WO2008057615A2

WO2008057615A2 - Highly transparent and conductive carbon nanotube coatings

Info

Publication number: WO2008057615A2
Application number: PCT/US2007/063343
Authority: WO
Inventors: Pual J. Glatkowski; David Alexander Britz
Original assignee: Eikos, Inc.
Priority date: 2006-03-03
Filing date: 2007-03-05
Publication date: 2008-05-15
Also published as: WO2008057615A3

Abstract

The present invention is directed to carbon nanotube (CNT) transparent conductive coatings and methods of improving the aggregate performance measure of resistance and transmittance by employing the electric field effect, chemical doping, nanotube alignment, and CNT linking via a metal atom, cluster, or nanoparticle.

Description

HIGHLY TRANSPARENT AND CONDUCTIVE

CARBON NANOTUBES COATINGS Reference to Related Applications

This application claims priority to United States Provisional Application No. 60/778,369 entitled "Methods of Forming Highly Transparent and Conductive Carbon Nanotube Coatings" filed March 3, 2006, and United States Provisional Application No. 60/780,284 entitled "Methods of Modifying Carbon Nanotubes Electronic Properties Coatings" filed March 9, 2006. The entirety of both provisional applications is specifically and entirely incorporated by reference.

Background

1. Field of the Invention

2. Description of the Background

Transparent conductors are essential components in many optoelectronic devices, including flat-panel displays, touch screens, electroluminescent lamps, solar cells and OLEDs. Transparent conducting oxides like indium tin oxide (ITO) have been a preferred choice for four decades. However, ITO has some significant limitations. ITO films are brittle and cannot be used for flexible display applications. Also, the procedure of fabrication of ITO coatings requires high vacuum sputtering followed by photolithographic etching; therefore the fabrication cost may be too high for high-volume or large area applications. Also, it is expected that ITO will become substantially more expensive due to closures of indium mines and reduction of indium supply.

Carbon nanotube dispersions can be prepared and then applied as a thin layer to form highly conductive transparent conductive films. Today, the highest quality CNT films results in 90-97% visible light transmittance and 50-500 O/sq sheet resistance or a volume conductivity of 5x10^Λ3 S/cm to 5 xlO^Λ4 S/cm - similar to the optoelectronic performance of sputtered ITO on plastic substrates. Such transparent CNT electrodes can be a viable alternative to ITO for many applications, officering ease of processing and cost effective technology for large area, flexible optoelectronic devices.

There is a strong demand to improve the resistance/transmittance (R/T) parameters of CNT coating to make it equal and even exceed the performance of ITO on a wide variety of substrates including glass and polymers.

U.S. Patent No. 6,139,919 to Eklund et al. relates to a method of doping single walled nanotubes (SWNT) with iodine by soaking SWNTs in molten iodine. The inventors describe soaking free-standing thick (>5 microns) mats of as-grown nanotubes in molten iodine, resulting in large decreases in resistance of the mat. The inventors disclose that the iodine can purify the SWNT, which would bring nanotubes into more intimate electrical contact, increasing inter-tube connectedness and increasing conductivity. In contrast, the present invention employs thin (<200 nm) films of nanotubes on substrates that have been previously purified and have very high connectivity. Eklund et al. show that iodine dopants leave the nanotubes at less than 100 degrees C, but do not disclose methods to trap the iodine at temperatures above 100 degrees C. Eklund et al. also demonstrate the doping method in a quartz vessel sealed under vacuum, which is not amenable to large scale production methods. Further, Eklund et al. limit the dopants used to iodine that is intercalated into nanotube bundles and into the interior cavity of the nanotube.

Luzzi et al. (U.S. Pat. No. 6,544,463 and 6,863,857) relates to hybrid materials wherein molecules are inserted into the interior cavity of carbon nanotubes. Luzzi limits his examples to the insertion of fullerenes C60 and La@C82. Although Luzzi indicates that the insertion of molecules into nanotubes could dope carbon nanotubes to change electronic properties, Luzzi does not measure any electrical properties on individual nanotubes, nor of mats or films on nanotubes. It has since been shown that the insertion of fullerenes into nanotubes can leave nanotubes essentially unaffected in electronic performance and properties. Luzzi's methods of inserting molecules into nanotubes are similar to that of Eklund, but are generally conducted at 325 degrees C and above and at sealed under high vacuum. Further, Luzzi only describes insertion of molecules that are in the solid phase into carbon nanotubes.

Green et al. (U.S. Pat. No. 6,090,363) relates to methods of opening and filling multi-walled nanotubes from solution. However, Green et al. do not discuss the possibility of changing nanotube electronic properties with the encapsulation of materials in the nanotube cavity. Further, Green et al. exclusively use multi-walled carbon nanotubes, which have a much larger interior cavity than single walled nanotubes. The larger cavity makes multi- walled CNTs easier to fill than single walled CNTs. Additionally, Green et al. do not describe a method to close multi-walled nanotubes, which limits the usefulness of their invention for some kinds of dopants. Moreover, multi-walled nanotubes are metallic and therefore will not have significant changes in conductivity in the presence of dopant materials.

Fan et al. (U.S. Pat. No. 7,029,751) relates to an isotope doped carbon nanotube having carbon- 12 and carbon- 13 isotopes. Fan et al. relates to methods of incorporating these atoms into the nanotube using different growth techniques. This type of doping can be considered a type of substitutional doping, where a normal carbon atom is replaced with a different atom. While Fan et al. indicate that non-carbon dopants could be incorporated into the nanotube, as well., they do not disclose or suggest the effect that the addition of dopant atoms could have on electrical conductivity. Specifically, they do not disclose or suggest that substitutional dopants improve the electrical conductivity of carbon nanotubes, nor do they suggest or disclose that substitutional dopants improve the conductivity of a carbon nanotube film or a carbon nanotube network.

Siddons et al. (Nano Lett, 4, 927, 2004) relates to methods to gate and dope nanotubes using polymer electrolytes polyethylene imine, polyethylene oxide, and polyacrylic acid. According to Siddons, these polyelectrolytes were deposited onto individual semiconducting SWNTs between two electrodes to determine their effect on conductance through a single nanotube. Additionally, solutions of the polymers were blended with SWNTs, and cast films were optically measured to determine doping. In the individual nanotube case, the Siddons et al. address how dopants affect conductance on an individual nanotube. The dopant can affect the nature of the nanotube-metal contact, which can change the apparent conductivity of the individual nanotube. Further, Siddons et al. use polymer electrolytes as top gates for nanotube field effect transistors. In contrast, in the present invention, it was surprisingly discovered that nanotubes with dopants, including polymers, maintain surface conductivity, which would electrically short out a top gated device because the present invention maintains surface conductivity of the nanotube film. The present invention's dopant coatings are significantly thinner than those described by Siddons et al., relative to the nanotube thickness, hi the case of the nanotubes blended in the polymer, no electrical measurements are presented to directly correlate doping with changes in conductivity for networks. Furthermore, the polymer to as-received nanotube weight ratio was found to be 1000:1, or 0.1%, which is very close to or below the percolation threshold, depending on nanotube length and purity and the polymer host. Conversely, Siddons et al. used Carbolex as-received nanotubes, which are typically about 40% carbon nanotubes by weight, indicating a nanotube to polymer weight ratio of about 0.04%, which is generally accepted to be below the percolation threshold, though with some exceptions for emulsion-nanotube blends. Thus, the samples are likely to have no bulk conductivity and do not represent a carbon nanotube network. The samples and methods described by Siddons et al. do not disclose or suggest the effects of dopants on the transparency or conductivity of thin carbon nanotube networks consisting of a plurality of nanotubes in electrical contact with each other.

Ozel et al. (Nano Lett, 5, 905, 2005) disclose using polymer electrolytes as a gate material for single walled nanotube network field effect transistors. Ozel describes how increasing the channel length from 0.5 microns to 100 microns (and channel width of 250 microns) improves the transistor characteristics. 100 microns is the largest fabricated channel length, compared to a distance of greater than 10 mm for the present invention. At short channel lengths, the devices tested show metallic behavior, i.e. an on/off ratio of 0. This observation of field effect transistor behavior is different to that seen by the present inventors, who find that distances between electrodes of >10 mm result in metallic behavior (linear I-V characteristics) and an on-off ratio of 0 (i.e. always on) of the nanotube network. Linear I-V characteristics and a 0 on-off ratio are good characteristics of a resistive metal, but are poor characteristics for a field effect transistor. Further in contrast from the present invention, Ozel et al. use polymer electrolytes as top gates for nanotube field effect transistors. In the present invention, it is surprisingly discovered that nanotube films with dopants, including polymers, maintain surface conductivity, which would electrically short out a top gated device because the present invention maintains surface conductivity of the nanotube film. Therefore, the dopant coatings of the present invention are significantly thinner than those described by Ozel et al. Moreover, Ozel et al. do not demonstrate or even suggest on the optical effects of polyelectrolytes on the nanotube networks. All of Ozel's samples are on silicon wafers, which could only allow for transparency in the infrared region.

Wu et al. (Science 305, 1273, 2004) disclose a transparent nanotube film. The authors comment that processing nano tubes leads to doping of the nanotubes in the film. The authors later remove the dopant by heating films to 600 degrees C. The doping process of Wu et al. is uncontrollable because it occurs concomitantly with purification, rather than in a separate process. This work is distinct from the present invention because Wu et al. do not intentionally dope the nanotubes. This work is also distinct from the present invention because Wu et al. dope unpurified nanotubes prior to forming a film. The method disclosed by Wu et al. has the disadvantage of possibly affecting nanotube dispersability, processing, and stability. Further in contrast from the present invention, Wu et al. do not disclose or suggest the location of the dopant. The work of Wu et al. differs from the present invention further in that Wu et al. only demonstrate dedoping, not doping. Doping nanotube networks is significantly more useful than dedoping because doping increases the conductivity of the nanotube transparent conductor. Further, the dedoping process is demonstrated only by optical measurements, not with electrical measurements. The present invention discloses that doping nanotube films has a direct effect on sheet resistance, conductivity, and nanotube optical properties. The present invention also discloses doping purified nanotubes, which is more advantageous and uses less doping material than doping an unpurified nanotube sample. This invention also discloses methods to dope nanotube networks or nanotube coatings in a controllable way that is independent of the purification process.

Thus, there is a need to improve the resistance/transmittance (R/T) parameters of CNT coating to make it equal and even exceed the performance of ITO on a wide variety of substrates including glass and polymers. Using a variety of techniques, many of which can be used additively, the present invention discloses methods to meet these RT performance goals. Further, approaches are described targeting one embodiment, R/T performance of 95% and 10 O/sq. While a number of advances have been made in the realm of CNT coatings, none have achieved the parameters or disclosed or suggested the methods of the present invention.

Summary

The present invention overcomes the problems and disadvantages associated with current strategies and designs and provides new tools and methods for improving the aggregate performance measure of resistance and transmittance by employing the electric field effect, chemical doping, nanotube alignment, and CNT linking via a metal atom, cluster, or nanoparticle.

One embodiment of the invention is directed to methods for forming coatings containing carbon nanotubes with stable electrical properties comprising doping the carbon nanotubes with a dopant.

Another embodiment of the invention is directed to a method of increasing the conductivity of a CNT-containing film comprising doping the carbon nanotubes with a dopant before, during or after forming the film.

Another embodiment of the invention is directed to methods for doping CNT films without significantly altering the surface conductivity of the film.

Another embodiment of the invention is directed to methods for doping films without electrically insulating the surface of the film upon doping.

Another embodiment of the invention is directed to methods for doping films without significantly altering the thickness of the film upon doping. Thickness is increased by less than 100%, less than 75%, less than 50%, or less than 10% in various embodiments.

Another embodiment of the invention is directed to methods for improving, varying, or stabilizing electrical properties of CNT coatings or films, wherein the electrical properties comprise resistance, conductivity, transparency, etc. Transparency of films of this invention is greater than 60%, greater than 70%, greater than 80%, greater than 90%, greater than 95%, or 100% or nearly 100%. In some embodiments of this invention, transparency is altered upon doping with methods of this invention by less than or equal to 10%, less than or equal 5%, less than or equal to 2%.

Another embodiment of this invention is directed to films or methods wherein the sheet resistance of the film decreases after doping. Sheet resistance in embodiments of this invention is decreased by greater than 5%, greater than 10%, greater than 20%, greater than 50%, greater than 100%, greater than 200%, or greater than 500% as measured in ohms/square.

Another embodiment of this invention is directed to methods wherein volume conductivity of a CNT film increases after doping. Volume conductivity in various embodiments is increased by greater than 5%, greater than 10%, greater than 20%, greater than 50%, greater than 100%, greater than 200%, or greater than 500%.

Other embodiments of the invention are directed to methods for reversibly or irreversibly doping CNTs.

Another embodiment of the invention is directed to single-walled, double-walled, few-walled or multi-walled carbon nanotubes.

Another embodiment is directed to methods for doping a CNT-containing film wherein the doped film is stable upon exposure to an ambient environment. This stability can also be exhibited with exposure to temperatures ranging from -200 to 600 degrees Celsius, and/or upon exposure to humidity from 0 to 100%. In various embodiments of this invention, stability of the doped film is directly proportional to the stability of the dopant.

Another embodiment of this invention is directed to methods comprising patterning of CNT coatings by doping to form variable electrical conductivity across the coating.

Another embodiment of this invention is directed to CNTs which function as solid electrolytes. Another embodiment of this invention is directed to methods for forming solid electrolytes comprising adding a dopant to the interior of carbon nanotubes. In embodiments of this invention, the dopant is a monomer that is polymerized. In various embodiments of this invention, a solid electrolyte is formed by filling carbon nanotubes with dopants, preferably wherein the dopant is introduced in a saturating amount to substantially fill said carbon nanotubes

Another embodiment of this invention is directed to a composition comprising a CNT-containing film that is electrically conductive and optically transparent wherein CNTs within the film are doped with a super-acid, a polymeric acid, a metal oxide, a polymer, a fluoropolymer, a particulate metal, or a combination thereof.

Another embodiment of this invention is directed to a method for forming a CNT- containing composition comprising forming a thin layer of CNTs on a substrate of soda lime glass and heating the CNT layer on the substrate such that alkali ions of the substrate dope the CNTs forming an alkali-doped CNT-containing film. In certain further embodiments, the alkali doped carbon nanotube-containing film has a resistivity of less than 10^~3 ohms-cm, of less than 10^"4 ohms-cm, or of less than 10^~5 ohms-cm. In further embodiments, the alkali-doped CNT-containing film has a transparency of greater than 60%, greater than 70%, greater than 80%, greater than 90%, greater than 95%, or nearly 100%.

Another embodiment of the invention is directed to increasing the conductivity of a CNT-containing film, comprising forming the film and doping the CNTs. Embodiments of this invention comprise methods wherein forming the film de-dopes the CNTs.

Another embodiment of the invention is directed to a carbon nanotube film comprising carbon nanotubes that are doped with a dopant, wherein the film is conductive and transparent, and stable to temperature changes.

Another embodiment of this invention is directed to doped CNTs in solar H2 generation. Another embodiment of this invention is directed to doping CNT networks during purification steps or during processing or when the CNTs are in ink/solutions.

Another embodiment of this invention is directed to doping CNT networks not only during purification steps or during processing or when the CNTs are in ink/solutions.

Another embodiment of this invention is directed to methods for doping CNTs without changing transparency. Preferably, in one embodiment of this invention, R is reduced by 10-100%, while T remains unchanged.

Another embodiment of this invention is directed to doping CNTs to increase transparency while increasing conductivity, especially near the IR range.

Another embodiment of this invention is directed to doping with a reduction in transparency while conductivity is increased, especially in the narrow parts of the spectrum.

Another embodiment of this invention is directed to methods for doping in highly purified CNTs, preferably greater than 70% pure, more preferably >90%, even more preferably >95%, most preferably >98% or >99%.Another embodiment of this invention is directed to doping CNTs to modify work function, to either increase or decrease work function of CNT-containing compositions to electronically match materials of a device.

Another embodiment of this invention is directed to methods of forming a coating comprising separating two transparent and conductive CNT layers with a dielectric material and enhancing resistance of the CNT layers by applying voltage between the two CNT layers.

Another embodiment of this invention is directed to transparent, electrically conductive coatings comprising carbon nanotubes with a diameter less than 5 nm and another component that decreases the sheet resistance of the carbon nanotube coating.

Another embodiment of this invention is directed to transparent, electrically conductive coatings comprising carbon nanotubes with a diameter less than 5 nm and another component that increases the volume conductivity of the carbon nanotube coating. Another embodiment of this invention is directed to a transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component that increases the conductivity of the carbon nanotube coating without affecting absorbance of light in the visible region

Another embodiment of this invention is directed to a transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component that increases the concentration of positive charge carriers in the nanotube coating.

Another embodiment of this invention is directed to a transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component that increases the concentration of negative charge carriers in the nanotube coating.

Another embodiment of this invention is directed to an electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component that changes the work function of the nanotube coating

Another embodiment of this invention is directed to a transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component, wherein the other component forms a redox couple with the carbon nanotube.

Another embodiment of this invention is directed to a transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component, wherein the composite is a solid electrolyte

Another embodiment of this invention is directed to a transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component, wherein the other component is located inside the carbon nanotube interior cavity and the other component improves the optoelectronic properties of the carbon nanotube coating

Another embodiment of this invention is directed to a transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component, wherein the other component is located in the interstices of the carbon bundle and the other component improves the optoelectronic properties of the carbon nanotube coating

Another embodiment of this invention is directed to a transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component, wherein the other component is trapped inside the carbon nanotube interior cavity by functional groups on the nanotube.

Another embodiment of this invention is directed to a transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and another component, wherein the other component is trapped in the interstices of the carbon bundle by functional groups on the nanotube.

Another embodiment of this invention is directed to transparent, electrically conductive coatings comprising carbon nanotubes with a diameter less than 5 nm and another component, wherein the other component at least partially forms a coating around the nanotubes and decreases the sheet resistance of the nanotube coating

Another embodiment of this invention is directed to methods of doping transparent conductive nanotube film wherein the carbon nanotube film is exposed to a dopant.

Another embodiment of this invention is directed to methods of doping transparent conductive nanotube films wherein the carbon nanotubes are exposed to a dopant prior to being formed into a film.

Another embodiment of this invention is directed to methods of doping transparent conductive nanotube film wherein the carbon nanotubes are exposed to a dopant during purification.

Other embodiments of this invention are directed to CNT layers, coatings or films formed by any of the methods herein.

Other embodiments and advantages of the invention are set forth in part in the description, which follows, and in part, may be obvious from this description, or may be learned from the practice of the invention.

Description of the Figures

Figure 1: Resistance vs. time for a glass slide coated with CNT, doped with OA, then monitored at ambient conditions. R_Sjl = 224 Ohm/sq, R_{S f} = 184.5 Ohm/sq over 2.8 days. Figure 2: Vis-Near IR spectra of intrinsic SWNT film (red) and OA doped SWNT film (blue). Note the suppressed Sn transition for the OA doped film.

Figure 3: Resistance vs. time for a glass slide coated with CNT, doped with OA, coated with Teflon AF binder, then monitored at ambient conditions. R₅,; = 227.5 Ohm/sq, R_s,f = 203 Ohm/sq over 8 days.

Figure 4: Resistance vs. time for a CNT film dipped in a solution of iodine in toluene.

Figure 5: Vis-NIR spectra of a SWNT film before Nafion coating (blue) and after Nafion coating at dip speed 0 (red). Note the apparent change in SWNT plasmon background is due to thickness effects (Fabry-Perot fringes) from the Nafion. Figure 6: Depiction of one embodiment of the invention.

Description of the Invention

As embodied and broadly described herein, the present invention is directed to novel methods for improvement of R/T performance of CNT transparent conductive coating, including but not limited to (a) electric field effect, (b) CNT chemical doping, (c) CNT alignment, and (d) CNT linking via a metal atom, cluster, or nanoparticle. The present invention also describes how a nanotube film, coating, paper, or network can be used as a solid electrolyte by using methods relevant to doping.

One embodiment of the invention provides enhanced conductivity and transparency by deposition of two conductive transparent CNT layers on both sides of a transparent substrate (which can be flexible). The resistance of CNT sheet on the one side can be controlled by an applied gate bias (voltage between two CNT layers). This design is similar to field effect transistors (FET) broadly applied in the optoelectronics and electronics industry. In particular, an FET including a transparent substrate and CNT electrodes has been described in the patent application [3]. However, in the present invention, the electric field effect is not employed to build a transistor device, but to enhance the conductivity of CNT transparent film. The field effect can provide a significant improvement of R/T performance of 10-1100% and reach a coating as high as 95% transmittance at 10 O/sq sheet resistance. One embodiment of the invention provides enhanced conductivity of the semiconducting nanotubes in the transparent coating. The enhancement is a consequence of changing the number of free carriers in the semiconducting CNTs by doping the CNTs. Doping is used in a broad sense: it is an addition of an impurity or impurities to another substance or mixture of substances, preferably a carbon nanotube film, in a controlled manner that produces desired properties. CNTs can be involved in a redox reaction to give either hole or electron conductivity, or carbon atoms in the nanotube sidewall can be replaced with other atoms to cause electron or hole conductivity. Preferably, the dopant is any electron acceptor (a Lewis acid), such as iodine, sulfuric acid, or p-toluene sulfonic acid, nitric acid, or Nafion. Also, the dopant is preferably any electron donor, such as sodium metal, ammonia, amine, or cobaltacene, or polyethylene imine. Dopants can also be any of the following: binders, polymers, acids, metal oxides, salts, slemion, thionyl chloride, TCNQ, oxygen, water, fluoropolymeric acids, polystyrene sulfonic acids, phosphoric acids, polyphosphoric acids, polyacrylic acids, any superacids, combinations of any dopants, etc. The dopant is preferably located inside the carbon nanotube to decrease the dopant optical cross-section and to shield the dopant and the dopant reaction products from further reaction with the environment. Dopants and dopant reaction products that do not absorb radiation in the visible region are preferred to dopants with absorbance in the visible region. Since transparency and other optical properties of the CNT film are dominated by the void space, rather than the optical properties of carbon nanotubes, changes in properties, such as refractive index, by doping will have a negligible effect on the bulk optical performance of the film. Finally, the dopant is preferably an atom that substitutes a carbon atom in the nanotube sidewall, such as nitrogen, boron, silicon and metals.

Dopant concentrations can range from 0.001 atomic percent per carbon atom to 50 atomic percent per carbon atom in the nanotubes sidewall. Substitutional doping does not substantially alter the optical transmission of CNTs but significantly increases the electrical conductivity of the individual nanotubes that are subsequently used to form the coating. Furthermore, substitution of a plurality of carbon atoms in the nanotubes can contribute to reduced tube to tube electrical resistance greatly enhancing the overall electrical conductivity of a collection of the nanotubes used to form a coating. Lastly, sidewall substitution will permit side wall functionalization at the site of substitution. More specifically the sites of substitution can be chemically reacted with molecules to enhance solubility, allow tube to tube cross linking, attaching the tubes to a substrate or other matrix material mixed with the nanotubes. All these processes enable the tubes to increase optical and electrical performance of the coating.

One embodiment of the invention improves conductivity by reducing the resistance between nanotubes in a bundle and between bundles. Conductivity is improved by creating a localized metallic contact between adjacent CNTs. hi a preferred embodiment, metallic nanoparticles, such as 5 ran nanogold, will co-precipitate with the carbon nanotubes as they dry from a fluid and will preferentially aggregate at the intersections of bundles. In another preferred embodiment, the particulates are added after the conductive nanotubes network forms by wetting the nanotubes layer with a dilute solution of particulates. The aggregation occurs at the intersection to minimize surface energy making it the last area to dry. Also, the metallic contact comes in the form of a metal coordination complex, where a transition metal atom or a metallic cluster is bonded to the nanotube sidewall. These methods of improving CNT electrical contact substantially alter the transmittance, since very little material is used and it is concentrated at the intersection of tubes and bundles of tubes. However, in sufficient quantities, this embodiment offers the ability to alter nanotube optical properties by strong electronic interactions between metals and nanotubes.

One embodiment of the invention provides enhanced conductivity, without detriment to optical transmission of light, by the orthogonal alignment of CNT ropes. In this case, the same CNT mass provides higher conductivity than randomly oriented CNTs. The mass of CNTs is proportional to optical absorbance (Beer's Law) for a given purity of nanotubes using the same deposition method. It has also been surprisingly discovered that dramatic differences in conductivity at the same optical absorbance can result from different deposition methods or different levels of sample purity. Different deposition methods will lead to films with different morphologies. In an analogy to other conductors, higher degrees of crystallinity will lead to better conductivity because of a reduction in electron scattering sites. The same idea of increasing order (i.e. crystallinity) will lead to increased conductivity. Order can be increased, for example, by thermally annealing CNTs to increase the bundle crystallinity. Alignment of nanotubes and bundles of nanotubes increases order over a longer length scale relevant to macroscopic conductivity. Alignment is achieved by means of imparting a force on the CNT film as it is forming. Centripetal force, shearing of a fluid, surface tension, electric field, or magnetic field cause CNTs to align. In addition, CNT alignment in one direction results in an enhanced conductivity in this direction and diminished conductivity in the orthogonal direction. Anisotropy of conductivity is accompanied by optical anisotropy, i.e. the ability of the CNT film to polarize light.

In one preferred embodiment of the invention, the interior cavities of carbon nanotubes are filled with the material or materials. Preferably, the CNT composite acts as a solid electrolyte comprising encapsulated or intercalated anions and nanotube cations. Embodiments of the present invention describe a material or materials added to carbon nanotubes that alter the electronic properties of the nanotubes and layer comprising the nanotubes. The electrical conductivity of the CNT composite is improved, relative to CNT without added material. The addition of material improves the performance of a device comprising conductive carbon nanotube layers or coatings. The type and quantity of encapsulated material allows work function control of the CNT layer to improve the performance of a device. Energy conversion efficiency of a photovoltaic device under varied sun and indoor light conditions is improved by using a CNT composite. The CNT composite is designed to replace liquid, solid, or gel electrolytes employed in dye sensitized or Graetzel solar cells and in other applications. The CNT composite is solid and environmentally stable; therefore it is not likely to degrade in performance from solvent evaporation, thermal cycling, damage, or breakage of the solar cell, hi one embodiment, iodine encapsulated inside carbon nanotubes converts to polyiodides, I₃ ^" and I₅ ^", and nanotube cations. The CNT-iodides are the redox active composite material in a dye sensitized solar cell.

The improvements in electrical conductivity of the nanotubes afforded by this invention composite enable the fabrication of transparent conductive coatings on a wide variety of substrates and as free standing film. The enhanced conductivity of the filled nanotubes is directly translated into macroscopic coatings by deposition from solvents by traditional and well know coating methods. The resulting coating is transparent to a wide range of the electromagnetic spectrum (from soft UV to the Far-IR and beyond) when deposited as a coating less than about 1 micron, more preferably less than about 0.1 micron. Such coatings are useful in making numerous consumer devices such as touch screens, LCD displays and others described in US6988925 and US Applications: 20030122111, 20050221016, 20050209392, which are incorporated by reference. These coating are also useful in solar cells as the top electrode.

Methods of changing the electronic properties of the nanotubes, and layers comprising nanotubes, are enabled by the addition of another material, forming a composite. More specifically, CNT composites improve the performance of photovoltaic devices through work function control and use as solid electrolytes. Furthermore, the composite nanotubes exhibit increase electrical conductivity thereby enabling formation of improved conductive coatings from said composite nanotubes. The conductive coatings are also transparent when deposited thinly on a substrate or suspended with a network of nanotubes or polymeric materials.

A conductive network comprising carbon nanotubes is deposited onto a substrate to form a layer, film, or coating. In a preferred embodiment, the carbon nanotubes consist of a single layer of graphene and have a diameter not greater than 3.5 nm and a length not less than 100 nm. The term CNT film does not imply that the film is exclusively CNTs. Rather, it may consist of a film greater than 50% CNT, where the other impurities are a result of the CNT production and purification process, hi a preferred embodiment, the CNT film is 20%-99.9% transparent and has a sheet resistance of 0.1 Ohm/sq to 10,000 Ohm/sq. In other embodiments, the CNT film is 20%-40% transparent and has a sheet resistance of 0.01 Ohm/sq to 500 Ohm/sq. In other embodiments, the CNT film is 40%- 60% transparent and has a sheet resistance of 0.1 Ohm/sq to 500 Ohm/sq. In other embodiments, the CNT film is 60%-80% transparent and has a sheet resistance of 1 Ohm/sq to 10,000 Ohm/sq. hi other embodiments, the CNT film is 80%-95% transparent and has a sheet resistance of 5 Ohm/sq to 7,000 Ohm/sq. hi other embodiments, the CNT film is 95%-99.9% transparent and has a sheet resistance of 10 Ohm/sq to 100,000 Ohm/sq.

In a preferred embodiment the substrate for depositing the CNT or CNT composite is transparent, but a transparent substrate is not a requirement for all applications. In other embodiments, the substrate is metal, ceramic, plastic, or a combination of metal, ceramic, or plastic. For example, the substrate may be glass with platinum particles deposited on the substrate. In one embodiment, the substrate incorporates refractive index matching layers to improve optical properties of the layered CNT-substrate structure. In one embodiment, the substrate is a functional material, mixture of materials, or layers of materials that absorb light and converts it to electron- hole pairs for the purpose of creating a useful current. Specifically, the substrate may be the active materials of a solar cell. In one embodiment, the substrate is rough such that it interpenetrates into the nanotube network.

Electric field effect (EFE) can be employed to increase CNT conductivity by controlling of the position of Fermi level of semiconducting CNTs. As 2/3 mass fraction of SWNTs consists of semiconducting nanotubes and only 1/3 of metallic; it could be very desirable to transform semiconducting nanotubes to quasi-metallic conductors. It is known that semiconducting nanotubes can function as transistors. The tubes exhibit increased conductivity when a negative bias is applied to the gate electrode (Fig.l) inducing a shift of the Fermi level toward the top of the valence band. In contrast, positive gate bias decreases conductivity as electron injection leads to the recombination of electrons with holes, which are major carriers in nanotubes (p-type semiconductors). Thus, at large enough negative gate voltages the hole concentration can be comparable with electron concentration in metallic tubes.

In one preferred embodiment two transparent and conductive CNT layers are separated by the insulating material. Then the resistance of CNT layer is enhanced by the applied voltage between two CNT layers due to EFE. As distinct from the field effect transistor described in the patent application [3], EFE is used herein to improve R/T performance of transparent conductive coating but not for fabrication of electronic devices like transistors. The basic principal of transistor function is the control of source- drain current by the gate bias, which implies the high off/on ratio, the absence of hysteresis, fast response time and other parameters characterizing the performance of such electronic device. For the present invention, which is an improvement of the R/T performance, there is no necessity to consider the above parameters since an electronic device is not a subject of the presented invention. For example, enhanced conductivity of CNT layer can be attained irrelevantly to hysteresis, on/off ratio and response time. CNT conductivity can be enhanced at least by the factor of two at appropriate design of CNT/insulator/CNT sandwich and relatively low gate bias in the range of 0.1 - 3.0 V.

Another embodiment of the invention provides an increase of transmittance of SWNT coating in vis-NIR range (0.5 - 2 μm) due to electric field effect. The SWNT absorbance of the light is defined by the energy gap between peaks of the density of states in valence and conducting bands [4], which is depends of CNT diameter. Usually, absorbance of SWNT layer (which used for conductive coating) consist of two bands (Sn and S₂₂) in NIR range (800-2000nm) and one (Mn) in the visible part of the light spectrum. The present invention establishes that an increase of the concentration of charge carriers in SWNT bands reduces an intensity of the absorbance bands (S₁₁, S₂₂ and in some cases Mn). The chemical doping resulting in hole injection is a cause of the considerable reduction of SWNT absorbance. Similar to chemical doping, EFE shifts the Fermi level toward the valence band (at negative gate bias) increases the hole concentration and consequently reduces SWNT absorbance, making conductive coating more transparent.

The present invention confirms that EFE, indeed, can induce an enhancement of conductivity of SWNT coating. In these experiments, SWNT were sprayed onto both side of transparent film of Nafion. Nafion (solid electrolyte) was specially taken to maximize charge injection under applied gate bias. A change of the I-V curve slope (resistance change) upon gate voltage that clearly indicates to charge injection in SWNT layer was observed. Under negative bias for SWNT sheet of 10x10 mm the resistance reduction up to 20% was observed at 1.2 V gate voltage.

Carbon nanotubes can be doped by bringing any substance or combination of substances into contact with the nanotube that will remove or add electrons or electronic density to the carbon nanotube. Lewis acids and Lewis bases meet the general requirement of removing and adding electronic density to the nanotube, respectively. It has been shown that doping is an exceptionally effective way to increase the conductivity of conjugated polymers, such as polyaceylene. However, doping conjugated polymers has several drawbacks, which are addressed in this disclosure. In a preferred embodiment, doping carbon nanotubes increases conductivity by a factor of 100 and does not cause a decrease in percent transmittance over the full visible spectrum.

Doped conjugated polymers experience many orders of magnitude improvement in conductivity. Polyacetylene (room temperature (RT) conductivity -10-5 S/cm) was doped with AsF5, iodine, AgClO4, bromine, or sodium napthalide, causing an increase in RT conductivity to values of 560, 360, 3, 0.5, or 80 S/cm, respectively. These values of conductivity are in the range of some metals. In all cases, polyacetylene was saturation doped to achieve maximum conductivity. These conductivity measurements were made in a controlled environment that prevented the degradation and loss of the dopant and prevented degradation of the conducting polymer. Other conducting polymers are stable when doped, but the dopants suffer from degradation or escape when the doped polymer under normal conditions. This example is illustrative of how doped structures may have improved conductivity, but their functional properties are not stable.

Carbon nanotubes are exceptionally stable structures with a hollow cavity that is capable of hosting molecules. These molecules interact with the electronic network of semiconducting nanotubes, increasing available electrons or holes for conduction. Carbon nanotubes can be effectively doped under certain conditions and the doping does not cause degradation of the nanotube upon exposure to air and water. Rather, dopants can leave the system (dedoping) or they can degrade by reacting with water or air.

In a preferred embodiment, the added dopant or encapsulated material does not substantially form covalent bonds with the CNT sidewall, which would cause a reduction in conjugation and thus a decrease in conductivity. In a most preferred embodiment, the material added to the CNT interacts via dispersion forces and via ionic and/or donor- acceptor bonding. However, the added material may interact with the CNT ends or defects, or functional groups on the CNT ends or defects via covalent bonding. In one embodiment, the nanotube does not form covalent bonds with the added material or materials. In another embodiment, the nanotube forms mostly ionic bonds with the added material or materials.

In other embodiments, the dopant forms a redox couple with the nanotube, which makes the nanotube-dopant composite an electrolyte. The dopant material either donates or accepts electrons from the carbon nanotube, depending on the application of the CNT composite. In a further preferred embodiment, the dopant material is encapsulated inside the nanotube and forms a redox couple with the carbon nanotube to form a CNT composite. This material is then used as the hole conducting solid electrolyte in dye sensitized solar cells, replacing the liquid electrolyte used in Graetzel cells, also known as dye sensitized solar cells.

One embodiment of this invention exposes a nanotube film to a Lewis acid or a Lewis base, preferably a Lewis acid. Preferably, the reaction spontaneously proceeds as follows for each dopant: X + SWNT → X^" + SWNT⁺, where X is the Lewis acid, and SWNT is the nanotube or nanotubes. This process p-dopes nanotubes, such that holes are the dominant charge carrier. Preferably, X^" is a stable anion that does not substantially react with air or water under usage conditions (-20 deg C to 100 deg C in air, 0% to 100% humidity). X^" can be in the form of (X^")_n where n is the number of units in a polymer chain. X can also convert to X^m~ in the presence of nanotubes, where m is an integer greater than 1. X may also interact with nanotubes such that X + SWNT — > X^d~ + SWNT^d+, preferably where X contains one or several halogen atoms, such as fluorine. X^d" can be in the form of (X^d")_n where n is the number of units in a polymer chain. Combinations of fluoropolymers and Lewis acids are also preferred as dopants. Preferably, the dopant reaction product does not convert back to the reactant, nor does it volatilize under standard usage conditions (-20 deg C to 100 deg C, 0% to 100% humidity). Nation is a preferred dopant, hi a preferred embodiment, the dopant reactant, the dopant reaction product, and any unintended side products are transparent in the visible region and do not impart any coloration to the nanotube film.

In the case of a Lewis base, the reaction, for each dopant, spontaneously proceeds as follows: X + SWNT → X⁺ + SWNT^", where X is the Lewis base, and SWNT is the nanotube or nanotubes. This process n-dopes nanotubes, such that electrons are the dominant charge carrier. Preferably, X⁺ is a stable anion that does not substantially react with air or water under usage conditions (-20 deg C to 100 deg C , 0% to 100% humidity). X⁺ can be in the form of (X⁺)_n where n is the number of units in a polymer chain. X can also convert to X^m+ in the presence of nanotubes where m is greater than 1. X may also interact with nanotubes such that X + SWNT → X^d+ + SWNT^d", preferably where X contains an amine (primary, secondary, or tertiary). X^d+ can be in the form of (X ⁺)_n where n is the number of units in a polymer chain. Combinations of nitrogen containing polymers and Lewis bases are also included as dopants. Preferably, the dopant reaction product does not convert back to the reactant, nor does it volatilize under standard usage conditions (-20 deg C to 100 deg C, 0% to 100% humidity). In a preferred embodiment, the dopant reactant, the dopant reaction product, and any unintended side products are transparent in the visible region and do not impart any coloration to the nanotube film.

In a preferred embodiment, dopants are located in the previously hollow interior of the carbon nanotube. Dopants and dopant reaction products, especially monomelic dopants and dopant reaction products, inside carbon nanotubes (endohedral dopants) are in a low energy configuration so that they are more thermodynamically stable than when located on the nanotube exterior. Additionally, endohedral dopants and dopant reaction products are more kinetically stable than exohedral dopants and dopant reaction products, since endohedral dopants must diffuse down a portion of the length of the carbon nanotube to reach an open end and desorb. Endohedral dopants and dopant reaction products are sterically shielded from oxygen and water in the environment, leaving them less vulnerable to reactions that could degrade the dopant and dopant reaction products. It was also discovered that cobaltacene, Co(Cp₂), can be encapsulated inside carbon nanotubes and will react solely with the nanotube to form Co(Cp₂)⁺ and SWNT^". Under atmospheric conditions, cobaltacenium would otherwise readily react. However, encapsulation enhances stability of the dopant.

In a preferred embodiment, the material inside the carbon nanotube accepts electrons from the carbon nanotube to become a negatively charged ion, and the nanotube becomes a positively charge ion.

In one embodiment, the dopant material exposed to the nanotube is iodine. In a further preferred embodiment, the iodine becomes encapsulated inside the nanotube, and the iodine converts to a mix of triiodide and pentaiodide ions when encapsulated. The more a carbon nanotube is filled with iodine, the greater the amount of pentaiodide, relative to triiodide. Thus the reaction can be viewed as follows: 3I₂ + SWNT → 2I₃ ^" + SWNT²⁺. (1)

As more iodine is added, the reaction continues:

Is^" + I₂^I₅ ^", (2)

Where a densely packed nanotube contains I₅ ^". This final reaction can be viewed differently when the quantity of the iodine is constant and no molecular I₂ is present, such that:

5I₃ ^" ~3I₅- + 2e^". (3)

This is one redox reaction for polyiodides inside nanotubes to form a nanotube-iodide electrolyte. The equilibrium for this reaction is controlled by two factors: the amount of iodine from reaction (2) and the potential applied to the iodine-SWNT composite from (3). The SWNTs may participate in the redox reaction upon application of a voltage bias. The amount of I₃ ^" present can be controlled initially by choosing how much of CNTs interior is filled with iodine, hi a dye sensitized solar cell, a dye is photoexcited and creates an electron-hole pair. The electron migrates to another material, such as titanium dioxide; the hole interacts with the electrolyte, oxidizing it. Next, the electrolyte is reduced at the interface with the conductive electrode, returning the cell to it initial state and creating a current flow.

In other embodiments, the nanotube is filled with an electrolyte solution used by those in the field of dye sensitized solar cells and in the field of liquid electrolytes. The filled CNT composite is used as a solid electrolyte in a dye sensitized solar cell. The electrolyte solution includes a materials of the formula M₃R_b, where a and b are variables greater than or equal to one. R is a suitable anion such as a halide, fluoride, chloride, bromide, iodide, trifluoromethyl sulfonate, hexafluorophosphate, sulfate, perchlorate, thiocyanate, carbonate, or phosphate, and M is a cation such as Li, Cu, Ba, Zn, Ni, lanthanides, Co, Ca, Al, Mg, other transition metals, or other suitable metals. The electrolyte may also be imidazolium iodide and derivatives thereof. Liquid crystal electrolytes and ionic liquid electrolytes may be used in combination with any of materials above or in combination with each other. On one embodiment, the combination of lithium iodide and elemental iodine is inserted into carbon nanotubes. It should be noted that no solvent is required for the electrolyte to function. Any combination of the above materials and any other known electrolyte may be encapsulated inside nanotubes to act as a solid electrolyte, including single elemental materials, such as iodine or bromine, that will form a redox couple with the SWNT or with themselves at room temperature.

CNT electrolyte in a Graetzel cell can be incorporated in a variety of ways. Material to form part of the redox reaction in the molten, gas or solution form may be exposed to CNTs. Preferably the CNTs are opened by an oxidative treatment. However, sufficiently oxidizing material, molten iodine for example, can open and fill CNT simultaneously. In one embodiment, the CNTs are shortened and filled with material to form a solid electrolyte. Next the CNT ends are functionalized, for example with octadecyl amine, to solubilize the CNTs. Next the CNTs are deposited by spray, dip, or spin coating. In this case, the CNT network is not as conductive as a network of full- length CNTs. However, the presence of electrolyte enhances the transport of holes to the electrode, therefore the cell will be more efficient than without an electrolyte. Alternatively, the CNT network can be formed, and then the network is exposed to material to form part of the redox reaction in the molten, gas or solution form. In a preferred embodiment, opened SWNTs are filled with gaseous or molten iodine, which forms the redox reaction described above. Further layers of the Graetzel cell are added, including the TiO2, dye, and transparent electrode. Note that the transparent electrode may consist of a network of carbon nanotubes or of a CNT composite. The cell may be fabricated to receive light on the side of the TiO2 or on the side of the electrolyte. In either geometry, the CNT may be used as the electrolyte and also may be used at the transparent electrode.

Ideally, the dopant, the dopant reaction product, and any unintended side products are transparent in the visible region and do not impart any coloration to the nanotube film. However, iodine and bromine are known to be effective dopants for carbon nanotubes, and both elements have a notable coloration in the visible region. Iodine inside nanotubes causes large increases in nanotube conductivity of nanotube mats, approximately by a factor of 40. These changes were observed on opaque, black, dense mats of carbon nanotubes and therefore any changes in optical transparency were not taken into account. Since endohedral dopants are located inside the carbon nanotube, they are less prone to interact with light and therefore less likely to absorb light. Also, dopants inside a carbon nanotube are preferably densely packed so that their optical cross section is minimized, hi a film of 500 Ohms/sq at 95%T, approximately 70% of the light that passes through the sample does not interact with carbon nanotubes. Therefore, endohedral doping can only affect the 30% of the light passing through the nanotubes that are present.

Since the average diameter of carbon nanotubes is 1.4 nm, no more than two atomic layers of a material can absorb light that passes through the nanotube sidewall. hi a preferred embodiment, visible light will pass through the endohedral dopant 90%- 100% of the time, causing a change in transparency of 0% to 3%. In other embodiments, visible light will pass through the endohedral dopant 80%-90% of the time. Furthermore, in a preferred embodiment, the film's optical properties are dominated by the void space, rather than the nanotubes, so changes in optical properties of the nanotubes will not affect the film's transmittance at a given film thickness.

Dopants encapsulated inside carbon nanotubes can be stabilized by closing the ends of the carbon nanotubes. End closing can be achieved through a variety of means. In a preferred embodiment, diamines, molecules with two or more primary amine groups and a spacer, reacted with acid chloride (-COC1) functionalized carbon nanotubes. The full reaction involved opening carbon nanotube end caps using an oxidative treatment (HN03, hot air, molten iodine, etc.) and terminating the opened ends or opened holes in the nanotube sidewall with carboxylic acids. These carboxylic acids were reacted with SOC12 or with PC15 to form COCl functionalized nanotubes. These acid chlorides were then reacted with diamines such as p-Xylylenediamine or Ethylenediamine to close the ends of carbon nanotubes or to link adjacent carbon nanotubes. In one embodiment, the CNTs are shortened and filled with dopant to form a solid electrolyte.

In a further preferred embodiment, nanotubes were opened and functionalized with carboxylic acids by nitric acid, then soaked in neat thionyl chloride. The thionyl chloride fills the interior of the nanotube and reacts with the carboylic groups to form acid chlorides. Next, p-Xylylenediamine or Ethylenediamine is exposed to the nanotubes to close the nanotube ends, trapping thionyl chloride inside the carbon nanotubes. hi another embodiment, fullerene molecules (C₆₀, C₇₀) of follerene molecules with functional groups are used to "close" the ends of carbon nanotubes. Fullerenes have a large exothermic energy of encapsulation in carbon nanotubes, so they will be trapped inside a carbon nanotube irreversibly. The functional groups on fullerenes may be provided to improve solubility of the fullerene or to increase the likelihood that the fullerene becomes trapped in the carbon nanotube end.

Doping nanotubes can occur through a variety of means. Dry nanotubes may be exposed to a vapor of the dopant, to a solution of the dopant, or to a neat liquid dopant, or to a dopant dissolved in a supercritical fluid. Dopants are mixed to optimize doping. For example, iodine dissolved in thionyl chloride is used as a dopant. Dopant exposure may occur at reduced temperature, room temperature, elevated temperature, in air, in inert atmosphere, or in vacuum. Dopants may be exposed to nanotubes while the nanotubes are dry, dispersed in an organic solvent, in water, or in a mixture of water and organic solvent. Doping may occur with as-produced nanotubes, air oxidized nanotubes, acid oxidized nanotubes, surfactant coated nanotubes, polymer coated nanotubes, or purified nanotubes. Dopants are added to nanotube films, nanotube electrodes, or nanotubes in a porous medium.

The addition of another material to the CNT can occur at a variety of points during the production and processing of the CNT. In a preferred embodiment, the as- produced film on a substrate is exposed to another material or combination of materials to form a composite that has altered RT performance. In other embodiments, another material is exposed to CNTs while CNTs are dispersed in solution (water, alcohol, THF, DMF, other organic solvents) or while the CNTs are solid, but not intentionally placed on a substrate (i.e. in a container for processing, as a film suspended in air, or as a film on a disposable or removable substrate). In a preferred embodiment, the ends of the carbon nanotubes have been opened so that the other material enters the interior cavity of the CNT. hi another embodiment, the exterior of the nanotubes are coated with a material or combination of materials, hi another embodiment, other materials are exposed to nanotubes during the production process (arc discharge, CVD, laser) to create a composite so that carbon atoms in the CNT sidewall are replaced with other atoms, including, nitrogen, boron, transition metals, and lanthanides. Other material or materials may be introduced to CNTs as a solid, liquid, gas, dissolved in solution, or dispersed in a liquid. Pressure, vacuum, and heat may be used to cause a phase transition to more easily incorporate the material into the CNT interior cavity or into a formed CNT network. The other material or materials are introduced to CNTs in air, in an inert environment, in an oxidizing environment, in carbon dioxide, or in vacuum. Molecules may be inserted in CNTs with the aid of supercritical fluids.

In another embodiment, nanotubes are filled with molecules that will be reacted at a later stage. In one embodiment, these molecules form covalent bonds with each other inside the carbon nanotube. Next, a reaction of the monomers is initiated to form a polymer inside the carbon nanotube. For example, nanotubes are preferably filled with acetylene, which polymerize inside carbon nanotubes at room temperature. Also, for example, fullerene epoxides are inserted into carbon nanotubes using supercritical fluids, then reacted to form a one-dimensional polymer upon heating to 250 degrees C. Another example is of C₆₀ or C₇₀ fullerenes inside nanotubes polymerizing to form a concentric nanotube inside the host nanotube. This polymerization is initiated by heat, UV light, electron irradiation, or a change on nanotube redox state through chemical or electrochemical means. The dopant simultaneously initiates polymerization and doping of the nanotube and/or polymer chain inside the nanotube.

A dopant can be exposed to the hybrid structure using methods described above, doping the nanotube and/or the encapsulated polymer. For example, polyacetylene chains inside carbon nanotubes are doped with iodine or bromine using methods described above. In either case, there must be enough space for dopants to enter nanotubes. There is ample evidence that dopants can enter nanotubes filled with fullerenes; individual potassium atoms were imaged by high resolution TEM inside carbon nanotube filled with fullerenes after exposure of the nanotube- fullerene structure to potassium vapor. Exposing fullerene-filled carbon nanotubes to potassium has been shown to shorten interfullerene distances and cause the fullerenes to polymerize into a metallic structure inside the nanotube. This method of making doped structures offers some advantages, since the hybrid structure benefits from nanotube conductivity, encapsulated molecule or polymer conductivity, the added conductivity from doping both structures, and the increased thermodynamic stability for dopants, due to the decreased free space inside the carbon nanotube cavity and the increased bonding interactions from both the nearest part of the nanotube sidewall and the encapsulated molecule or polymer chain.

Dopants located on the exterior of the nanotube may be stabilized physically and chemically by further treating the nanotube film. The treatment consists of a step to polymerize monomelic dopants, by coating the film, or by sealing the film. The coating can consist of any solid material that can be deposited on the surface without substantially destroying functional properties of the doped film. These solid materials are preferably polymers, fluoropolymers, chloropolymers, ionomers, inorganic films, metal oxide films, metal films, monocrystallyine films, polycrystalline films, and amorphous films, metal particles, metal oxide particles, and polymer particles. The coating also imparts some doping in addition to the already present dopant. The coating is applied by dipping the doped nanotube film in a solution of the solid material, by spraying a solution of the solid material, by electrostatic painting, by sputtering, by depositing a precursor on the film and reacting it (polymerizing monomers), or any means known to those skilled in the art. A coating also increases the kinetic and thermodynamic stability of endohedral doping.

In one preferred embodiment, a CNT network is employed as the transparent electrode in a solar cell. Amorphous silicon, CIGS, CdS, Graetzel, organic, exitonic, multijunction, and quantum dot-based solar cells all require a transparent electrode. In all cases, electron-hole (e-h) pairs are created from photons in the active material(s) and must be transported to different electrodes before recombining. In this preferred embodiment, one of the electrodes of a solar cell must be transparent.

Each solar cell mentioned above has unique electronic properties and responses to light and thus has different requirements for transparent electrodes. These requirements are: transparency, conductivity, work function match, type of charge carrier, and electrical contact to the active material. Ideally, a transparent electrode would have maximum transparency to create the largest amount of e-h pairs possible. Also, the transparent electrode would have maximum conductivity to turn the maximum amount of e-h pairs into useable current. Other considerations include the effect of work function matching and the type of charge carrier to be transported to the electrode. Adding another material to the nanotube tunes these properties, which improves efficiency of the solar cell beyond what would be realized by just improving resistance and/or transparency of the transparent electrode.

In a preferred embodiment, the composite CNT material used as a transparent electrode leads to a greater conversion of solar energy to electrical energy, as compared to using only a CNT film. In a further preferred embodiment, the CNT composite causes additionally improved solar efficiency due to work function matching with the active layer. Work function matching is achieved by adding a material to the CNT film that changes the work function of the CNT composite. This effect can be achieved, for example, by adding a dopant to increase the number of conducting holes or the number of conducting electrons in the CNT network, hi one embodiment, the work function of the CNT composite is increased to increase the built-in voltage potential of the solar cell.

In a further preferred embodiment, the added material or dopant increases the number of charge carriers most useful to the cell design. For example, CIGS and Graetzel solar cells transport electrons to the transparent electrode. CNTs are p-doped in air, but would be more efficient as n-doped materials to conduct electrons. CNTs are converted to n-type conductors by doping with a Lewis base, encapsulating an electron donating molecule, such as cobaltacene, incorporating alkali metals, incorporating nitrogen- containing polymers, or by replacing carbon atoms in the CNT sidewall with nitrogen atoms. As another example, exitonic solar cells transport holes to the transparent electrode. Therefore, a greater number of holes contributing to CNT conductivity would enhance current through the cell and thus increase efficiency. Holes may be added by doping with a Lewis acid, encapsulating iodine, bromine, TCNQ, superacid, Lewis acid- Bronstead acid mixtures, or any electron withdrawing molecule or element.

In a preferred embodiment, the material exposed to the nanotube forms a redox couple with the carbon nanotube to form a CNT composite. The material either donates or accepts electrons from the carbon nanotube, depending on the application of the CNT composite, hi a further preferred embodiment, the material is encapsulated inside the nanotube and forms a redox couple with the carbon nanotube to form a CNT composite. This material is then used as the hole conducting solid electrolyte in dye sensitized solar cells, replacing the liquid electrolyte used in Graetzel cells, also known as dye sensitized solar cells.

The CNT composite is solid and stable; therefore it does not significantly leak from thermal cycling, damage, or breakage of the cell. The electrolyte is chemically and physically shielded from interactions with air and water, so performance degradation is minimized. CNTs are deposited from solution in the form of spraying or film coating (Gravure, slot die, etc.) and the electrolyte or redox active material can in turn be encapsulated inside CNTs while the nanotube is a film. Alternately, the electrolyte can be encapsulated prior to deposition. In a preferred embodiment, the CNT composite is deposited onto a substrate in a continuous, roll-to-roll process.

In a preferred embodiment, the nanotube layer or coating is exposed to a material that increases the conductivity of the nanotube network, as measured by DC sheet resistance. In a further preferred embodiment, the nanotube film sheet resistance decreases from 80% to 10% upon exposure to another material or materials, hi other embodiments, the CNT composite is deposited on a substrate and the composite has a lower sheet resistance for a given mass of CNT, compared to a CNT film without any added material on the same substrate. In other embodiments, the CNT composite has identical sheet resistance to the CNT without added material. In the embodiment of CNTs acting as a solid electrolyte, transparency of the resulting film is not a requirement for all solar cell geometries.

In a preferred embodiment, the encapsulated material does not affect the percent transmittance of the film, compared to a film with the same mass of CNT. In other embodiments, the encapsulated material increases percent transmittance from 0.00001% to 5% for a film that is initially 95% transparent, with appropriate scaling for films of lower %T. In other embodiments, the encapsulated material decreased %T from .0000001% to 1% based on a film that is initially 95%T. In other embodiments, the encapsulated material decreased %T from 1% to 3% based on a film that is initially 95%T. Ln other embodiments, the encapsulated material decreased %T from 3% to 6% based on a film that is initially 95%T. In other embodiments, the encapsulated material decreased %T from 6 to 15% based on a film that is initially 95%T. In other embodiments, the encapsulated material decreased %T from 15% to 40% based on a film that is initially 95%T.

Electronic changes to CNT films imparted by adding another material are useful for improving material properties beyond RT performance. Furthermore, electronic changes to CNT films imparted by adding another material are useful for applications beyond solar cells. Correct addition of another material imparts changes to electronic structure of the film that improve thermal stability, UV and visible light stability, changes from humidity, and changes from thermal cycling, as measured by changes in sheet resistance or resistivity over time. Appropriate materials inside CNTs will change the ease with which the CNT can be oxidized by air. In a preferred embodiment, Lewis acids or Bronstead acids increase the thermal, UV, and visible light stability of CNT coatings.

In other embodiments, the change in work function of the CNT composite is beneficial to organic light emitting displays (OLED) and improves the OLED efficiency at emitting light. In one embodiment, the work function of the CNT composite is decreased to use the CNT composite as an electron injecting electrode in an OLED device. The change in work function will benefit LCD displays using transparent electrodes. The work function of the CNT composite can be adjusted to be close to the work function of a reflective pixel electrode in an LCD. An improvement in electroluminescent (EL) lamp lifetime and/or brightness occurs when using CNT composites as the transparent electrode described in this disclosure, as compared to bare CNT electrode or other organic alternatives.

Changing the charge carriers of the CNT composite to make the material a p-type conductor or an n-type conductor is useful for some applications besides solar cells. Most transparent conductive oxides (TCOs) are n-type conductors. P-type TCOs have much lower conductivities, and therefore are not used to make transparent circuit elements. It is possible to integrate CNTs with TCOs to make heterostructures, but similar effects can be achieved by changing the carriers in CNT networks. Transparent p-n junctions, transistors, diodes, including light emitting diodes are fabricated with CNT composite acting as one or both of the materials. Also, smart windows or electrochromic windows take advantage of different carriers in the transparent conductors. Solid CNT electrolytes have potential for use in a variety of applications beyond Graetzel solar cells. One example of an application is the use of CNT electrolytes as high dielectric materials for capacitors. Additionally, the CNTs have high dielectric constants (-10 for semiconducting CNTs and -1,000 for metallic CNTs), which would give an additive effect of electrolytic capacitors made of high dielectric constant CNTs. Fuel cells generate current by ionic conduction through a proton exchange membrane (PEM). Proton conduction of a CNT network employed as a PEM may be enhanced, for example, by incorporating Nafion or another polymeric electrolyte into the CNT network.

Carbon nanotubes are doped by replacing a carbon atom in the nanotube sidewall with another, different atom. For example, nitrogen has been incorporated into nanotube sidewalls by adding a nitrogen source to arc discharge and chemical vapor deposition growth of nanotubes. Doped nanotubes have similar or the same bandgap as undoped counterparts. Growth conditions with dopant present may favor one or several chiralities, but do change the broad observation that 1/3 of the SWNTs are metallic and 2/3 of the SWNTs are semiconducting. Also, the diameter distribution of doped nanotubes are similar to that of undoped nanotubes grown with the same method and under similar conditions. Therefore, doping leads to similar optical properties (e.g. refractive index) of the film. Furthermore, in a preferred embodiment, the film's optical properties are dominated by the void space, rather than the nanotubes, so changes in optical properties of the nanotubes do not affect the film's transmittance at a given film thickness.

In a preferred embodiment, using the standard film RT performance of 500 Ohms/square at 96% transparency, a dopant or combination of dopants decreases Rs from 90% to 10% and increases %T from 0% to 30%. In other embodiments, a dopant or combination of dopants decreases Rs from 99.9% to 90% and increases %T from 0% to 30%. In other embodiments, a dopant or combination of dopants decreases Rs from 50% to 10% and increases %T from 0% to 5%. In other embodiments, a dopant or combination of dopants decreases Rs of 90% to 50% and increases %T from 0% to 5%. hi other embodiments, a dopant or combination of dopants decreases Rs from 90% to 10% and decrease %T from 0% to 4%. In one preferred embodiment CNT alignment is an important factor which increases nanotube conductivity. Three major reasons of this phenomenon are the following:

1) from geometrical consideration (Fig. 2) it is clear that two-dimensional orthogonal alignment should result in higher conductivity than random CNT network at the same mass fraction (it should be noted that films deposited in such a method maintain anisotropic conductivity as a random network)

2) the CNT rope alignment significantly increases nanotube conductivity approaching quasi-ballistic transport, probably due to large area of inter-rope contacts. Such effect was observed experimentally for CNT aligned in magnetic field.

3) nanotube ordering effectively displaces unwanted impurities (catalyst particles, amorphous carbon) from the alignment area as it was reported by Hadberg et al.

4) alignment is achieved by means of imparting a force on the CNT film as it is forming. Centripetal force, shearing of a fluid, surface tension, electric field, or magnetic field cause CNTs to align. In addition, CNT alignment in one direction should result in an enhanced conductivity in this direction and diminished conductivity in the orthogonal direction. Anisotropy of conductivity is accompanied by optical anisotropy, i.e. the ability of the CNT film to polarize light.

One embodiment of the invention improves conductivity by reducing the resistance between nanotubes in a bundle and between bundles. Conductivity is improved by creating a localized metallic contact between adjacent CNTs. In a preferred embodiment, metallic nanoparticles, such as 5 nm gold, platinum, silver, nickel, tin, lead, copper, chromium, indium, beryllium, and other metals, will co-precipitate with the carbon nanotubes and preferentially aggregates at the intersections of bundles. Though not bound by theory, one assumption of the present invention is that aggregation occurs at the intersection because it is the last area to dry, due to the Kelvin effect. Also, the metallic contact can come in the form of a metal coordination complex, where a transition metal atom or cluster is bonded to the nanotube sidewall. The addition of small amounts of nanoparticles for improving CNT bundle electrical contact will not substantially alter the transmittance, since very little material will be used. However, it should be noted that the optical spectra of the hybrid material will not always be the sum of the individual spectra, since the two materials will be electronically coupled. In this sense, controlled ratios of nanoparticle to nanotube can lead to tailored optical properties of conductive films.

In a preferred embodiment, nanoparticles dispersed in a fluid are brought into intimate contact with carbon nanotubes and the fluid is removed. In a further preferred embodiment, carbon nanotubes and nanoparticles are co-dispersed in a fluid and then deposited onto a surface and dried. In another embodiment, a fluid dispersion of nanoparticles is coated onto a preformed film of carbon nanotubes and dried. The essential points of the method are that dispersed nanoparticles are mixed with carbon nanotubes and the fluid is removed to allow the nanoparticles to aggregate onto the nanotubes, especially at the nanotube intersections.

In another preferred embodiment, a metal atom or cluster of metal atoms forms a monohapto, dihapto, pentahapto, or hexahapto, bond with one carbon nanotube sidewall or end and a monohapto, dihapto, pentahapto, or hexahapto covalent bond with a different carbon nanotube sidewall or end. The metal is preferably f group elements, transition elements, main group elements, silicon, boron, selenium, germanium, arsenic, antimony, and tellurium. In addition, several metal atoms may be bonded to each other, and one or more of these metal atoms may be bonded to two or more carbon nanotubes. Film transparency is tuned to optimize conductivity, where linking metal atoms make up a small percentage of the absorbing material. Conductivity increases due to a decreased electrical resistance in interbundle junctions and intertube junctions. Transparency is dominated by void space, and film coloration from metallic elements can be tailored by judicious choice of elements or combination of elements to minimize coloration.

An example of a similar compound comes from reactions of platinum-ruthenium and ruthenium with C₆₀ fullerenes. The interaction of C₆₀ with Ru₅C(CO)₁₅ or PtRu₅C(CO)₁₄(COD), followed by treatment with solubilizing phosphines, gave compounds with a hexahapto coordination of C₆₀ to a Ru₃ face of the square pyramidal Ru₅C or octahedral PtRu₅C cluster framework. In other work, C60 fullerenes dimerized with a metal cluster bridge to form a metal sandwich complex. These works demonstrates that all-carbon structures can be functionalized with coordinated metal complexes and metal clusters, though this work cannot be extended to conductivity of carbon nanotubes or transparency of nanotube films.

Another synthetic method to link to carbon nanotubes may be borrowed from C60 fullerene organometallic synthesis. For example, Rh₆(CO)₉(dppm)₂(μ₃-η²: η²: η²-C₆₀) can be prepared by direct thermal interaction of Rh₆(CO) ₁₂(dppm)₂ with C_O0. A metal cluster such as Rh6 (CO)j₂- (dppm)₂, which has enough electron-donating ligands to compensate for the electron- withdrawing effect of C60, can form a stable bisfullerene adduct. Excess C₆₀ reacts with Rh₆(CO)9(dppm)₂(μ₃-η²: η²: η²-C₆o) in refluxing chlorobenzene, followed by treatment with benzyl isocyanide at room temperature, forms a fullerene-metal cluster sandwich complex, Rh₆-(CO)₅(dppm)₂(CNCH₂C₆H₅) (μ₃-η²: η²: η²-C₆₀)₂ Rh-C distances of the complex imply that the electron density within the cluster unit is highly delocalized, which would increase conductivity between adjacent nanotubes. In this case, one metal atom is bonded to both fullerene cages, whereas the other 5 metal atoms are bonded to one or none of the other cages. In other embodiments, C₆₀-Ir₄ metal sandwich complex with a novel μ₄-η : η : η :η -C₆₀ bonding mode and further enhancement of the interfullerene electronic communication by inserting two metal atoms as a bridge between two C₆₀ cages. Generally, heavier metals will coordinate with two carbon nanotubes, and these metals are preferred.

These metal atom or small metal cluster coordination complexes with carbon nanotubes may not be stable to atmospheric conditions and may require special coatings to inhibit degradation of the metal linkage. Chromium, molybdenum, and tungsten form hexahapto sandwich complexes that are air sensitive, for example, and would require isolation from air to provide long-term benefits to nanotube conductivity. These coatings included fluorinated polymers and other polymers and surfactants that will effectively provide a barrier impervious to air and moisture. Organotin compounds are expected to be more stable and would be preferred as metallic elements to bond between to carbon nanotubes. Organotins are also used for increasing stability of PVC in light and heat. Organomercury compounds are also attractive, due to their stability. It should also be noted that any materials or combination of materials that are not volatile, but are air sensitive may be appropriate for space applications, such as in solar cells, sails, in detectors, or in antistatic coatings.

The following examples illustrate embodiments of the invention, but should not be viewed as limiting the scope of the invention. Example 1

Doping films with triethyloxonium hexachloroantimonate

40 mg of triethyloxonium hexachloroantimonate ((C₂Hs)₃O⁺SbCl₆ ^"), OA, was dissolved in 200 mL methylene chloride. A glass slide with silver epoxy contacts was sprayed with CNT to 448 Ohm (2 squares, 224 Ohms/sq) at 93 %T. The resistance of the slide was monitored during dipping, rinsing and drying of the slide (Figure 1).

The slide was dipped in the OA solution for seconds and immediately rinsed with methylene chloride. The resistance of the slide dropped to 305 Ohms, a 32% decrease, upon immersion in the OA solution. A binder coat was not added. Over a period of 2.8 days, the sheet resistance increased to a measured value of 369 Ohm, a net decrease of 18%. The loss of conductivity over time is due to dedoping, where the doping reaction products are degrading or desorbing in air and humidity. The doping-induced increase in absorbance may be due to subtle changes in the electronic structure, which could increase reflectance of the doped film, for example.

The OA doped sample was compared to an undoped sample by looking at the Vis-near IR spectra of the two samples. As is typical with partially doped sample, the Sn semiconducting transitions were suppressed, while the S₂₂ and metallic transitions were largely unaffected. The discontinuity at 820 nm is due to the instrument switching to a different detector.

Example 2

In this example, it was attempted to increase the R_s stability by coating the doped sample with Teflon AF to act as a binder. A slide with silver epoxy contacts was sprayed to 455 Ohms (2 squares, 227.5 Ohm/sq) and monitored. The slide was dipped in OA solution, rinsed with methylene chloride, then dip coated in Teflon AF. Similar to before, the sample resistance dropped 25% during exposure to OA solution, and the resistance started to increase immediately. Adding Teflon AF marginally slowed the R_s drift. After 2.8 days, the resistance of the sample was approximately 18% of the starting value, which indicates the topcoat did not substantially assist in slowing the dedoping process. The net change in resistance, 10% over 8 days, appears to be stable. Simply adding Teflon AF to a CNT film can cause a decrease in R_s of 5-10%. The transparency of the sample was not monitored due to constraints of the data acquisition equipment and because Teflon AF acts as an antireflective coating, artificially increasing %T.

Example 3

Iodine doped SWNTs

Initial experiments to dope CNTs with iodine were conducted with purified CNTs dried from paste. CNTs were mixed with ioidine in three different solvent environments and as molten iodine. In all cases, dried CNT paste was mixed with a tenfold excess of iodine, by mass. In all cases, the dried material was substantially more difficult to disperse, and the resulting RT performance was considered to be poor. A control sample of CNTs dried from paste, then redispersed back in 3:1 IP A/water performed substantially worse than CNTs sprayed without drying. A loss of 3% to 10% of the transparency at 500 Ohms/sq was observed upon redispersing CNTs. This result emphasizes that processing of the CNTs can be more important than the changes imparted by a dopant. Other experiments for iodine doping were conducted on as-produced films, which allows more controlled monitoring of RT performance changes for a given sample with a nominally static morphology and known baseline RT performance.

SWNT film was soaked in a toluene solution of iodine. 5 g of iodine was dissolved in 500 mL of toluene. A slide with silver metal contacts was sprayed to 1055 Ohm (2 squares, 527.5 Ohm/sq) at 94.7%T. A control sample to be soaked in toluene was sprayed to 1040 Ohm (2 squares, 520 Ohm/sq) at 94.4 %T. The slide placed in iodine solution was monitored during the doping process (Figure 4). Immediately after placing the slide in solution, the resistance dropped to 905 Ohms (452.5 Ohms/sq) and remained constant for 30 minutes. After 7 minutes of soaking in iodine solution, the slide was taken out. As the slide dried in air, the resistance continued to drop to a low value of 860 Ohms (430 Ohm/sq) once the slide was visibly dry. The slide resistance began to increase immediately after drying, and returned to 1055 Ohm, the original resistance prior to doping. The iodine desorbs from the CNTs once the film is dry, which leads to dedoping. It should be noted that the doping is reversible, indicating that the doping and dedoping did not damage the CNT network. The film was left to soak in iodine solution overnight. The slide resistance was monitored, and it was seen the resistance increased linearly from 905 Ohm to 1226 Ohm overnight. This linear increase in resistance can be attributed, at least partly, to the conversion of the silver contacts to silver iodide, which is less conductive. After the film was removed from solution, rinsed, and the contacts were repainted, the resistance was measured to be 1198 Ohm (599 Ohm/sq) at 93.4%T, showing the film lost some functional performance during soaking.

Iodine was inserted inside SWNTs using gas phase filling of the CNTs after an opening procedure. CNTs were sprayed on a small quartz slide to approximately 254.1 Ohms (1 square, 254.1 Ohm/sq) at 86.6%T. The sample was annealed in air at 350 degrees C for 30 minutes to open the ends of the CNTs. The RT performance was measured to be 664 Ohms at 94.7%T after air annealing, a notable change in both resistance and transmittance. Next, the sample was sealed in a test tube with iodine at the bottom and heated at 120 degrees C. After an overnight exposure to dense I₂ vapor, the sample was removed from the vessel and warmed under vacuum to remove excess iodine. The slide was measured to be 210.0 Ohms at 86.8%T. The renormalized values to 500 Ohms/sq are displayed below in Table 1.

Table 1: Iodine doped CNT film. %T values were renormalized to 500 Ohm/sq.

Example 4

During experiments with Nafion field effect devices, it was attempted to make thin Nafion films on SWNT coatings. In the process, it was observed that the sheet resistance of the SWNT would drop 20-30% upon coating with a solution of Nafion polymer. Also, an increase in transparency of approximately 5% was observed after Nafion coating.

The former effect, a drop in R_s, is ascribed to p-type doping by the polymer. Nafion is a unique polymer in that it is water soluble, essentially fully fluorinated, and has periodic sulfonic acid groups. The combination of the electron-withdrawing fluorinated backbone and the sulfonic acid makes Nafion a solid superacid, the strongest class of acids. Thus Nafion is almost an ideal polymeric dopant, since it is solvent compatible, stable at high temperature, and will readily withdraw electrons from SWNTs. Unlike OA doping, it was observed that Nafion doped SWNT films are stable over time, making it a better candidate for SWNT doping. Examining Nafion coated SWNT films showed that the Sn transition was fully suppressed, indicating full doping of the SWNT film.

The latter effect, an increase of 5-7% in transparency is due to Nation's optical properties. Nafion is a low index of refraction material (ca. 1.34) similar to that of Teflon AF. Vis-NIR spectra are shown for various thicknesses of Nafion coatings on glass slides Using a four point probe, SWNTs with a Nafion binder showed a sheet resistance of 318 Ω/D with painted silver contacts, whereas the four point probe gave a sheet resistance of 303 Ω/D. This test indicates that surface resistivity is maintained and that there may be a SWNT-Ag contact resistance of approximately 7.50.

SWNT films on glass were prepared. The Vis-NIR spectra were measured prior to Nafion doping and after Nafion doping. The RT performance of the SWNT network changed from 7.6 kΩ/α at 94.6%T at 550 nm to 2.3 kΩ/α at 101%T at 550 run upon doping, a drop of 70% in sheet resistance and an increase of 6.4%T.

In this experiment, the presence of the Sn peak after doping shows that Nafion has not fully doped the nanotubes. For complete doping using a polymeric dopant, a higher concentration of Nafion could be used, SWNT bundles may be decreased in size or a combination of small dopants, such as nitric acid, iodine, etc. and polymeric dopants can be used. Exploration of small dopant molecules to intercalate inside SWNTs, in addition to polymeric dopants to coat and bind SWNTs, can be continued.

Example 5

More detailed experiments were conducted with dip coating Nation. The dip speed was varied to control wet coating thickness (and thus amount of Nafion deposited) on glass and plastic substrates. It was found that the sheet resistance dropped 15.5% to 20.7% for these samples with dip speeds ranging from 0.475 inches/min to 4.25 inches/min. The results of these experiments are displayed below in Table 2. Table 2: Experiments with dip coating 5% Nafion at different speeds.

Example 6

More detailed experiments were conducted with dip coating Nafion. The number of dips were varied to vary the amount of Nafion deposited on glass and plastic substrates. It was found that the sheet resistance dropped 14% to 17.8% for these samples. The results of these experiments are displayed below in Table 3. Table 3: Experiments on 5% Nafion dipped at 4.25 inches/minute.

Example 7

More detailed experiments were conducted with dip coating Nafion. The concentration of Nafion in solution was varied to control the amount of Nafion deposited on glass and plastic substrates. It was found that the sheet resistance dropped 12.2% to 20% for these samples. Higher concentrations than 5% may dope the nanotubes more, resulting in lower sheet resistance. Results from experiments varying the amount of Nafϊon deposited on glass and plastic substrates can be seen below in Table 4. Table 4: Experiments on varied concentrations of Nafion.

Example 8

More detailed experiments were conducted with dip coating Nafion. The sheet resistance of nanotube films was varied to determine if the doping effect was limited by film thickness or network density. It was found that the sheet resistance dropped >10% over a wide range of sheet resistances, indicating that the entire network is being doped by Nafion, as seen below in Table 5. Table 5: Experiments with ranges of sheet resistances.

Example 9

Schematic of Dye sensitized solar cell using nanotubes as an electrolyte. In the present invention nanotubes are used as the electrolyte, as well as the top and bottom transparent electrodes.

Example 10 Nanocyl double walled nanotubes (DWNTs) were purified and dispersed in a solution of alcohol and water. A transparent conductive film was made by spray coating the DWNT dispersion onto a glass slide 3" x 1" with electrodes 2" apart. The sprayed sample had a transparency of 95.2%T at 7,535 Ohms/sq at room temperature. The sample was flow coated with 5% Nafion and placed in a 100 degree C oven to remove solvents. The sample was removed and cooled back to room temperature. The measured sheet resistance was 5,250 Ohms/sq after Nafion coating, a decrease in sheet resistance of >30%.

Other embodiments and uses of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. All references cited herein, including all publications, U.S. and foreign patents and patent applications, are specifically and entirely incorporated by reference. It is intended that the specification and examples be considered exemplary only with the true scope and spirit of the invention indicated by the following claims.

References:

1. C M. Trottier, P. Glatkowski, P. Wallis, and J. Luo, Journ. of the SID 13(2005)759.

2. P. Glatkowski, " Coating Containing Carbon Nanotubes", Intern. Patent Appl., # WO02/076724 Al

3. G. Gruner et al., " Flexible Nanostructure Electronic Devices" US Patent Appl. # US205/01844641 Al.

4. R. Saito, G. Dresselhaus, and M. S. Dresselhaus " Physical Properties of Carbon Nanotubes", Imperial College Press, 1998.

5. R.E. Smalley et al, Appl. Phys. Lett. 77(2000) 666.

6. J. Hedberg, L. Dong, and J. Jiao, Appl. Phys. Lett. 2005, 86, 14311

7. K. Lee, H. Song, B. Kim, J.T. Park, S. Park, M.G. Choi, JAm.Chem.Soc. 2002, 124, 2872.

Claims

1. A method of increasing the conductivity of a carbon nanotube-containing film comprising doping the carbon nanotubes with a dopant before, during or after forming the film.

2. The method of claim 1, wherein the dopant is a binder, a polymer, an acid, a metal oxide, a salt, or a combination thereof.

3. The method of claim 1, wherein the dopant is nafion, slemion, thionyl chloride, TCNQ, oxygen, water, nitric acid, sulfurinc acid, a fluoropolymeric acid, polystyrene sulfonic acid, phosphoric acid, polyphosphoric acid, polyacrylic acid, and combinations thereof.

4. The method of claim 1 , wherein surface conductivity of the film is not significantly altered after doping.

5. The method of claim 1, wherein a surface of the film is not electrically insulated after doping.

6. The method of claim 1, wherein transparency of the film is not significantly altered after doping.

7. The method of claim 6, wherein transparency decreases by less than 10%, by less than 5%, or by less than 2% after doping.

8. The method of claim 6, wherein transparency is greater than 60%, greater than 70%, greater than 80%, greater than 90%, greater than 95%, or nearly 100%.

9. The method of claim 1, wherein the film has a sheet resistance that decreases after doping.

10. The method of claim 9, wherein sheet resistance in ohms/square is decreased by greater than 5%, greater than 10%, greater than 20%, greater than 50%, greater than 100%, greater than 200%, or greater than 500%.

11. The method of claim 1 , wherein the film has a volume conductivity that increases after doping.

12. The method of claim 1 1, wherein volume conductivity is increased by greater than 5%, greater than 10%, greater than 20%, greater than 50%, greater than 100%, greater than 200%, or greater than 500% .

13. The method of claim 1, wherein thickness of the film is not significantly altered upon doping.

14. The method of claim 13, wherein the thickness of the film after doping is increased by less than 100%, less than 75%, less than 50%, or less than 10%.

15. The method of claim 1, wherein the conductivity of the doped film is stable upon exposure to an ambient environment.

16. The method of claim 15, wherein exposure is for greater than 12 hours, greater than 24 hours, greater than one week, greater than one month or greater than one year.

17. The method of claim 1, wherein the doped film is stable upon exposure to temperatures of from -200 to 600 degrees Celsius.

18. The method of claim 1 , wherein the doped film is stable upon exposure to a humidity

19. The method of claim 1, wherein stability of the conductivity of the doped film is directly proportional to stability of the dopant.

20. The method of claim 1, wherein the doping is reversible.

21. The method of claim 1 , wherein the doping is irreversible.

22. The method of claim 1, wherein forming the film de-dopes the carbon nanotubes.

23. The method of claim 1, further comprising patterning the coating to form variable electrical conductivity across the coating.

24. The method of claim 1, wherein doping generates hydrogen production.

25. The method of claim 1, wherein the doping is performed during purification of the carbon nanotubes, during processing of the carbon nanotubes, or while the carbon nanotubes are in solution.

26. The method of claim 1, wherein, after formation of the doped film, resistance of the film is reduced by at least 10% and transparency of the film is unchanged.

27. The method of claim 1, wherein, after formation of the doped film, transparency and the conductivity of the film are both increased.

28. The method of claim 1, wherein, after formation of the doped film, transparency of the film is increased to infrared or near infrared.

29. The method of claim 1, wherein, after formation of the doped film, transparency of the film is reduced in narrow portions of the electromagnetic spectrum.

30. The method of claim 1, wherein the carbon nanotubes are at least 70% pure.

31. The method of claim 1 , wherein the carbon nanotubes are at least 80% pure.

32. The method of claim 1, wherein the carbon nanotubes are at least 90% pure.

33. The method of claim 1, wherein the carbon nanotubes are single- walled, double walled, few-walled or multiwalled.

34. A film formed by the method of claim 1.

35. A method of forming a solid electrolyte comprising introducing a dopant to the interior of carbon nanotubes.

36. The method of claim 35, wherein the dopant is introduced in a saturating amount to substantially fill said carbon nanotubes.

37. The method of claim 35, wherein the dopant is a monomer and further comprising polymerizing said monomer.

38. A solid electrolyte formed by the method of claim 35.

39. A composition comprising a carbon nanotube-containing film that is electrically conductive and optically transparent wherein carbon nanotubes within the film are doped with a superacid or a polymeric acid.

40. The composition of claim 39, wherein the superacid is nafion.

41. The composition of claim 39, which further contains a metal oxide, a polymer, a fluoropolymer, a particulate metal, or a combination thereof.

42. A method of forming a carbon nanotube-containing composition comprising: forming a thin layer of carbon nanotubes on a substrate of soda lime glass; and heating the carbon nanotube layer on the substrate such that alkali ions of the substrate dope the carbon nanotubes forming an alkali-doped carbon nanotube-containing film.

43. The method of claim 42, wherein the alkali-doped carbon nanotube-containing film has a resistivity of less than 10^~3 ohms-cm, of less than 10^"4 ohms-cm, or of less than 10^"5 ohms-cm.

44. The method of claim 42, wherein the alkali-doped carbon nanotube-containing film has a transparency of greater than 60%, greater than 70%, greater than 80%, greater than 90%, greater than 95% or nearly 100%.

45. A method of forming a coating comprising separating two transparent and conductive carbon nanotube layers with a dielectric material; and enhancing resistance of the CNT layers by applying voltage between the two CNT layers.

46. A transparent, electrically conductive coating comprising carbon nanotubes with a diameter less than 5 nm and a component that varies optoelectronic properties of the coating.

47. The coating of claim 46, wherein the component decreases sheet resistance of the coating.

48. The coating of claim 46, wherein the component increases volume conductivity of the coating.

49. The coating of claim 46, wherein the component increases conductivity of the carbon nanotube coating without affecting absorbance of light in the visible region of the electromagnetic spectrum.

50. The coating of claim 46, wherein the component increases concentration of positive charge carriers in the coating.

51. The coating of claim 46, wherein the component increases concentration of negative charge carriers in the nanotube coating.

52. The coating of claim 46, wherein the component forms a redox couple with the carbon nanotube.

53. The coating of claim 46, wherein the coating forms a solid electrolyte.

54. The coating of claim 46, wherein the component is located inside the carbon nanotube interior cavity.

55. The coating of claim 46, further comprising an additional component for improving the optoelectronic properties of the coating.

56. The coating of claim 46, wherein the component is located in interstices of carbon bundles.

57. The coating of claim 46, wherein the component is trapped inside the carbon nanotubes interior cavity by functional groups on the carbon nanotubes.

58. The coating of claim 46, wherein the component is the interstices of the carbon bundles by functional groups on the carbon nanotubes.

59. The coating of claim 46, wherein the component at least partially forms a coating around the carbon nanotubes and decreases the sheet resistance of the coating.

60. A method for doping transparent conductive carbon nanotube film wherein the film is exposed to a dopant.

61. A method for doping a transparent conductive film comprising carbon nanotubes wherein the carbon nanotubes are exposed to a dopant prior to being formed into the film.

62. A method for doping a transparent conductive film comprising carbon nanotubes wherein the carbon nanotubes are exposed to the dopant during purification.

63. A method of forming a coating comprising separating two transparent and conductive carbon nanotube layers with an insulating material and enhancing resistance of the carbon nanotube layers by applying voltage between the carbon nanotube layers.