High strength low density fiber

Abstract

A welded nanotube fiber and a method of forming such a fiber is disclosed. The method comprises applying a voltage or current signal to a nanotube bundle, either in an evacuated chamber or in a chamber in which additional process gasses are introduced. The fiber comprises a plurality of substantively parallel nanotubes, in which a plurality of the subtantially parallel nanotubes has been covalently bonded to other nanotubes in the fiber by welding.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to U.S. Provisional Patent Application Ser. No. 60/875,047, entitled “A High Strength Low Density Fiber” filed Dec. 15, 2006.

FIELD OF THE INVENTION

This disclosure relates to an apparatus comprising a high strength, low density, conductive fiber, and a method of fabricating such an apparatus. More specifically, in embodiments, the apparatus comprises nanotubes spun into a fiber and subsequently welded together to significantly strengthen the fiber. In embodiments, the method includes current welding the spun fiber, in an environment which promotes the active welding of the nanotubes while discouraging damage to the tubes such as oxidation or vaporization. In embodiments, the welded fiber can be subjected to other processes, such as the incorporation of the welded fiber into a polymer matrix, or threaded together to form a fabric.

BACKGROUND

Significant interest in carbon nanotubes has occurred since Iijima demonstrated in 1991 that carbon nanotubes are formed during arc-discharge synthesis of C60 and other fullerines. The subsequent introduction of catalytic plasma-enhanced chemical vapor deposition (PECVD) enabled better control over the growth of carbon nanostructures, which are now very actively researched. Carbon nanotubes have interesting properties and the potential to enable a vast array of technological advancements in electromechanical devices such as sensors, textiles such as stain repellants, electrical devices such as transistors, and energy converters such as solar cells. Nanotubes exhibit remarkable properties, such as their ability to tolerate high electric current density; their semi-conducting or metallic electrical characteristics (depending on the type of nanotube), their high thermal conductivity, their high modulus and tensile strength in the direction of their long axis. These properties make them ideal candidates for key elements in the next generation of numerous electronic, thermal, and mechanical devices.

Significant advance in the growth of ordered nanotubes has occurred recently with the advent of PECVD grown nanotube “forests”, in which a high density of nanotubes are grown in a vertically oriented and order fashion from a common substrate. A nanotube forest might be thought of as a “marine's haircut” or a “bed of nails”, reflective of the parallelism and approximate common height of the nanotubes grown from the common substrate. A method of harvesting nanotubes from such a forest and spinning the tall structures into a yarn has been described in “Multifunctional Carbon Nanotube Yarns by Downsizing an Ancient Technology”, M. Zhang, et al, [Zhang] and in U.S. Pat. No. 6,979,709, by Smalley, et al., both of which are hereby incorporated in their entirety for reference.

SUMMARY

A welded nanotube fiber and a method of forming such a fiber is disclosed. The method comprises applying a voltage or current signal to a nanotube bundle, either in an evacuated chamber or in a chamber in which additional process gasses are introduced. The fiber comprises a plurality of substantively parallel nanotubes, in which a plurality of the substantially parallel nanotubes has been covalently bonded to other nanotubes in the fiber by welding.

DETAILED DESCRIPTION

Macroscopic Carbon Fiber fabricated using the spinning technique described in Zhang reportedly exhibits a strength of 460 MPa. The strength of such yarns are limited by the forces binding together the individual nanotubes, and fall short of the theoretical strength of a single nanotube, reported be between 30 and 150 GPa. In fact, according to Zhang, frictional forces may obey classical laws of strength of yarn, and the ratio of yarn tensile strength (σy) to the tensile strength of the component fibers (σf) may be given by

σ_y/σ_f≈cos²α[(1−(k cosec α)],

“. . . where k=(dQ/μ)½/(3L), α is the helix angle that fibers make with the yarn axis, d is the fiber diameter, μ is the friction coefficient between fibers, L is the fiber length, and Q is the fiber migration length, i.e. the distance along the yarn over which a fiber shifts from the yarn surface to the deep interior and back again.” Although not exact, a reasonable estimate of this equation applied to the spun nanofibers of Jiang suggests the strength of the fibers should be about ½ the strength of the individual tubes. However, experimentally, the fibers have significantly less strength than this, at an apparent two orders of magnitude lower than the reported theoretical nanotube strength.

In a preferred embodiment, the macroscopic carbon fiber produced as described herein is made up of an aggregate of large number of twisted nanotubes in close proximity, as one might expect from a twine being comprised of a number of threads or strands twisted together. While the twisting angle of individual nanotubes may be quite high, say as high as 45°, the fiber has a clear long range orientation along the direction of the axis of the fiber. The term long range is used here in a relative sense, i.e. long range in this case would be a distance of at least many multiples of the fiber diameter, such as 100 times or 1000 times the fiber diameter.

In another preferred embodiment, the macroscopic carbon fiber produced as described herein is made up of an aggregate of a large number of untwisted nanotubes in close proximity. In yet another preferred embodiment, the macroscopic fiber produced as described herein is made up of nanotube fibers formed by dispersing the nanotubes in a surfactant solution, and then recondensing the nanotubes in the steam of a polymer solution. The raw material used in this process can be produced with an electric-arc technique, which produces nanotubes in the form of bundles. The material is sonicated in an aqueous solution of sodium dodecyl sulfate (SDS), which adsorbs at the surface of the nanotube bundles.

Efforts to increase the strength of the nanotubes have often focused on the binding of the nanotubes within a polymer matrix, a process which has reportedly resulted in a fiber strength of about 4 GPa, according to Zhang. Evidently, this strength is still one order of magnitude lower than the theoretical strength of the nanotubes themselves.

The density of nanotubes is low compared to other materials commonly used in high strength applications, such as steel. For example, the density of nanotubes may be around 0.8 g/cm³, compared to the density of steel, around 7-8 g/cm³. Thus a compelling use for nanotubes is those applications which demand high strength/weight or high strength/density ratios, such as aerospace applications. Other applications may make use of other properties of nanotubes, such as the reported high electrical conductivity of nanotubes. For example, if a cable using nanotubes could be made to conduct the same amount of electrical current as a presently commercially available cable, but where the nanotube-based cable is lower in weight, the nanotube-based cable could have a higher commercial value than the presently commercially available cable. Thus improving the strength/weight ratio of materials, including nanotube-based materials, is desirable.

In light of the above-described disadvantages associated with using macroscopic carbon fibers fabricated from pulling and/or spinning from a nanotube forest, the present invention is generally related to a welded fiber, made by a process which preferably includes the use of a macroscopic carbon fiber bundle, in which the nanotubes are in substantively parallel alignment and in close proximity to one another such that essentially all nanotubes in the fiber are in intimate contact with at least one other nanotube. The welded fiber may be completely welded, such that any nanotubes comprising the fiber are covalently bonded contiguously throughout the entire length of the fiber, or may be partially welded, such that at least one nanotube harvested from a nanotube forest is welded to another, different, nanotube, harvested from the same or another nanotube forest. The welding process preferably results in at least one additional, new, covalent bond adjoining the two nanotubes.

Adherence of two nanotubes has been reported by Hirayama, et al., in “Nanospot welding of carbon Nanotubes” [Hirayama], the disclosure of which is hereby incorporated by reference. Hirayama reports on the “. . . study of electrostatic touching of SWNTs . . . ”. Hirayama reportedly accomplishes this by touching a tungsten tip to a molten tin sample and retracting the tip slowly, forming self-aligned tin apexes in a “. . . homemade STM apparatus operating in a commercial TEM.” Then he drops a solution of SWNTs dispersed in ethanol onto the apexes outside the TEM. He indicates that this results in layers of SWNTs covering the apexes. To carry out the step reported to result in welded nanotubes, he indicates that “Finally, the aligned apexes with the SWNT layers were brought to a distance of 30 nm, and bias voltage was applied to the tip in the TEM-STM.”

In order to demonstrate adherence between nanotubes, Hirayama indicates: “Applying a bias voltage above 2.0 eV to the tip resulted in the straight bundles jumping into contact with the bundle loop on the sample. The bundles stuck to each other and the bundle loop was pulled so as to rotate around its vertical axis {. . . }. Once the straight SWNT bundle had contacted the looped bundle, it did not detach when the bias voltage was lowered to 0 V {. . . }. The looped bundle was dragged by the piezo motion of the straight bundle {. . . }. Meanwhile, the straight bundle did not make contact with the looped bundle at bias voltage below 1.5 V . . . . The straight bundle moved freely with the piezo motion with no relation to the looped bundle.”

Furthermore, Hirayama demonstrates that current travels through the tips at the moment enough voltage is applied to bring the nanotubes into contact. Although the current is roughly constant throughout the duration of the applied voltage, the IV response of the current is non-linear, suggesting the existence of a junction.

The adhesion of singular nanotubes has also been reported by Cox, et al., in “Study of the current stressing in nanomanipulated three-dimensional carbon nanotube structures” [Cox], the disclosure of which is hereby incorporated by reference. In Cox is reported the “. . . fabrication of free-standing carbon nanotube structures.”. Cox reports running current through arc-discharge-produced nanotubes loaded into a scanning electron microscope (SEM) configured with electrical feedthroughs connected to a pair of sharp tungsten tips and the specimen substrates. A voltage is applied to a given nanotube, causing current to flow. When a voltage of about 4.5V is reached, the signal is oscillated. With each oscillation the voltage needed to maintain a constant current decreases to a value of about 2.5 V. At that voltage, the current becomes stable with further oscillations. A resistance of the joined nanotube of as low as 12 kΩ is achieved, which is comparable with resistances reported on singular nanotubes welded on each end to titanium electrodes, as reported by Chen, et. al., in “Ultrasonic nanowelding of carbon nanotubes to metal electrodes”, the disclosure of which is hereby incorporated by reference.

In a preferred embodiment, the starting material is a fiber comprising a plurality of nanotubes maintained in a basically parallel grouping, and held in proximity by forces such as Van der Waals forces. The generally parallel grouping could be held together by a secondary material, such as a polymeric matrix in which the grouping is embedded. The nanotubes can be tightly packed or loosely packed in the grouping, but the grouping should be close enough together that the fibers do readily fall apart when the fiber is handled. It is further stipulated that essentially all of the nanotubes in the bundle are in intimate contact with at least one other nanotube in the bundle. In this context, intimate contact approximately means close enough to be affected by Van der Waals or frictional forces for a distance along the length of the nanotubes for more than one half the length of the shorter nanotube.

The fiber preferably comprises a sufficient number of nanotubes that it is visible to the naked eye. For example, the fiber should not have a smaller diameter, or smallest cross section of 5 μm, but could have a larger smallest cross-section, such as 25 □m. In embodiments, the fiber is made by a process comprising being pulled from a nanotube forest, either using a procedure which includes twisting during the pulling process, or one in which a group of nanotubes is pulled from the forest without twisting. For example, Zhang reported that a nanotube forest formed using a silicon wafer as a substrate, with about 100 μm high nanotubes, is capable of producing 50 meters of a 2 μm diameter fiber per square cm. Further, fiber of any cross-sectional size can threaded into fiber of greater diameter using standard thread-making techniques.

In a preferred embodiment, the fiber is then subjected to a current in a chamber effectively isolated from atmospheric air. The chamber can be evacuated to a low pressure, such as 10-11 torr or 10-10 torr or 10-9 torr, or different gasses can be introduced into the chamber during processing. For example, H₂ may be introduced into the chamber during the application of current. Gaseous H2 can combine with gaseous O₂ to form water vapor. This combining process can “scour” the existing atmosphere in the chamber for O₂, which is detrimental to nanotubes. Gaseous O₂ can oxidize nanotubes by reacting with the carbon to for CO₂.

In order to foster an environment favorable for nanotube growth, a growth gas such as methane (CH₄) can be introduced into the welding chamber, either in place of H₂, or in addition to H₂. Methane is one of the gasses used to supply carbon to grow nanotube forests, typically using a silicon substrate with very thin (1 nm) titanium or tungsten layers, which form islands upon annealing. The islands act as catalysts and determine the density of the forest. The presence of methane during the welding process may provide supplemental carbon to promote the replacement of vacancy defects in the nanotube contact points with a carbon atom, rather than another type of atom.

The application of current to the fiber can be carried out by connecting two points on the fiber to electrical leads which traverse the chamber from the interior to the exterior. The length of fiber between the two leads will be the portion of the fiber which is modified. After the chamber is sealed and prepared with the desired evacuation and gas flows and pressures, the signal may be applied in various fashions. For example, the signal may be applied using a simple application of constant voltage, in which the current is adjusted to maintain the applied voltage, the application of an oscillating voltage in which the current is adjusted to maintain the applied voltage, or by the application of a controlled current. The controlled current can be constant or alternating, or any combination of both. In a preferred embodiment, the current is maintained at a value less than that current which results in nanotube degradation. One such current is 80 μA.

The duration of the signal should be enough to achieve a steady state response. For a voltage application, the response is in steady state when the current is in proportion to the applied voltage. For the current application, the response is in steady state when the voltage is proportional to the applied current. It is expected that the response signal will change with repeated application of the applied signal as the welding is taking place. When the welding is complete (that is, for the applied conditions), the response signal will be a function of the applied signal. In other words, the fiber will no longer exhibit hysteresis.

Generally, the macroscopic carbon fiber produced in a preferred embodiment begins with a fiber which is envisioned to consist of a sufficient number of entwined nanotubes that it is large enough in diameter to be handled as an individual fiber. Further processing of such a fiber, such as further entwining with additional such fiber, and/or with other fibers which have not undergone the strengthening, conductivising procedure of the preferred embodiment, is also contemplated.

The macroscopic carbon fiber of the preferred embodiment is expected to be at least several centimeters in length, and could be orders of magnitude longer, as long as tens of miles or longer. In one application, the fiber may be used as an integral component of a space elevator, a cable of tens of thousands of miles long, intended to extend from a point at the surface of the earth to a point well past geosynchronous orbit, which occurs at an altitude of about 23,000 miles from the Earth's surface. The elevator stays “up” by virtue of competing forces between that caused by the centripetal acceleration of the Earth's rotation, and gravity. A space elevator requires a cable to have an extremely high strength to density ratio, so that its own weight will be supported under the effect of gravity.

In other applications, the fiber may be used as a low weight conductive current carrier for use in any application where weight must be strictly controlled, such as aerospace applications. In the case of aerospace applications, the fiber may have a length of several centimeters to several tens of meters. Like other strong fibers, the fiber of the preferred embodiment may be embedded into a matrix of other material, such as a polymeric matrix.

A preferred embodiment of this fiber provides much higher strength that the fiber before the application of the strengthening mechanism, and much higher strength to density than the fiber before the application of the strengthening mechanism, when the pre-treated fiber is embedded in a polymeric matrix. The carbon nanotube fiber of the preferred embodiment provides extremely high tensile strength at low weight, for example, 10 GPa/(g/cm³) or 100 GPa/(g/cm³). Presently, some of the highest strength/density products which are commercially available have less than about 3.5 GPa/(g/cm³). Exemplary commercially available materials include Zylon, available from Toyobo (Tsuruga-city, Japan), Spectra, available from Honeywell (Morristown, N.J.), and T-1000, available from Toray Carbon Fiber America (Flower Mound, Tex.). By way of comparison, steel piano wire has a strength-to-density ratio of about 0.4 GPa/(g/cm³). The reported strength of the pre-treated fiber has a tensile strength of 0.5 to 0.6 GPa before being embedded in a polymer matrix, and a reportedly achieved value of about 4 GPa has been reported for nanotubes embedded in a high strength polymer matrix. The reported density of the pre-treated fiber is 0.8 g/cm³, which would result in a strength-to-density ratio of about 0.625 GPa/(g/cm³). Assuming the polymer does not change the fiber density, the reported strength of 4 GPa for the embedded structure is expected to have a strength-to-density ratio of 5 GPa/(g/cm³).

In a preferred embodiment, the treated fiber has an electrical conductivity greater than that of the untreated fiber. In embodiments, it has a conductivity greater than twice that of the untreated fiber, and in additional embodiments, the treated fiber has a conductivity of greater than three times the conductivity of the untreated fiber. High conductivity at low weight is particularly useful in the aerospace industry, where conductive cabling in aircraft constitutes a significant amount of weight.

In a preferred embodiment, the treated fiber has a thermal conductivity greater than that of the untreated fiber. In general the carbon fiber of the preferred embodiment will exhibit improved properties over conventional carbon fibers because the large collection of twisted nanotubes of the pretreated fiber are held together primarily by proximity forces such as friction and Van der Waals forces, whereas the nanotubes of the post treated fiber of the preferred embodiment are welded into a smaller plurality of covalently bonded, much longer, twisted nanotubes that extend over macroscopic distances.

Applications of the post-treated fibers of the preferred embodiment include all those that would be available for the pre-treated fiber, including strengthening fibers, conductive fibers which can be further incorporated into insulative or protective sheathing. Post-treated fibers of the preferred embodiment can also be further spun into fibers of greater diameter, or used as threads to form textiles.

The following detailed example describes a welding process applied to a twisted nanotube carbon fiber comprised of twisted nanotubes numbering about 100,000 through any given cross-section of the fiber.

EXAMPLE 1:

A twisted fiber comprising nanotubes of individual length 100 μm, spun into a fiber of diameter 5 μm, has a length of 2 meters. The fiber is placed inside a chamber capable of being evacuated, such that each end of the fiber is attached to electrodes which feed through the chamber walls, and are suitable for the attachment of an electrical signal. The chamber is sealed and evacuated, and the leads are electrically connected to a current supply, such as the Keithley 238, available from Keithley instruments (Cleveland, Ohio). The Keithley 238 can be controlled by controlling software such as Labview™, available from National Instruments (Ausin, Tex.). The chamber is evacuated to 10-11 torr, and an oscillatory voltage signal at a frequency of 0.2 Hz is applied to the leads. Current is regulated such that it does not exceed 10 μA. The current is monitored, and when the current cycles through the same range with each successive voltage cycle, the voltage signal is cut off.

The present invention has been described with reference to preferred embodiments. However, it will be readily apparent to those skilled in the art that it is possible to embody the invention in specific forms other than as described above without departing from the spirit of the invention. The preferred embodiment is illustrative and should not be considered restrictive in any way. The scope of the invention is given by the appended claims, rather than the preceding description, and all variations and equivalents which fall within the range of the claims are intended to be embraced therein.

Claims

1. A welded nanotube fiber, comprising a plurality of substantively parallel nanotubes, in which a plurality of the substantially parallel nanotubes has been covalently bonded to other nanotubes in the fiber by welding.

2. The fiber of claim 1, wherein the fiber has a strength-to-density ratio of greater than 10 GPa/(g/cm3).

3. The fiber of claim 1, in which the fibers are twisted.

4. The fiber of claim 1, in which the fibers comprise a macroscopic carbon fiber bundle, wherein substantively all the nanotubes are in intimate contact with at least one other nanotube.

5. The fiber of claim 1, in which the fiber is at least 5 μm in cross-section everywhere along it length.

6. The fiber of claim 1, in which the nanotubes are harvested from at least one nanotube forest.

7. A method of forming a welded nanotube fiber, comprising applying a voltage or current signal to a macroscopic nanotube bundle, the bundle comprising a plurality of nanotubes in substantively parallel alignment, wherein substantively all the nanotubes are in intimate contact with at least one other nanotube in the bundle.