppress macroscopic conductivity. The spectrum of a highly doped PEDOT (Cl- counterion) shows a narrow resonance at 80 pmm and two broad resonances, one centered near 150 pmm and the other near 50 pmm (FIG. 52A). We attribute the narrow peak to diamagnetic -CH20- carbons, and the 150 pmm broad peak to conductive carbons (FIG. 52A, insets). The conductive carbons support delocalized charge distribution which leads to broadening. The 150-pmm peak is assigned to conductive sp2 carbons directly bonded to oxygen or sulfur (the 50- pmm peak will be assigned later in this discussion). Of the two broad peaks, only the 150-pmm peak substantially survives interrupted decoupling (FIG. 52B). This Solid state NMR is an ideal quantitative characterization technique for conducting polymers providing exquisite molecular and structural information; therefore, we conducted solid state 13C NMR to understand PEDOT’s backbone architecture and doping levels. NMR spectra are obtained with a 4-frequency spectrometer described in detail earlier. Each 100-mg sample consists of one part PEDOT and four parts sulfur ground and mixed uniformly to suppress macroscopic conductivity. The spectrum of a highly doped PEDOT (Cl- counterion) shows a narrow resonance at 80 pmm and two broad resonances, one centered near 150 pmm and the other near 50 pmm (FIG. 52A). We attribute the narrow peak to diamagnetic -CH20- carbons, and the 150 pmm broad peak to conductive carbons (FIG. 52A, insets). The conductive carbons support delocalized charge distribution which leads to broadening. The 150-pmm peak is assigned to conductive sp2 carbons directly bonded to oxygen or sulfur (the 50- pmm peak will be assigned later in this discussion.) Of the two broad peaks, only the 150-pmm peak substantially survives interrupted decoupling (FIG. 52B). This result indicates that conductive sp2 carbons are not proximate to either intra- or interchain protons.

If the chloride counter-ion is exchanged by an acetate ion, the NMR spectrum permits an estimate of the degree of doping. The theoretical maximum doping level of PEDOT is 33%, in which there is one acetate methyl carbon for every six oxygen-substituted sp3 carbons, suggesting bipolaronic states throughout the chains (FIG. 52C). The observed ratio of integrated peak intensities at 20 pmm (methyl carbon) and 80 pmm (sp3 carbon) is about 1 to 12 (FIG. 52D). Thus, the PEDOT of panel c is approximately half doped at 15%. This doping level is lower than 30% as determined by X-ray photoelectron spectroscopy suggesting partial exchange of chloride ions with acetate ions. Possibly, this is the result of incomplete dedoping of chloride ions and sterically hindered acetate ions.

A short H C cross-polarization contact results in a spectrum arising from just protonated carbons. Two peaks are observed (FIG. 52E), one narrow at 80 pmm already assigned to -CH20- carbons, and the other broad, centered near 50 pmm. We assign the broad peak to -CH20- carbons near conductive loops formed by the proximity of two PEDOT chains (FIG. 52C, dotted lines and yellow highlight). We have pictured nearest-neighbor chains in separate planes, one above that of the paper (black) and the other below that of the paper (green).

The current flowing in the black part of the loop is left-to-right from the charge- insertion center, and the current in the green part of the loop is right-to-left (FIG. 52C, insert right). The induced magnetic field (Bi) for both parts is therefore opposed to the static field (Bo). This means that the Larmor frequency for - CH20- carbons near the conductive path (red dots in FIG. 52C) will be shifted to higher field, as observed in FIG. 52E. The Lamor frequencies of the sp2 carbons within the conductive paths are dominated by local electron density and are therefore broadened but not shifted.

Lower doping levels can be achieved by using acetate as the initial dopant. The observed ratio of intensities at 20 and 80 pmm for a lightly doped PEDOT (FIG. 52F) is increased by a factor of 4 relative to that of the PEDOT in FIG. 52D. For 15% doping, an increase of 100 would be expected because of 13C isotopic enrichment over natural abundance. Thus, the PEDOT of panel f has a doping of approximately 1%. This low level of doping means that some relatively narrow lines are observed (FIG. 52F) for non-conductive sp2 carbons directly bonded to oxygen (150 pmm) and sulfur (120 pmm), consistent with earlier assignments. The former survives interrupted decoupling but the latter does not (FIG. 52G). Much of the methyl-carbon peak survives interrupted decoupling because internal C3 motion decreases H-C dipolar coupling.

PEDOT particles as an additive for thermoplastic and cementitious composites

Unlike PEDOT:PSS, AVP-PEDOT particles afford high processability in organics and serve as stable colloids in common organic solvents (FIG. 53A). Thermoplastic composites are readily formulated using a polycaprolactone matrix and 25 wt% loading of PEDOT particles (FIG. 53B); these are homogeneously distributed as shown by scanning electron microscopy (FIG.

53B inset). Polycaprolactone is a flexible thermoplastic utilized as a cell substrate in biological applications and as a 3D printing filament in fused deposition modeling. The incorporation of particles leads to a percolation network rendering this film electrically conductive (FIG. 53B) and results in a sensor for detecting vapors of acids and bases. The current-voltage profile of this composite film shows ohmic behavior as well as reversible doping and dedoping in the presence of HCI and NH3 vapors, respectively (FIG. 53C). PEDOT particles are also easily processed in molten sulfur (FIG. 53D) for construction applications. Sulfur is the thermoplastic cementitious matrix in sulfur concrete, a promising structural material possessing tensile strength greater than that of traditional Portland cement-based concrete and is also acid resistant. Sulfur concrete is produced by mixing gravel, sand, molten sulfur and solidifies immediately upon cooling reaching maximum strength within 24 h after casting (FIG. 53E). A stifling limitation for the commercialization of this type of concrete lies in sulfur’s low melting point (120 °C) and its thermoplastic nature which leads to a loss of structural integrity at 88 °C. The addition of PEDOT particles to sulfur concrete enables engineering of a chemoresistive sensor that indirectly monitors temperature; the electrical resistance of PEDOT decreases with increasing temperature due to the promotion of charge carriers to the conducting band. A PE DOT-sulfur concrete composite contains PEDOT particles, sand, and sulfur; the addition of conducting polymer particles increases the weight of sulfur concrete by 10%. Mixing in a bath sonicator at 140 °C results in a solid-state electrically conductive composite upon casting with PEDOT particles homogeneously dispersed in the matrix (FIG. 53F). Two iron wires embedded in the composite, connected to a common handheld multimeter serving as electrodes (FIG. 53G) enable detection of electrical resistance as a function of temperature (FIG. 53H). The response to temperature is linear and reproducible over months due to the chemical and physical stability of PEDOT particles (FIG. 53I) providing a robust baseline for measuring the internal temperature of sulfur- concrete and detecting any ensuing structural deformation.

The results of these experiments demonstrated a scalable continuous synthetic approach to producing submicron-sized PEDOT particles of high conductivity (330±70 S/cm) by combining monomer vapor and an oxidizing aerosol. Our approach utilizes an independent feedstock of reactants where water droplets, produced from a common household ultrasonic nebulizer, serve as structure directing agents. We take advantage of fast polymerization kinetics from the vapor phase to produce exceptionally high chemical and physical stability in our particles which remain doped for months. AVP-PEDOT particles are discreet and collected as a solid-state powder that is ideal for 1) producing aqueous charge-stabilized colloids that obviate the need for surfactants and 2) developing semiconducting organic thermoplastic and cementitious composites. The functionality provided by these particles serving as additives enables engineering of robust pH and temperature sensors.

Methods

Reagents

Iron(lll) chloride (reagent grade, 97%), ethanol (200 proof, anhydrous) and 3,4-ethylenedioxythiophene (EDOT, 97%) were purchased from Sigma Aldrich and used without further purification. Ultrapure water was obtained from a Millipore filter (18.2 MW).

Particle Purification

An ethanol dispersion containing PEDOT particles was reduced using a rotary evaporator to a 50 ml_ aliquot and subsequently centrifuged in 1 M hydrochloric acid until the supernatant became colorless or light blue. Hydrochloric acid removes any iron from particles and dopes the polymer. Particles were then washed with water and centrifuged until a dispersion of pH of 7 was obtained. A purified dispersion is frozen and lyophilized resulting in a blue colored powder.

Microscopic characterization

Scanning electron microscopy and energy-dispersive X-ray spectroscopy data were collected using a JEOL 7001 LVF FESEM. Samples were dispersed in water and drop cast on gold coated polyimide tape. Off-line size distribution measurements were carried out using ImageJ.

Spectroscopic characterization

Fourier-transform infrared spectra of the dry powder samples were collected with a Bruker ALPHA Platinum-ATR. Powder X-ray diffraction spectra were collected using a Bruker d8 advance X-ray diffractometer at room temperature, with a Cu Ka radiation source (l = 1.5406 A) and LynxEyeXE detector, operating at 40 kV and 40 mA. Dry powders of each sample (20 mg) were doctor-bladed on a zero-intensity silicon wafer and the sample holder was rotated at 30 rmm/min using a scan step of 0.02°. Ultraviolet-visible-near infrared spectra were collected on a Cary 5000 UV-vis-NIR spectrophotometer using a parallel liquid cell. Solid powders were dispersed in 1 M HCI and 1 M KOH aqueous solutions and sonicated; data was collected in a quartz cuvette. X-ray photoelectron spectroscopy was conducted on solid samples using a PHI 5000Versaprobe II with an Al 1486.6 eV Mono-X-ray source at 51 .3 W, a beam diameter of 100-200 mm and a 1 V neutralizer at 15 pA.

Pellet fabrication and conductivity measurements

A pellet required 10 mg of PEDOT particles and was formed using a 13 mm diameter die by pressing under a hydrostatic pressure of 3,500 kg/cm2 for 2 min. The pellet thickness ranged between 30 mm and 40 mm and two 13 mm Teflon liner plates were used during pressing. Two-point resistance and current- voltage profile measurements were performed by attaching two leads to a sample using a collinear four-point probe station. Configuration (FIGS. 66A and 66B) required a Keithley 2450 Source and Measurement Unit and a Signatone four-point probe station. Conductivity was averaged over 10 measurements by rotating a pellet; detailed calculations for determining electrical conductivity were previously reported by our group.

Particle size and zeta potential measurements

A Portable Aerosol Spectrometer (GRIMM, model 11-C) was used to measure the in-line particle size distribution. Off-line size distribution was carried out using imageJ on 10 scanning electron micrographs and over 800 particles are measured. The zeta potential of a PEDOT particle dispersion was measured at pH 7 using a Malvern Zetasizer Nano series instrument.

Composite Fabrication

PEDOT-polycaprolactone composite required 25 wt% of PEDOT particles, 75 wt% polycaprolactone and trifluoroethanol mixed using sonication for 1 h. A homogeneous dispersion was drop cast on a silicone mat and air dried at room temperature for 6 h; the resulting composite film peeled off the silicone mat easily.

PEDOT-sulfur concrete composite required mixing 40 wt% sulfur powder, 50 wt% sand and 10 wt% PEDOT in a glass vial serving as a mold. This mixture was sonicated for 30 min in an oil bath heated to140 °C using a cartridge heater. The molten mixture was allowed to cool to room temperature and the solidified sample was removed from the glass mold.

Derivation of theoretical EDOT vapor concentration

EDOT at ambient temperature is a semi-volatile liquid and bubbling nitrogen through the liquid is a simple and effective method for delivering its vapor into a reactor. Based on ideal gas theory, the concentration is dependent on the partial pressure of a gas and the saturated vapor pressure of liquid and determined by the environmental temperature using the Clausius-Clapeyron Equation:

Literature shows that there are two saturated vapor pressures for EDOT reported at two different temperatures i.e. , 1) 20 °C = 0.04 Kpa,102) 90 °C =1.333 Kpa.

Then, AHvap is calculated as 44.35 kJ/mole and this is consistent with the predicted value from ACD/Labs Percepta platform, which is 42.9D3.0 kJ/mol.12

We assume EDOT is saturated in the carrier gas and therefore its concentration is: where, PEDOT and PN2 are the saturated vapor pressures of each gas at a specific temperature.

EXAMPLE 13: RED PIGMENT-DERIVED HIERARCHICAL THICK ELECTRODE FOR ORGANIC FLEXIBLE SUPERCAPA CITORS

The following example describes the data for methods of synthesizing scalable submicron-sized particles of the conducting polymer poly(3, 4- ethylenedioxythiophene) (PEDOT).

Increasing capacitance and energy density is a major challenge in developing supercapacitors for portable electronics. Thick electrode with high mass loading is necessary for the device to store a large amount of energy; however, it is typically restricted by mechanical stiffness and hampered ion diffusion that terminate flexible applications. Here, we show a precursor design of a conventional electrode synthesis for creating a hierarchical and mechanically flexible structure that increases mass loading by 15 times and areal capacitance by 17 times. We use a-Fe2O3 from a commercially available red pigment as an oxidant precursor for the oxidative radical vapor-phase polymerization of conducting polymer poly(3,4-ethylenedioxythiophene)

(PEDOT). Using a-Fe2O3-impregnated carbon cloth as a substrate, we produce state-of-the-art flexible nanofibrillar PEDOT-based supercapacitors characterized by high areal capacitance (2243 mF/cm2 for two-electrode vs. 6210 mF/cm2 for three-electrode) and high areal energy density (412 mWh/cm2).

A supercapacitor is an energy storage device featuring rapid charging- discharging capability, outstanding cycling stability and is characterized by lightweight and safety. [1] Unfortunately, this type of device suffers from low energy density that stifles its application for electric vehicles and portable electronics. [2] Current efforts to increase a supercapacitor’s energy density (E) are focused on modulating specific capacitance (C) or output voltage (V) as governed by the equation E = 1/2 CV2.[3, 4] Strategies that extend output voltage include asymmetric device engineering [3, 5] and using electrochemically stable electrolytes[4] whereas advancing electrode material and/or the cell architecture, increases capacitance. [6]

The emergence of technologies based on flexible and wearable electronics requires energy storage devices possessing both high energy density and mechanical flexibility.[1] For flexible supercapacitors, areal capacitance (Ca, mF/cm2) is the standard metric for evaluating energy storage using Ca = Csp x M / A, where Csp (F/g) is the specific capacitance of the active material and M / A (mg / cm2) is the electrode mass loading. [1 , 7] In the past few decades, efforts at increasing Ca share the common approach of targeting Csp by designing pseudocapacitive nanostructured materials.[2, 6] Improving M / A as an alternative for enhancing capacitance, is rarely discussed. In most cases, flexible electrodes are ultrathin films produced via self-assembly, [8] chemical vapor deposition, [9, 10] templated growth, [8, 11] electrodeposition[6] or drop- casting.^, 13] A low mass loading (< 1 mg/cm2) ensures a high material utilization yield however, the sparse amount of active material limits total energy stored and commercial applications which typically require a mass loading > 10 mg/cm2.[14] Increasing mass loading diminishes electrode mechanical stability, ion transport and electrical conductivity, making high flexibility and high mass loading a fundamental challenge. [7, 14] A recent report shows a high mass loading electrode is achieved by 3D printing hollow architectures that facilitate ion transports 5, 16] unfortunately, this electrode structure is rigid and impractical for flexible applications.

Here, we demonstrate a solid-state-precursor strategy for attaining flexible electrodes that increases mass loading (15 x) and areal capacitance (17 x) versus conventional liquid-precursor synthesized electrodes. Our approach produces a hierarchical electrode structure in one step resulting in a mechanically flexible supercapacitor of high energy density. This is achieved by depositing a nanofibrillar coating of the conducting poly(3,4- ethylenedioxythiophene) (PEDOT) via vapor phase polymerization on a carbon cloth substrate.

Typical vapor phase polymerization utilizes an oxidant solution such as FeCI3 on substrate to initiate oxidative radical polymerization of the monomer vapor 3,4-ethylenedioxythiophene (EDOT).[17-19] We previously demonstrated the hydrolysis of Fe3+ during vapor synthesis produces FeOOH nanospindles that guides PEDOT nanofiber growth resulting in electrodes of high surface area and enhanced electrical conductivity. [20-22] Our precedent work proves that common iron rust (comprised of iron oxides, oxyhydroxides and hydroxides) serves as an inert precursor for producing nanofibrillar PEDOT thin coatings on rigid electrodes. [23] In this work we show that a-Fe2O3 particles from a commercially available red pigment serve as a ubiquitous and cost-effective oxidant precursor that leads to flexible nanofibrillar PEDOT thick electrodes when supported on a carbon cloth substrate.

Our nanofibrillar PEDOT flexible electrode is characterized by a mass loading (30.2 mg/cm2) three times greater than the 10 mg/cm2 of commercial electrodes[14, 24] affording the potential for storing large amounts of charge.

The intertwined 3D hierarchical architecture in the electrode combines PEDOT nanofibers and carbon cloth micropores that increase surface area and provide a percolation network for conducting both electrons and ions.[7, 14, 25-27] This architecture is characterized by a layered carbon cloth-PEDOT structure that mitigates mechanical deformation where the inner core of a rectangular planar carbon cloth is void of polymer and characterized by a large free volume. Our flexible electrode shows state-of-the-art areal capacitance (three electrode measurement = 6210 mF/cm2), contributing to a symmetric bendable supercapacitor (2243 mF/cm2) possessing areal energy density of 412 mWh/cm2 using 1 M H2SO4 aqueous electrolyte. Notably, the areal capacitance and energy density reported here are the highest among flexible organic supercapacitors and exceed most current inorganic pseudocapacitors[1 , 8-13,

25, 26, 28-45]

Results and discussion

Synthesis of hierarchical nanofibrillar PEDOT -coated carbon cloth electrode. We show the difference between using an aqueous FeCI3 solution versus solid-state a-Fe2O3 particles as oxidative sources for synthesizing PEDOT on carbon cloth (FIG. 106). When FeCI3 solution is utilized, ferric ion serves as initiator, oxidant and dopant[46] to polymerize EDOT monomer at temperatures ranging from 50 oC to 150 oC (FIG. 106B).[17, 18, 20, 21 , 47] This synthetic approach requires a dilute (10-1 M) solution of FeCI3 to control hydrolysis and generation of FeOOH templates, resulting in a low PEDOT mass loading. [20-23, 47]

In this work, we replace the reactive and unstable oxidant solution (FeCI3) with an inert solid-state oxidant precursor (a-Fe2O3 particles) that continuously releases oxidizing ions (Fe3+) by dissolution of acid vapor (HCI) during synthesis (Figure 106 and Figure 111). This releasing mechanism buffers ferric ions directly in contact with EDOT enabling continuous Fe3+ hydrolysis and growth of PEDOT nanofibers using EDOT concentration (0.85 M) 10 x higher than solution-based synthesis. Processing of a-Fe2O3 particles (Figure 112) is carried out using a slurry to produce a high packing density of impregnated carbon cloth leading to a contiguous polymerization network within the spaces of the carbon fibers near cloth surface. The core of the carbon cloth electrode is void of polymer due to the low kinetics of static vapor diffusion during synthesis. [48] Purification of the polymer-coated electrode in concentrated acid solution removes all iron species resulting in a mechanically robust polymer/carbon composite.

Scanning electron micrographs reveal the morphology of carbon cloth before/after a-Fe2O3 impregnation and polymerization (Figure 107). Pristine carbon cloth (Figure 107a inset) is comprised of carbon fibers (~ 10 mm diameter) that assemble into bundles (~ 500 mm) weaving orthogonally between one another to form a mat (Figure 107a). A carbon mat is ~ 450 mm thick as shown by cross-sectional scanning electron microscopy (Figure 107b) with a mesh size of ~ 200 mm (determined via optical microscopy and shown in Figure 107b inset). Oxygen plasma treatment increases the hydrophilic character of carbon cloth and facilitates the impregnation of a-Fe2O3 particles (~ 200 nm diameter) (Figure 107c) between carbon fibers (Figure 107d). Impregnation changes the grey color of carbon substrate to red (Figure 107c inset) and effectively decreases the mesh size to ~ 100 mm (Figure 107d inset). Thermogravimetric analysis at 800 °C in air decomposes carbon completely demonstrating an 18 wt% a-Fe2O3 particle loading (Figure 113a,b). Our synthesis produces PEDOT nanofibers (~ 300 nm diameter) that deposit as a conformal coating with characteristic blue color becoming apparent after purification (Figure 107e). Cross-sectional scanning electron micrographs show that nanofibrillar PEDOT coatings (~ 100 mm thick) deposit simultaneously on both sides of a carbon cloth substrate resulting in the polymer/carbon/polymer electrode architecture (Figure 107f). The carbon cloth interlayer (250 mm thick) is void of polymer affording large free volume for mitigating mechanical deformation and the surface-localized PEDOT coating consists of soft semiconducting nanofibers ideal for flexible electrodes. [49, 50] Optical microscopy shows that a nanofibrillar polymer network effectively renders the carbon cloth mesh size to ~ 100 mm due to high mass loading of a-Fe2O3 particles during impregnation and structural support afforded by a-Fe2O3 particles for polymerization (Figure 107f inset).

For comparison, saturated FeCI3 solution is drop-casted on carbon cloth and tested as a substrate for vapor-phase polymerization. [46, 51] The electrode shows a dark blue color (Figure 107g inset) comprised of PEDOT nanofibers (~ 300 nm diameter) similar to a-Fe2O3-derived polymer, however nanofibrillar packing density is lower and coating is heterogeneous (Figure 107g) lacking polymer between carbon fibers (Figure 107h). The scarce presence of polymer mass bears minimal effect on mesh size resulting in an electrode with a mesh size similar to pristine carbon cloth (~ 200 mm) (Figure 107h inset).

The difference of decomposition temperatures between PEDOT and carbon cloth enable quantitative determination of PEDOT mass loading via thermogravimetric analysis (Figure 108a). Dopant loss commences at 120 °C and the polymer completely decomposes at 500 °C in air while carbon cloth remains stable up to 600 °C.[20, 22] Isotherm at 500 °C (1 h) provides complete subtraction of polymer mass for electrode mass loading calculation (Figure 113c, Supporting Information). The a-Fe2O3-derived PEDOT-coated carbon cloth shows ~ 70 wt% PEDOT content (mass of polymer and dopant / overall electrode mass) and a mass loading of 30.2 mg/cm2 (mass of polymer and dopant / electrode area), significantly higher than the ~ 15 wt% and 2.1 mg/cm2 from the FeCI3-derived electrode. Three-electrode characterization of PEDOT-coated carbon cloth electrodes. Our electrode exhibits an areal capacitance of 6210 mF/cm2 (206 F/g based on PEDOT’s mass) and a quasi-rectangular cyclic voltammogram at 2 mV/s calculated after removal of all iron species (Figure 108c). This exceedingly high areal capacitance originates from high mass loading, nanofibrillar morphology and high surface area. We illustrate the setup for three-electrode characterization in Figure 3b and methods section with information of all electrodes and devices summarized in Supporting Table 1 and Supporting Table 2. Step-by-step calculations are included in Supporting Calculations based on Supporting Equations. The rectangular shape of the cyclic voltammogram is retained at 25 mV/s scan rate and deformed to fusiform at 100 mV/s. A Nyquist plot shows real impedance Z’ versus imaginary impedance -Z” under a sinusoidal disturbance at the open circuit potential. Nyquist plots obtained via electrochemical impedance spectroscopy (Figure 108d red curve) pin point the cause for this deformation to fast scan rate that limits ion diffusion kinetics which is especially prevalent in thick electrodes. [26, 52, 53] The ~ 45° slope line linking the semicircle and low-frequency domain represents a Warburg region, characteristic of a tortuous ion diffusion path in a thick electrode. [54] Nyquist plots are also collected for bare carbon cloth and FeCI3-derived PEDOT-coated carbon cloth electrode for comparison of internal resistance (x-axis intercept) against a-Fe2O3-derived PEDOT-coated carbon cloth electrode. The a-Fe2O3- derived electrode shows a similar internal resistance to bare carbon cloth (~ 1.1 W) demonstrating intimate contact between polymer and carbon (Figure 108d black curve) whereas FeCI3-derived electrode exhibits a higher internal resistance (Figure 108d blue curve).

Cyclic voltammograms for carbon cloth and FeCI3-derived PEDOT electrodes show negligible (20 mF/cm2) and low (376 mF/cm2) capacitance respectively at 2 mV/s. At 25 mV/s, an a-Fe2O3-derived PEDOT electrode significantly exceeds both FeCI3-derived PEDOT and carbon cloth electrodes in capacitance (Figure 108e). Figure 108f summarizes the areal and gravimetric capacitances obtained from three-electrode cyclic voltammograms. At scan rates of 2, 25 and 100 mV/s, the superior capacitance of a-Fe2O3-derived electrode is clearly observed, higher than the FeCI3-derived electrode by ~ 17, 12 and 11 times in areal capacitance and by ~ 5, 4 and 3 times in total electrode’s gravimetric capacitance, respectively (Figure 108, top and middle row). The total electrode’s gravimetric capacitance is normalized by PEDOT mass plus carbon cloth mass in which carbon cloth serving as current collector and structural support provides negligible contribution to capacitance. When PEDOT mass loading is low, the carbon cloth’s mass dominates the total electrode’s gravimetric capacitance resulting in a value orders of magnitude lower than the intrinsic PEDOT’s gravimetric capacitance. [14] The high PEDOT mass loading of our a-Fe2O3-derived PEDOT electrode enhances the contribution of PEDOT mass to overall electrode mass and results in a high total electrode’s gravimetric capacitance.

The bottom row of Figure 108f shows PEDOT’s gravimetric capacitance normalized by PEDOT mass without considering carbon cloth mass. This parameter reflects the capacitive performance of PEDOT as an intrinsic active material and sets an upper limit of the electrode’s gravimetric capacitance enhanced by increasing mass loading. The PEDOT’s gravimetric capacitances of a-Fe2O3-derived and FeCI3-derived electrodes (Figure 108f, bottom row) are similar, originating from their analogous PEDOT morphology and conductivity as previously shown by microscopy and impedance analysis. The value of a- Fe2O3-derived electrode is slightly higher than FeCI3-derived electrode at 2 mV/s, possibly due to a better electrical contact between PEDOT and carbon cloth as previously confirmed via electrochemical impedance spectroscopy (Figure 108d). At higher scan rates (25 and 100 mV/s), the comparison result is reversed because the impeded ion diffusion in the thick electrode slows down the charging kinetics of a-Fe2O3-derived electrode, leading to a lower PEDOT’s gravimetric capacitance.

We compare total electrode’s and PEDOT’s gravimetric capacitances of both a-Fe2O3-derived and FeCI3-derived electrodes (Figure 108f, middle and bottom row respectively). The ratios of total electrode’s to PEDOT’s gravimetric capacitances are ~ 70% and 15% for a-Fe2O3-derived and FeCI3-derived electrode, indicating their total electrode’s gravimetric capacitances are dominated by capacitances from PEDOT and carbon cloth, respectively. Symmetric nanofibrillar PEDOT-coated carbon cloth supercapacitor. A flexible symmetric supercapacitor is developed using two rectangular a-Fe2O3- derived PEDOT-coated carbon cloth electrodes (12.70 mm x 3.175 mm x 0.45 mm, 17.25 mg each containing 12.18 mg PEDOT) and 1 M H2SO4 aqueous electrolyte. The electrode is coated with poly(vinyl alcohol) to enhance PEDOT adhesion (Figure 109a). We summarize detailed device geometries, mass loadings and electrochemical metrics in Supporting Table 1 and 2.

Cyclic voltammogram for a flexible PEDOT supercapacitor shows a device areal capacitance of 2096 mF/cm2 (at 2 mV/s) with rectangular shape (at 25 mV/s) characteristic of capacitive behavior (Figure 109b). This device exhibits state-of-the-art areal capacitance (2243 mF/cm2) and energy density (312 mWh/cm2) determined from a galvanostatic charge-discharge curve collected at 0.5 mA/cm2 (Figure 109c). Triangular curve symmetry is retained up to a current density of 50 mA/cm2 with coulombic efficiency near 100% demonstrating efficient charging-discharging rates. During repeated charging-discharging, ions move in (doping) and out (de-doping) of PEDOT, causing physical expansion- contraction that leads to electrode degradation. [14, 55, 56] Our carbon cloth- PEDOT sandwich electrode is characterized by an open architecture with free volume and nanofibrillar PEDOT structure that mitigates PEDOT volume change resulting in a device with high cycling stability. After 10,000 charging-discharging cycles, a supercapacitor retains -89% of original capacitance (collected at 25 mA/cm2) (Figure 109d). Capacitance increases by 10% after the first 1000 cycles because of improved wetting and ion diffusion[29] and slowly decreases thereafter possibly due to electrode degradation, electrolyte evaporation and loss of dopant.

We expand the voltage window from 1 V to 1.2 V to store more charges in the device and maximize the energy density.[3, 4] The upper limit 1.2 V is selected to prevent the hydrolysis of aqueous electrolyte generating H2 and 02. At 1.2 V, galvanostatic charge-discharge curves and cyclic voltammograms show triangular symmetry and rectangular shape at different current densities and scan rates, respectively (Figure 109e and 109f). Expanding voltage window causes an increased IR drop (Figure 109e inset) that decreases the amount of charge stored per unit voltage i.e., device capacitance (Figure 109g). However, the overall charges stored in our device is maximized, as shown by the increased energy (412 mWh/cm2 at 1 mA/cm2) and power (30 mW/cm2 at 50 mA/cm2) densities. A Ragone plot shows the superior areal energy density of our flexible supercapacitor versus state-of-the-art devices based on materials such as carbon allotropes, metal oxides as well as a metal-organic framework, MXene and transition metal dichalcogenide (Figure 109h).[1 , 8-13, 25, 26, 28-45]

Flexibility test and a tandem supercapacitor. We test the flexibility of our supercapacitor via mechanical bending from Oo to 180° while charging and discharging, showing consistent cyclic voltammograms at all bending angles (Figure 110a). Internal resistance in Nyquist plot decreases from 1.9 to 1.6 W as bending angle changes from 0° to 180° because device components are compacted during bending thereby enhancing electrical contact and ion transport (Figure 110b). We bend our supercapacitor repeatedly to examine the mechanical stability; after 500 bending cycles between Oo and 180°, our device shows a capacitance retention of 94% (Figure 110c).

We connect three supercapacitors in series and roll into a cylinder (2.25 mm radius) that shows a 3.6 V voltage window (Figure 110d inset) determined from a quasi-rectangular cyclic voltammogram. The tandem supercapacitor’s current decreases to one-third of a single supercapacitor because the internal resistance is tripled (Figure 110d). We visualize the energy storage performance of our tandem supercapacitor by lighting up a light-emitting diode. A 4.5 V external supply charges the tandem supercapacitor to 3 V in 5 s, enabling the supercapacitor to light up a white light-emitting diode for ~ 100 s with a “turn-on” voltage of 2.546 V (Figure 110e). We show the discharging profile in Figure 110f together with a Supporting Movie.

Conclusion

In summary, we present an electrode synthesis and engineering strategy to overcome the mechanical rigidity caused from high mass loading. The sandwich hierarchical structure of our electrode provides free volume in the interlayer and nanofibers on the surface to mitigate mechanical deformation meanwhile maximizes charge storage sites to ensure a high capacitance. We believe this work is a general solution for designing flexible high mass loading electrodes that will contribute to high energy flexible supercapacitors and batteries for powering flexible and wearable electronics.

Methods

Materials. Chlorobenzene (99%), 3,4-ethylenedioxythiophene (97%), poly(vinyl alcohol) (Mw 89, 000-98,000, 99+% hydrolyzed), methanol (>99.8%) and hydrochloric acid (37%) were purchased from Sigma-Aldrich; sulfuric acid (AR) was purchased from Macron. All chemicals were used without further purification. Platinum foil (0.025 mm thick, 99.9%) was purchased from Alfa Aesar and utilized for engineering electrode leads and Celgard 3501 membrane was used as a separator. The ELAT hydrophilic plain carbon cloth was purchased from FuelCellStore (College Station, Texas). Hematite particles (□- Fe2O3) produced by NewLook Inc. were purchased at The Home Depot Inc.

Characterization. Scanning electron micrographs were collected with a JEOL 7001 LVF FE-SEM. Thermogravimetric analysis was conducted on a Discovery TGA (TA Instruments). Cyclic voltammetry, galvanostatic charge- discharge measurements and electrochemical impedance spectroscopy were performed in a BioLogic VMP3 multi-potentiostat with Ag/AgCI reference electrode and Pt mesh counter electrode. A Pt current lead was attached to the PEDOT-coated carbon cloth with polyimide tapes for working electrode, leaving a square area (6.35 mm x 6.35 mm) exposed to 1 M H2SO4 electrolyte. For electrochemical impedance spectroscopy, the sinusoidal disturbance was 10 mV with frequencies scanned between 100 kHz and 0.1 Hz.

Synthesis of a-Fe2Q3-derived PEDOT-coated carbon cloth electrode. Carbon cloth (1 .27 cm x 0.635 cm) was treated with 02 plasma for 30 s and impregnated with 50 mL of a 0.25 g/mL aqueous dispersion of a-Fe2O3 particles; two glass plates were utilized to press particles into carbon cloth with ~50 N of force. This process was repeated thrice to achieve a high loading, excess liquid was removed in between pressings using Kimwipe before a coated substrate was dried at 50 °C for 0.5 h. Deposition of a nanofibrillar PEDOT coating was carried out in a 25 mL Teflon-lined stainless-steel hydrothermal autoclave at 160 °C for 14 h containing 200 mL of a 0.85 M EDOT in chlorobenzene solution and 20 mL of 11 M HCI. A PEDOT-coated carbon cloth was washed by sequential immersion in methanol (1 h) and concentrated hydrochloric acid (1 h) and repeated thrice at room temperature before three-electrode characterization.

Synthesis of FeCI3-derived PEDOT-coated carbon cloth electrode. We refer this synthesis to our previous synthesis on hard carbon fiber paper.[21] A carbon cloth (1.27 cm x 1.27 cm) was treated with 02 plasma for 30 s, coated by 1 ml_ saturated FeCI3 / nitromethane solution (~ 4.5 M) via drop casting, dried at ambient conditions for 12 h and transferred to a CVD reactor on a Si wafer substrate. A 150 mL water droplet was placed on the FeCI3-coated carbon cloth and 500 mL 0.0674 M EDOT / chlorobenzene solution was introduced into two glass reservoirs (250 mL each) in the reactor. After closing the reactor lid, the temperature was ramped from 30 °C to 130 °C with a rate of 600 °C / h and kept at 130 °C for 1 hour before taking out the substrate and purifying in concentrated hydrochloric acid and methanol.

Flexible supercapacitor fabrication. Two electrodes were soaked in 1 g/mL poly(vinyl alcohol) / 1 M H2SO4 solution at 70 °C for 12 h followed by ambient drying for 3 h. The electrodes were then swelled in 1 M H2SO4 solution at room temperature for 12 h prior to device assembling using Celgard 3501 separator, Pt current lead, 25 mL 1 M H2SO4 electrolyte and polyimide tape sealing.

EXAMPLE 14: BRICK-BASED DEVICES AND FABRICATION METHODS

The following example describes the data for brick-based devices that incorporate PEDOT coatings formed by any of the methods described above.

FIG. 116 shows a brick 102 containing one or more cavities 104, in which one of the cavities 104 contains a PEDOT-based device including any of the devices describes herein. In some aspects, the surrounding cavities 104 surrounding the device 108 may contain heating elements 106, including, but not limited to, resistive heating elements, to provide homogenous heating to facilitate the formation of the PEDOT-based device 108.

FIG. 117 and FIG. 118 are a close-up view and a cross-sectional view of a device 108 within a brick that is sealed by a cap 110 containing at least one opening 112 that may act as an inlet and/or outlet for introducing gases or liquids used to form the PEDOT-based device 108, or to introduce gases or liquids to be processed or used within the PDEOT-based device including, but not limited to, an electrolyte solution or electrolyte gel, a solution containing reactants to be catalyzed, or any other suitable liquid or gas composition. FIG. 119 is a schematic illustration of a method of forming an electrical device within an opening 104 within a brick that includes that formation of a PE DOT layer 114, removal of a portion of the PE DOT layer 114 to form electrically separated anodes and cathodes, fabrication of current collectors 118, fabrication of sealer/insulator materials between the anodes, cathodes, and associated current collectors 118, and insertion of a gel electrolyte 122.

FIG. 120 is a graph summarizing the performance of a brick-based water collection device comprising a PEDOT-coated brick.