POLYMER COMPOSITE COMPRISING THREE-DIMENSIONAL CARBON NANOTUBE NETWORKS, METHOD FOR PREPARING THE POLYMER COMPOSITE AND STRAIN SENSOR USING THE POLYMER COMPOSITE

[ Technical Field]

The present invention relates to a polymer composite comprising carbon nanotube networks. More specifically, the present invention relates to a polymer composite comprising three-dimensional carbon nanotube networks with good and uniform electrical connection, a method for preparing the polymer composite, and a strain sensor using the polymer composite.

[ Background Art] Strain sensors are apparatuses that measure strains (e.g., tensile or compressive strains) applied to various structures, including aircraft structures, automotive structures, machine tool structures, bridge structures, ship structures, etc., to effectively conduct static and dynamic tests and stability tests on the structures. A typical strain sensor is manufactured by attaching a strain gauge to a thin epoxy film and connecting the strain gauge to the epoxy film through a wire.

When an electric current is applied to the wire to produce a resistance, the resistance of the wire is inversely proportional to the cross-sectional area of the wire and is proportional to the length of the wire. When a load is applied to a test sample attached to the strain sensor, the length of the test sample is increased or decreased in response to the applied load, leading to a change in the length and cross-sectional area of the wire. This causes changes in the resistance of the wire. The strain sensor measures the resistance changes of the wire to determine the strain. This is the fundamental principle of the strain sensor.

Strain gauges based on nickel-chromium alloys have been commercially available. Such strain gauges have drawbacks in terms of flexibility and sensitivity. Another drawback is that only strains in particular directions in limited portions of samples to which the strain gauges are attached can be measured.

Under these circumstances, attempts have been made to use a mixture of a polymeric material and a carbon material to manufacture strain sensors with improved flexibility that can be applied to large areas. The first attempt was a strain sensor using a mixture of carbon black or carbon nanofibers and a polymer. However, the first strain sensor has the problem of poor mechanical properties of the materials.

To solve this problem, research on strain sensors using a composite of carbon nanotubes and a polymer has been conducted. Carbon nanotubes are allotropes of carbon that is one of the most abundant elements on the earth. Carbon nanotubes are tubular structures in which carbon atoms are bonded to each other in a hexagonal honeycomb shape and have an extremely small diameter in the nanometer range.

Carbon nanotubes are very elastic and bendable and exhibit better physical properties (for example, higher tensile strength) than any other carbon material.

Carbon nanotubes exhibit varying electrical properties depending on their rolled-up shape and diameter. Particularly, multi-walled carbon nanotubes (MWCNTs) can find application in subminiature sensors, etc. due to their high conductivity.

The strain sensors using carbon black or carbon nanotubes as carbon materials are based on the piezoresistivity of the carbon materials and are designed to measure changes in the resistance or conductivity of the carbon materials in order to sense strains applied to samples. In an area where a small strain is applied, the resistance of the carbon materials is largely dependent on changes in contact resistance between the polymer as a base material and the carbon materials. Meanwhile, in an area where a large strain is applied, the resistance of the carbon materials is changed by the disconnection of the carbon materials (e.g., the carbon nanotubes). The sensors exhibit linear response characteristics when compressive stress is applied thereto, but exhibit nonlinear response characteristics when a tensile force is applied thereto. Accordingly, the sensors may suffer from serious reproducibility problems when dynamic loads (repeated cycles of compressive stress and tensile force) are applied to the sensors.

Specifically, strain sensors using a composite of carbon nanotubes and a polymeric material can be broadly classified into two types: i) sensors that are manufactured using a mixture of a carbon nanotube powder and a polymeric material; and ii) sensors that are manufactured by applying a polymeric material to a two-dimensional planar sheet of carbon nanotubes. In either case, the carbon nanotubes are in a conducting state because they are in physical contact with each other within the polymer. This physical contact is simply due to the van der Walls forces between the carbon nanotubes, leading to weak structural strength of the sensors when an external strain is applied. Therefore, the sensors have very limited strain measurement ranges. Particularly, in the sensors using a sheet of carbon nanotubes, the connection state of the carbon nanotubes within the carbon nanotube sheet is varied continuously during repetitive measurements, and as a result, there is a very strong tendency that the signal characteristics of the sensors shift in particular directions. Further, the transparency of the sensors is severely impaired, which makes it impossible to use the sensors in electronic devices, such as touch panels.

In the sensors in which a carbon nanotube powder is dispersed in a polymer, low reliability inevitably results in because it is difficult to disperse the carbon nanotubes. A chemical substance (e.g., a dispersant) used to uniformly disperse the carbon nanotubes may remain in the final strain sensors, causing low sensitivity of the sensors. Further, the high percolation threshold of the carbon composite material requires the use of a larger amount of the carbon composite material. The increased amount of the carbon composite material impairs the transparency of the sensors, which makes it impossible to use the sensors in electronic devices, such as touch panels.

[Disclosure] [ Technical Problem] A first object of the present invention is to provide a polymer composite comprising carbon nanotube networks that are formed so as to have a three- dimensional network structure at the initial stage of preparation, thus achieving a good electrical connection therebetween.

A second object of the present invention is to provide a method for preparing the polymer composite.

A third object of the present invention is to provide a strain sensor using the polymer composite.

[ Technical Solution] According to the present invention, the first object can be accomplished by the provision of a polymer composite comprising carbon nanotubes and a polymer wherein the carbon nanotubes grow in horizontal and vertical directions to form three-dimensional networks.

In an embodiment, the three-dimensional carbon nanotube networks may be embedded in the polymer and the carbon nanotubes may be surface modified.

In an embodiment, the polymer may be selected from the group consisting of a silicone polymer, an acrylic polymer, a copolymer thereof, and mixtures thereof.

In an embodiment, the silicone polymer may be selected from the group consisting of polysilane, polysiloxane, polysilazane, polycarbosilane and mixtures thereof. In an embodiment, the acrylic polymer may be selected from the group consisting of poly(meth)acrylate, polymethyl (meth)acrylate, polyethyl (meth)acrylate, polypropyl (meth)acrylate, copolymers thereof, and mixtures thereof.

In an embodiment, the polymer may be a transparent elastomer.

In a preferred embodiment, the three-dimensional carbon nanotube networks are at least five layers of carbon nanotube networks connected to each other and stacked together in a vertical direction.

In an embodiment, the carbon nanotubes may be double-walled carbon nanotubes or multi-walled carbon nanotubes.

In a preferred embodiment, the number density of the three-dimensional carbon nanotube networks per unit volume of the polymer composite is at least

1.5/μm ³ .

In an embodiment, the polymer may be in the form of a film whose at least one surface is patterned with nanoholes.

In a preferred embodiment, the three-dimensional carbon nanotube networks are partially exposed at the entrances of the nanoholes or inside the nanoholes.

In an embodiment, the carbon nanotubes may be single-walled carbon nanotubes.

In an embodiment, the nanoholes may have a depth of 2 to 200 μm and may be spaced apart from each other at intervals of 50 to 2,000 nm. According to the present invention, the second object can be accomplished by the provision of a method for preparing a polymer composite comprising three- dimensional carbon nanotube networks, the method comprising (a) adsorbing one or more metal catalysts to a substrate formed with nanorods, (b) supplying a carbon source gas to the surface of the substrate to form carbon nanotubes having a three- dimensional network structure, (c) coating a prepolymer solution on the surface of the substrate and curing the prepolymer solution to form a polymer film, and (d) separating the polymer film and the substrate from each other.

In an embodiment, the metal catalysts may be adsorbed by dipping the substrate formed with nanorods in a Fe-Mo catalyst solution containing

Fe(NO ₃ ) ₃ -9H ₂ O and an aqueous solution of a molybdenum (MO) salt, and the concentration ratio of Fe to Mo in the Fe-Mo catalyst solution may be between 5: 1 and 0.5: 1.

In an embodiment, the method may further comprise, after step (a), annealing the substrate adsorbed by the metal catalysts and supplying ammonia (NH ₃ ) or hydrogen (H ₂ ) gas to the annealed substrate to reduce the metal catalysts. In an embodiment, the carbon source gas may be selected from the group consisting of methane gas, ethylene gas, acetylene gas, benzene gas, hexane gas, ethanol gas, methanol gas, propanol gas, and mixed gases thereof.

In an embodiment, the nanorods may have a height of 2 to 200 μm and an aspect ratio of 2 to 100 and may be spaced apart from each other at intervals of 50 to 2,000 nm.

In an embodiment, the polymer film may be transparent and may be formed of a material selected from the group consisting of a silicone polymer, an acrylic polymer, a copolymer thereof, and mixtures thereof.

According to the present invention, the third object can be accomplished by the provision of a strain sensor comprising the polymer composite.

In an embodiment, the strain sensor may comprise a device having two or more metal electrodes disposed on at least one surface of the polymer composite.

In an embodiment, the metal electrodes may be spaced apart from each other at both ends of the polymer composite and may be made of a metal selected from the group consisting of Au, Cr, Ti, Al and alloys thereof. In a preferred embodiment, the metal electrodes are in electrical communication with the three-dimensional carbon nanotube networks present within the polymer composite.

[ Description of Drawings] FIG. 1 illustrates schematic diagrams for explaining a method for forming three-dimensional carbon nanotube networks of a polymer composite according to an embodiment of the present invention;

FIG. 2 illustrates schematic diagrams for explaining a method for preparing a polymer composite comprising three-dimensional carbon nanotube networks according to an embodiment of the present invention and a method for fabricating a device of a strain sensor according to an embodiment of the present invention in which metal electrodes are deposited on the polymer composite;

FIG. 3 shows scanning electron microscope (SEM) images of carbon nanotubes formed on a silicon substrate in Example 1 -(1); FIG. 4 shows cross-sectional SEM images of a polymer composite prepared in Example l-(3);

FIG. 5 is a photograph of a strain sensor device fabricated in Example l -(4); FIG. 6 shows the average numbers of three-dimensional carbon nanotubes between two adjacent nanorods, which were measured in Test Example 1 ; FIG. 7 shows photographs of a characterization system for measuring changes in the electrical conductivity of a strain sensor device fabricated in Example l -(4) in response to an applied strain;

FIG. 8 is a graph showing changes in the electrical conductivity of a strain sensor device fabricated in Example l -(4), which were measured in Test Example 2; and FIG. 9 is a graph showing the reproducibility of electrical conductivity measured in Test Example 3.

[Mode for Invention]

The present invention provides a polymer composite comprising carbon nanotubes and a polymer wherein the carbon nanotubes grow in horizontal and vertical directions to form three-dimensional networks. No impurities, such as a dispersant, are present between the carbon nanotubes to ensure a very good and uniform electrical connection between the carbon nanotubes.

The three-dimensional carbon nanotube networks may be present on the surface of the polymer. Alternatively, the three-dimensional carbon nanotube networks may be embedded in the polymer, which is preferred because the polymer can protect the three-dimensional carbon nanotube networks from pollution and damage from external environmental factors. The carbon nanotubes may be surface modified to introduce functional groups, such as carboxyl groups or carboxyl groups, into the surface of the carbon nanotubes. This surface modification improves the interaction between the carbon nanotubes and the polymer to increase the durability of the polymer composite. Alternatively, the carbon nanotubes may be coated with a metal. This coating improves the conductivity of the carbon nanotubes. The metal can be coated on the carbon nanotubes by any suitable process known in the art, for example, atomic layer deposition. The coating process is not particularly limited.

Any commercially available polymer may be used in the polymer composite present invention. There is no particular restriction on the kind of the polymer. The polymer is preferably a flexible polymer. For example, the polymer may be selected from the group consisting of a silicone polymer, an acrylic polymer, a copolymer thereof, and mixtures thereof. Specifically, the silicone polymer may be selected from the group consisting of poly(meth)acrylate, polymethyl (meth)acrylate, polyethyl (meth)acrylate, polypropyl (meth)acrylate, copolymers thereof, and mixtures thereof. The general structures of the polysilane, polysiloxane, polysilazane and polycarbosilane are represented by Formulas 1 to 4, respectively:

[-Si(R)(R')-] _n (1) wherein R and R' are independently a hydrogen atom, a Cj-C ₅ alkyl group or a C ₅ -C ₂₀ aryl group;

[-Si(R)(R')-O-] _n (2) wherein R and R' are independently a hydrogen atom, a C ₁ -C ₅ alkyl group or a C ₅ -C ₂₀ aryl group; [-Si(R)(R')-(NR")-] _n (3) wherein R, R' and R" are independently a hydrogen atom, a C ₁ -Cs alkyl group or a C ₅ -C ₂₀ aryl group; and

[-Si(R)(R')-(C _x H _y )-] _n (4) wherein R and R' are independently a hydrogen atom, a Ci-C ₅ alkyl group or a C ₅ -C ₂₀ aryl group, and C _x H _y is a Ci-C ₆ alkylene group.

The polymer may be a transparent elastomer. The transparent elastomer is preferably polydimethylsiloxane (PDMS). PDMS is a polymer suitable for use in contact lenses due to its high flexibility, elasticity, transparency and UV stability.

The polymer composite of the present invention can be used in a variety of applications, including transparent electrodes, touch panels, strain sensors, etc. High conductivity of the polymer composite is needed for use in a transparent electrode. The polymer composite is preferably required to have semiconducting properties when used in a strain sensor. Changes in the conductivity of the carbon nanotubes in response to an applied external force are measured in the strain sensor. Since such conductivity changes are attributed to changes in the energy band gap of the carbon nanotubes, the polymer composite having semiconducting properties is preferably used to increase the sensitivity of the strain sensor.

Unlike a prior art polymer composite in which carbon nanotubes as pillars are dispersed in a polymer, the polymer composite of the present invention can achieve sufficient conductivity even when it is used in a small amount in a transparent electrode or a touch panel. That is, the polymer composite of the present invention can ensure sufficient conductivity without deteriorating the inherent mechanical properties of the polymer due to its low percolation threshold. As described above, the polymer composite of the present invention is required to have improved conductivity when used in a transparent electrode or a touch panel. To meet this requirement, it is preferred in terms of conductivity that the three- dimensional carbon nanotube networks are at least five layers of carbon nanotube networks connected to each other and stacked together in a vertical direction. It is more preferred that at least ten layers of carbon nanotube networks are stacked together. Where high conductivity is required, double-walled carbon nanotubes or multi-walled carbon nanotubes than single-walled carbon nanotubes.

The number density of the three-dimensional carbon nanotube networks present inside or on the surface of the polymer composite per unit volume is preferably at least 1.5/μm ³ . When the number density exceeds 1.5/μm ³ , improved conductivity of the polymer composite is achieved, thus enabling the polymer composite to be directly applied to various electronic devices.

When it is intended to use the polymer composite of the present invention in a strain sensor, the polymer may be in the form of a film whose at least one surface is patterned with nanoholes. The nanohole pattern may be regular or irregular. The individual nanoholes can serve as release points against external strains in the manufacture of a strain sensor. The nanoholes amplify the same strains applied to the strain sensor to improve the sensitivity of the strain sensor.

The polymer composite of the present invention may include a plurality of nanorods embedded inward from at least one surface thereof. The nanorods are not necessarily made of a conductive material. When the polymer composite of the present invention is applied to an electronic device, it is preferred that the nanorods are made of a conductive material taking into consideration the electrical communication between the polymer composite and external electrodes. The conductive material may be exemplified by silicon (Si).

On the other hand, it is preferred that the three-dimensional carbon nanotube networks are partially exposed at the entrances of the nanoholes or inside the nanoholes. This partial exposure allows the polymer composite to be in electrical communication with external metal electrodes when the polymer composite is applied to a strain sensor. The metal electrodes can be formed by suitable techniques, such as thermal evaporation. According to thermal evaporation, a metal is applied to the entrances of the nanoholes and inside the nanoholes (by deposition) so that the metal electrodes are directly connected to portions of the three-dimensional carbon nanotube networks exposed at the entrances of the nanoholes or inside the nanoholes. That is, the metal electrodes are in electrical communication with the three- dimensional carbon nanotube networks.

The carbon nanotubes of the three-dimensional carbon nanotube networks are physically/chemically bonded to each other to form three-dimensional networks. The carbon nanotubes may be single-walled carbon nanotubes, double-walled carbon nanotubes or multi-walled carbon nanotubes. When the polymer composite of the present invention is used in a strain sensor, the use of single-walled carbon nanotubes is preferred. It is believed that this preference is because changes in the conductivity of a strain sensor using the polymer composite may depend on the possibility that the energy band gap of the carbon nanotubes may be changed by an external strain rather than depend on changes in contact area between the individual carbon nanotubes or between the carbon nanotubes and the polymer.

There is no particular limitation on the depth, intervals and the diameter/depth ratio of the nanoholes that may be present on the surface of the polymer composite according to the present invention. Preferably, the nanoholes have a depth of 2 to 200 μm and are spaced apart from each other at intervals of 50 to 2,000 nm. If the nanoholes are not deeper than 2 μm, there is the risk that the nanoholes may not be in electrical contact with metal electrodes. Meanwhile, if the nanoholes are deeper than 200 μm, there is the risk that the mechanical properties of the polymer composite may deteriorate. It is not easy to space the nanoholes apart from each other at intervals of less than 50 nm. If the nanoholes are spaced apart from each other at intervals of less than 50 nm, the durability of the polymer composite may be poor. Meanwhile, if the nanoholes are spaced apart from each other at intervals of more than 2,000 nm, there is the risk that the nanoholes may not be in electrical contact with metal electrodes.

The present invention also provides a method for preparing a polymer composite comprising three-dimensional carbon nanotube networks. Specifically, the method comprises (a) adsorbing one or more metal catalysts to a substrate formed with nanorods, (b) supplying a carbon source gas to the surface of the substrate to form carbon nanotubes having a three-dimensional network structure, (c) coating a prepolymer solution on the surface of the substrate and curing the prepolymer solution to form a polymer film, and (d) separating the polymer film and the substrate from each other. According to the method of the present invention, three- dimensional carbon nanotube networks can be directly formed within the polymer composite. The intervals between the adjacent nanorods or nanoholes may be in the range of 10 nm to several tens of μm, but are not particularly limited to this range. The substrate is not particularly limited so long as it is commonly used in the art. For example, the substrate may be a silicon (Si) or silica (SiO ₂ ) substrate. It should be understood that substrates made of other materials can be used in the present invention.

FIG. 1 illustrates schematic diagrams for explaining a method for forming three-dimensional carbon nanotube networks of a polymer composite according to an embodiment of the present invention. FIG. 2 illustrates schematic diagrams for explaining a method for preparing a polymer composite comprising three- dimensional carbon nanotube networks according to an embodiment of the present invention and a method for fabricating a device of a strain sensor according to an embodiment of the present invention in which metal electrodes are deposited on the polymer composite. Referring to FIG. 1 , the three-dimensional carbon nanotube networks are formed by (a) etching a substrate to form nanorods in a three- dimensional structure, (b) dipping the etched substrate in a solution to introduce metal catalyst particles on the etched substrate, and (c) supplying a carbon source gas to the substrate, on which the metal catalyst particles are introduced, to produce carbon nanotubes in a three-dimensional network bridge structure. The etching process is not particularly limited so long as it is commonly used in the art. The substrate can be etched by suitable processes known in the art, for example, the

Bosch process. A direct growth process may also be used when the substrate is made of silicon (Si). According to the direct growth process, a catalyst is formed on the Si substrate and a Si source is supplied to grow Si nanorods directly on the Si substrate.

Referring to FIG. 2, the strain device can be fabricated by (a) preparing a substrate having a uniform array of nanorods, (b) growing carbon nanotubes in a horizontal direction on the surface of the nanorods to form a three-dimensional network bridge structure of the carbon nanotubes, (c) applying a prepolymer solution to the substrate to surround the carbon nanotubes and curing the prepolymer solution to form a polymer film, (d) separating and removing the substrate from the carbon nanotubes/polymer composite, and (e) depositing metal electrodes on portions of nanoholes through which the surfaces of the carbon nanotubes are exposed. The electrodes of the strain device are connected to a power source and a measuring instrument to measure changes in the electrical signal (conductivity) of the strain device in response to an applied strain.

When it is intended to produce carbon nanotubes by chemical vapor deposition (CVD) using one or more metal catalysts, there is a limitation that a substrate, on which the carbon nanotubes are to grow, should not be sintered together with the metal catalyst when heat is applied to grow the carbon nanotubes. For example, a silicon substrate is sintered together with Fe as a metal catalyst to form Fe _x Si _y during growth of carbon nanotubes. That is, the catalyst loses its catalytic activity for the growth of the carbon nanotubes, resulting in a decrease in the density of the grown carbon nanotubes. For these reasons, the prior art methods use silica (SiO ₂ ) substrates rather than silicon substrates. The surface of silica nanorods or nanoholes formed after etching is not electrically conductive because the silica per se is a nonconductor. In contrast, despite the use of the silicon substrate in the method of the present invention, the catalysts are protected from inactivation to grow three-dimensional networks of carbon nanotubes in high density even at the lowest portions of the nanorods or the nanoholes. It is believed that the reason why the Fe particles can be prevented from sintering despite the use of the silicon substrate is because the Mo acts as a barrier to the sintering. The composition of the Fe-Mo catalyst solution is not particularly limited. In an embodiment, the Fe- Mo catalyst solution may contain Fe(NO ₃ ) ₃ -9H ₂ O and an aqueous solution of molybdenum (Mo).

The nanorods of the silicon substrate can be formed by any suitable method commonly used in the art, for example, electrochemical etching, photolithography or direct synthesis. There is no particular restriction on the height, shape and intervals of the nanorods. In order to form three-dimensional networks of carbon nanotubes, it is preferred that the nanorods have a height of 2 to 200 μm and an aspect ratio of 2 to 100 and are spaced apart from each other at intervals of 50 to 2,000 nm. If the nanorods are not higher than 2 μm, the spaces defined by the nanorods are too narrow to form carbon nanotubes in a three-dimensional network structure.

Meanwhile, if the nanorods are higher 200 μm, there is the risk that carbon nanotubes may not be uniformly formed at the lower portions of the nanorods. If the adjacent nanorods are spaced apart from each other at intervals of less than 50 nm, they are too close to each other to form carbon nanotubes. If the adjacent nanorods are spaced apart from each other at intervals of more than 2,000 nm, there is the risk that carbon nanotube bridge networks may be difficult to form. There is a need to limit the aspect ratio of the nanorods in order to increase the number density of three-dimensional carbon nanotube networks per unit space. If the nanorods have an aspect ratio lower than 2 or higher than 100, there is the risk that the density of carbon nanotubes may decrease.

Thereafter, the patterned silicon substrate is cleaned with suitable solvents, such as acetone, ethanol and deionized water, and is treated with piranha solution, UV-ozone or oxygen plasma to modify the surface of the Si substrate into Si-OH. The -OH groups interact with the metal catalysts or the catalyst ions to prevent the metal catalysts from being separated from the surface of the nanorods in the subsequent cleaning step. The piranha solution is a mixture of sulfuric acid and hydrogen peroxide.

The molar concentration ratio of Fe to Mo in the Fe-Mo catalyst solution is between 5 : 1 and 0.5 : 1. If the Mo concentration is less than the lower limit, Fe is sintered to lose its activity, resulting in a decrease in the density of carbon nanotubes. Meanwhile, if the Mo concentration is greater than the upper limit, the

Mo cannot function as a seed for the growth of carbon nanotubes, there is the risk that the density of the carbon nanotubes may decrease.

In an alternative embodiment, the Fe-Mo catalyst solution may be a mixture of an ethanolic solution of Fe(NO ₃ ) ₃ -9H ₂ O and an aqueous solution of molybdenum (Mo). In this embodiment, sonication may be performed in a state in which the Si substrate is dipped in the catalyst solution. This sonication permits uniform adsorption of the metal catalysts on the Si substrate.

The method of the present invention may further comprise, after step (a), annealing the substrate adsorbed by the metal catalysts in a reactor and feeding NH ₃ or H ₂ gas into the reactor to reduce the metal catalysts. The annealing is conducted under vacuum or an atmosphere containing oxygen. Typically, the annealing may be conducted at about 300 to about 500 ⁰ C for about 10 to about 60 min. The reason why the annealing is performed is to remove organic/inorganic chemical substances attached to the metal catalysts and the substrate. Another reason for the annealing is to oxidize the surface of the catalyst particles. The annealing inhibits the metal catalysts from migrating at high temperatures and prevents the metal catalysts from aggregation. The metal catalysts are not sufficiently annealed at a temperature lower than 300 ⁰ C, while excessive thermal energy is created at a temperature higher than 500 ⁰ C to activate the thermal motion of the metal catalysts, posing the risk that the metal catalysts may aggregate. The oxygen-containing atmosphere for the annealing is advantageous in removing organic chemical substances, but it increases the risk that the surface of the silicon substrate may be oxidized. Despite this risk, the short annealing time minimizes the amount of the silicon oxidized to a negligible level.

As a result of the annealing, the metal catalysts are oxidized on the substrate surface. H ₂ or NH ₃ gas is fed into the reactor to reduce the metal catalyst oxides. Specifically, after the annealing, the reactor is heated to about 700 to about 900 ⁰ C while reducing the internal pressure to about 10 ^"2 torr or lower. For example, H ₂ or NH ₃ gas can be fed into the reactor when the reactor is stabilized at about 800 ⁰ C. Alternatively, the gas may be fed into the reactor while increasing the reactor temperature. The internal pressure and temperature of the reactor are not limited to the ranges defined above.

After the metal catalysts are reduced, a carbon source gas is supplied to the reactor to produce carbon nanotubes. Any suitable carbon source gas that is commonly in the art may be used without limitation. For example, the carbon source gas can be selected from the group consisting of methane gas, ethylene gas, acetylene gas, benzene gas, hexane gas, ethanol gas, methanol gas, propanol gas, and mixed gases thereof.

The carbon nanotubes may be single-walled carbon nanotubes, but are not necessarily limited thereto. For example, the carbon nanotubes may be double- walled carbon nanotubes or multi-walled carbon nanotubes. Multi-walled carbon nanotubes are advantageous due to their high conductivity.

The polymer film used in the polymer composite of the present invention is transparent and may be formed of a material selected from the group consisting of a silicone polymer, an acrylic polymer, a copolymer thereof and mixtures thereof. Preferably, the transparent polymer film is formed by coating a prepolymer solution (a composition state prior to curing) on the surface of the substrate and curing the prepolymer solution. The curing can be performed by thermal curing, photocuring or room-temperature curing. The curing is preferably performed by allowing the coated substrate to stand at room temperature for 24 hr. This room-temperature curing facilitates the release of bubbles from the prepolymer solution.

The present invention also provides a strain sensor comprising the polymer composite. The strain sensor of the present invention is characterized by high signal reproducibility, good stability, broad spectrum sensing characteristics, and very fast and accurate response and recovery.

The strain sensor of the present invention may comprise a device having two or more metal electrodes disposed on at least one surface of the polymer composite. The metal electrodes may be spaced apart from each other at both ends of the polymer composite. The metal electrodes may be made of a metal selected from the group consisting of Au, Cr, Ti, Al and alloys thereof. The strain sensor of the present invention can also be manufactured by depositing the electrodes on the polymer composite in the form of a film to fabricate a strain device and attaching the strain device to a structure such as a metal foil strain gauge.

It is preferred that the metal electrodes are in electrical communication with the three-dimensional carbon nanotube networks present within the polymer composite. An external strain applied to the strain sensor induces changes in the resistance of the carbon nanotubes, eventually resulting in a change in the conductivity of the carbon nanotubes. The strain sensor measures the conductivity changes to sense the external strain. It is believed that the conductivity changes may depend on the possibility that the energy band gap of the carbon nanotubes may be changed by the external strain rather than depend on changes in contact area between the individual carbon nanotubes or between the carbon nanotubes and the transparent polymer. The metal electrodes can be deposited by suitable processes commonly used in the art, such as thermal evaporation. According to the thermal evaporation, a metal is applied to the entrances of the nanoholes and inside the nanoholes (by deposition) so that the metal electrodes are directly connected to portions of the three-dimensional carbon nanotube networks exposed at the entrances of the nanoholes or inside the nanoholes. That is, the metal electrodes are in electrical communication with the three-dimensional carbon nanotube networks.

Hereinafter, the present invention will be explained in more detail with reference to the following examples. However, these examples serve to provide further appreciation of the invention but are not meant in any way to limit the scope of the invention.

EXAMPLES

Example 1

1 -(1): Preparation of silicon substrate formed with three-dimensional carbon nanotube networks

An n-type Si wafer was etched by common photolithography and the Bosch process to form nanorods having a height of 2 μm and a diameter of about 1 μm. The nanorods were spaced apart from each other at intervals of 1 μm. Then, the etched Si wafer was cleaned with acetone, ethanol and deionized water, treated with piranha solution for 30 min to modify the surface of the Si wafer into Si-OH, and cleaned with deionized water. An ethanolic solution of Fe(NO ₃ ) ₃ -9H ₂ O (Junsei, Japan) was mixed with an aqueous solution of a molybdenum salt (ICP/DCP standard solution, 10,000 μg/mL Mo in H ₂ O, Aldrich) to prepare a catalyst solution. The molar concentration ratio of Fe to Mo in the catalyst solution was adjusted to 4: 1. The surface-modified Si wafer was dipped in the catalyst solution to uniformly adsorb the catalysts over the entire surfaces of the wafer and the nanorods, cleaned with ethanol, and mounted in a horizontal quartz tube reactor. The Si wafer adsorbed by the catalysts was annealed in air at 400 ⁰ C for 30 min. The reactor was heated to 800 ⁰ C while maintaining the internal pressure at 1.0 x 10 ^"2 Torr or less. Then, the reactor was stabilized at a temperature of 800 ⁰ C. NH ₃ gas was fed into the reactor at 300 seem for 10 min to reduce the metal oxide catalysts to their pure metals. C ₂ H ₂ as a carbon source gas was fed into the reactor at 20 seem for 10 min to form three- dimensional networks of single-walled carbon nanotubes. At that time, the internal pressure of the reactor was maintained at 3.3 x 10 ^" ' Torr. FIG. 3 shows SEM images of the carbon nanotubes on the silicon substrate. It can be seen from the images of FIG. 3 that a three-dimensional network structure of the carbon nanotubes was formed on the silicon nanorods as supports.

l -(2): Formation of transparent polymer film on the carbon nanotubes A solution of PDMS prepolymers was applied to the surface of the silicon substrate, on which the carbon nanotubes were grown, followed by curing to mold

(i.e. transfer) the carbon nanotube networks formed on the surface of the silicon substrate in the polymer. Specifically, the silicon substrate formed with the three- dimensional carbon nanotube networks was held in close contact with the bottom of a Petri dish, and then 3 mg of a mixture of svlgard 184a (Dow Corning) and svlgard 184b (Dow Corning) (10: 1 , w/w) as the PDMS propolymers was poured into the

Petri dish to form a PDMS film having a thickness of about 2 mm. Then, the resulting structure was cured in a container under vacuum at room temperature for about 1 hr to remove air bubbles from the prepolymer solution. The cured structure was further cured on a table in a horizontally balanced position at room temperature for 24 hr to ensure the thickness uniformity of the PDMS film.

l -(3): Preparation of polymer composite

After the cursing, sonication was performed to separate the substrate and the polymer composite film from each other. As a result of the sonication, a polymer composite comprising three-dimensional carbon nanotube networks was prepared. Cross-sectional scanning electron micrograph (SEM) images of the polymer composite are shown in FIG. 4. The images clearly show that hollowed spaces (nanoholes) were left at positions of the polymer composite from which the silicon nanorods were removed. However, the shape of the three-dimensional carbon nanotube networks embedded in the polymer was not observed in the images.

Example l -(4)

Fabrication of strain sensor device

Chromium was pre-deposited to a thickness of 20 nm on the polymer composite prepared in Example 1 by thermal evaporation, and gold (Au) was deposited to a thickness of 80 nm on the chromium layer to complete the fabrication of a strain sensor device. FIG. 5 is a photograph of the strain sensor device. The photograph demonstrates that the strain sensor device was highly flexible and transparent.

Example 2

Preparation of silicon substrate formed with three-dimensional carbon nanotube networks

A silicon substrate formed with three-dimensional carbon nanotube networks was prepared in the same manner as in Example 1 -(1), except that the height of the nanorods formed on the n-type Si wafer was changed to 5 μm.

Example 3

Preparation of silicon substrate formed with three-dimensional carbon nanotube networks

A silicon substrate formed with three-dimensional carbon nanotube networks was prepared in the same manner as in Example 1 -(1), except that the height of the nanorods formed on the n-type Si wafer was changed to 7 μm.

Test Example 1

The average numbers of the three-dimensional carbon nanotubes between the nanorods

Images of the silicon substrates prepared in Examples 1 -3 were taken by scanning electron microscopy (SEM). Based on the SEM images, the average numbers of the three-dimensional carbon nanotubes between the two adjacent nanorods of the silicon substrates were counted. The results are shown in FIG. 6. The average numbers were 11 (Example 1), 17 ((Example 2) and 21 (Example 3). From these results, it can be concluded that the average number of the carbon nanotubes tends to increase with increasing height of the nanorods.

Test Example 2 Evaluation of strain sensitivity

Aluminum lead electrodes were formed on the metal electrodes of the strain sensor device fabricated in Example l -(4). The metal electrodes were connected to a measuring instrument through the Al lead electrodes. As illustrated in FIG. 7, the strain sensor device was fixed to an apparatus on which a holder was moveable horizontally by the action of a stepping motor. Then, the number of rotations of the stepping motor was adjusted to move the holder by a desired length. After a strain was applied in a direction parallel to the strain sensor device, the electrical conductivity of the strain sensor device was recorded to measure changes in the resistance (i.e. in conductivity) of the strain sensor device. Specifically, the strain sensor device was maintained for 10 sec in a state in which a strain (0.1%) was applied thereto. The application of the strain was stopped to allow the strain sensor device to return to the initial state. The strain-free state was maintained for 10 sec. This procedure was repeated five times. The same tests were conducted except that the strain was changed to 0.5, 1.0, 2.0 and 3.0%. Changes in the electrical conductivity of the device in response to the applied strains were measured. The results are shown in FIG. 8. Referring to FIG. 8, it can be confirmed that the strain sensor device showed fast response speeds and reproducible changes in conductivity from the low strain zone (0.1 %) to the very high strain zone (3.0%).

Test Example 3 Evaluation of response stability

In this test example, the response stability of the sensor device fabricated in Example l -(4) against a strain was evaluated. Specifically, the strain sensor device was maintained for 10 sec in a state in which a strain (1.0%) was applied thereto. The application of the strain was stopped to allow the strain sensor device to return to the initial state. The strain-free state was maintained for 10 sec. This procedure was repeated at least forty times. The results are shown in FIG. 9. Referring to FIG. 9, it can be confirmed that the sensor device showed very stable operational characteristics in terms of signal reproducibility against the relatively large strain (1.0%).

In the strain sensor of the present invention, carbon nanotubes are directly networked with each other in a three-dimensional space. Due to this direct networking, the polymer composite has higher mechanical strength than prior art polymer/carbon nanotube composites. When an external strain is applied to the strain sensor, the individual carbon nanotubes undergo physical deformation, which causes changes in the energy band gap of the carbon nanotubes, leading to an immediate change in conductivity. Therefore, the strain sensor is very fast and accurate in response and recovery, has high signal reproducibility and good stability, and exhibits broad spectrum sensing characteristics.

[ Industrial Applicablity]

The polymer composite of the present invention comprises carbon nanotubes having a three-dimensional network structure. No impurities are present between the carbon nanotubes because the three-dimensional network structure is formed simultaneously when the carbon nanotubes are synthesized. The absence of impurities ensures a very good and uniform electrical connection between the carbon nanotubes and makes the polymer composite highly flexible and transparent. Therefore, the polymer composite of the present invention is suitable for use in a touch panel. In addition, the strain sensor of the present invention has the advantages of high signal reproducibility, good stability, broad spectrum sensing characteristics, and very fast and accurate response and recovery.