US 20060237749 A1
Electrical devices comprised of nanoscopic wires are described, along with methods of their manufacture and use. The nanoscopic wires can be nanotubes, preferably single-walled carbon nanotubes. They can be arranged in crossbar arrays using chemically patterned surfaces for direction, via chemical vapor deposition. Chemical vapor deposition also can be used to form nanotubes in arrays in the presence of directing electric fields, optionally in combination with self-assembled monolayer patterns. Bistable devices are described.
90. An article, comprising:
an array of memory defined at least in part by an array of nanoscopic wires reversibly moveable between a first stable state and a second stable state, the wires positioned across a series of trenches.
91. The article of
94. An article, comprising:
a memory element comprising a plurality of nanoscopic wires crossing a plurality of conductors, the memory element able to rectify signals between the plurality of nanoscopic wires.
95. An article, comprising:
an array comprising a wire disposed proximate a conductor, the wire and the conductor together defining a memory element able to be switched between at least two readable states, the array free of means addressing the memory element to effect switching of the memory element between the at least two states.
96. An article comprising:
an array comprising a wire disposed proximate a conductor, the wire and the conductor together defining a memory element able to be switched between at least two readable states, the array free of auxiliary circuitry defining the memory element.
This application is divisional of U.S. patent application Ser. No. 10/973,665, filed Oct. 26, 2004, which is a divisional of U.S. patent application Ser. No. 10/812,653, filed Mar. 29, 2004, which is a continuation of U.S. patent application Ser. No. 10/033,369, filed Oct. 24, 2001, which is a continuation of International Patent Application Serial No. PCT/US00/18138, filed Jun. 30, 2000, which claims priority to U.S. Provisional Patent Application Ser. No. 60/142,216, filed Jul. 2, 1999.
This invention was sponsored by the Office of Naval Research/Defense Advanced Research Projects contract number N00014-99-0495. The government has certain rights in the invention.
The present invention relates generally to the controlled formation and/or orientation of large molecules, such as nanotubes, on surfaces, and more particularly to formation of carbon nanotubes on surfaces for making nanoscopic-scale electronic devices such as memory arrays, configurable logic and other computer elements.
During the past several decades there has been a nearly constant exponential growth in the capabilities of silicon-based microelectronics leading, for example, to tremendous advances in our computational capabilities. Silicon-based microelectronics, however, can be made only so small. That is, there is a size limitation smaller than which silicon-based microelectronics cannot be fabricated. Specifically, the National Technology Roadmap for Semiconductors (SEMATECH, Austin, 1997) suggest that silicon-based microelectronics, which are typically said to follow “Moore's Law”, will continue only to about the year 2010. At this time, two factors are expected to bring Moore's scaling to an end. First, fundamental physical limitations will be reached for both device elements and wire interconnects that will prevent current designs from functioning reliably. Second, the concurrent exponential increase in fabrication (FAB) facility cost is expected to make it uneconomical to consider increasing integration levels further (using silicon technology) even if it is physically meaningful.
These factors, and the expected benefits that could be derived from further dramatic increases in computational power in the relatively near future, have led many to consider new devices and computer architectures. In particular, there has been considerable interest in developing the concept of molecular electronics. Molecular-based electronics can in principle overcome the fundamental physical and economic limitations of silicon-based microelectronics; it is physically possible to have single molecular devices. For example, a conformational change that varies the conjugation in a molecule could behave as a switch or rectifier.
Investigation has taken place into manipulation of molecules at surfaces for electronic applications. Liu, et al, in “Controlled Deposition of Individual Single-Walled Carbon Nanotubes on Chemically Functionalized Templates,” Chem. Phys. Lett. 303 (1999) 125-129 report procedures for producing individual, short carbon nanotube segments and for their deposition on chemically functionalized nanolithographic templates. Specifically, a patterned self-assembled monolayer is formed on a surface and a carbon nanotube is adsorbed onto the surface in an orientation corresponding to the pattern. The authors also describe connection of an individual carbon nanotube between two electrodes.
Monolayers have been used to provide molecular electronic devices. Collier et al. describes the use of a Langmuir-Blodgett film of rotaxane molecules interposed between lithographically fabricated wires of micron-scale diameter (Science Vol. 285, p. 391, 1999). This system is useful for read-only memory devices, however, as the configurable elements involve irreversible oxidation of the rotaxane.
Other studies report a single carbon nanotube constructed as an electronic switch (Collins, et al., Science 278 (1997)100), and a room-temperature transistor (Tans, et al., Nature 393 (1998) 49).
To date, there has been considerable progress in characterization of the electrical behavior of individual or small numbers of molecule devices. However, a significant need exists for improvement in molecule-scale electronics, especially for integrating bistable and switchable devices for high-density memory arrays.
The present invention provides a series of nanoscopic-scale electronic elements, methods of making nanoscopic-scale electronic elements, and methods of use of nanoscopic-scale electronic elements.
In one aspect, the invention provides a nanoscopic-scale electronic device. The device is defined by an electrical crossbar array that includes at least one nanoscopic wire. The crossbar array can be of a variety of configurations such as a 1×8 array, 8×8 array, etc. The array can include contact electrodes in electrical contact with various wires, for example, by covalent attachment. Crossbar arrays provided according to the invention have densities up to about 1012/cm2.
In another aspect the invention provides techniques for making nanoscopic-scale electronic devices. In one embodiment, the invention involves forming a nanoscopic wire on a surface in a pattern dictated by a chemically patterned surface. The nanoscopic wire can be a pre-formed wire, in which case the method involves applying the pre-formed wire to the surface in the pattern. Alternatively, the nanoscopic wire can be grown on the surface in the pattern. The chemically patterned surface can be patterned to direct assembly or growth of the nanoscopic wire in a predetermined orientation useful for a particular electronic device.
In another embodiment the invention provides a method involving growing a nanoscopic wire in the presence of an electric field. The field is of intensity sufficient to orient the growth of the wire. This method can, optionally, be used in combination with a method involving growing a nanoscopic wire on a self-assembled monolayer. In all methods, nanoscopic wire growth can be carried out via chemical vapor deposition (CVD).
In other embodiments, the invention provides a method involving forming a nanoscopic wire on a surface in a pattern dictated by a mechanically patterned surface or by gas flow.
In another aspect, the invention provides methods of using electronic devices. In one aspect, a method of the invention involves providing a crossbar array comprising at least two wires in crossbar array orientation, where the wires are free of contact with each other, and bringing the wires into contact with each other. The wires are contacted at a crossbar array junction at which they are alternately brought into contact with each other and released from contact with each other. In one embodiment the wires are nanoscopic wires.
Another aspect of the present invention provides an article comprising a self-assembled monolayer defining a delineated pattern. At least two crossed wires are associated with the self-assembled monolayer in which at least one of the wires is a nanoscopic wire. In another aspect, the invention provides an article comprising an electric crossbar array comprising at least two crossed wires defining a memory element able to be switched between at least two readable states. The device is free of auxiliary circuitry other than the at least two crossed wires defining the memory element.
In another aspect, the present invention provides a method comprising switching a memory element of a crossbar array between “on” and “off” states by alternatively biasing, at similar and opposite polarity, wires that cross the array to define the element.
In another aspect, the present invention provides an article comprising an electric crossbar array comprising at least two crossed nanoscopic wires defining a memory element capable of being switched reversibly between at least two readable states.
In another aspect, the present invention provides an article comprising an electrical crossbar array comprising at least two crossed nanoscopic wires defining a memory element capable of being switched between at least two readable states. The memory element is non-volatile.
In another aspect, the present invention provides an article comprising an electrical crossbar array comprising at least two crossed wires defining a diode. The device is free of auxiliary circuitry other than the at least two crossed wires defining the diode.
In another aspect, the present invention provides a method comprising providing a mixture of metallic nanotubes in semiconducting nanotubes. The method also involves separating the metallic nanotubes from the semiconducting nanotubes.
In all of the embodiments of the invention, preferred nanoscopic wires are nanotubes.
Other advantages, novel features, and objects of the invention will become apparent from the following detailed description of the invention when considered in conjunction with the accompanying drawings, which are schematic and which are not intended to be drawn to scale. In the figures, each identical or nearly identical component that is illustrated in various figures is represented by a single numeral. For purposes of clarity, not every component is labeled in every figure, nor is every component of each embodiment of the invention shown where illustration is not necessary to allow those of ordinary skill in the art to understand the invention.
The present invention provides a series of techniques and devices involving nanoscopic-scale wires, also referred to herein as “nanoscopic wires”. “Nanoscopic-scale” or “nano” is meant to include elements of widths or diameters of less than 1 μm. Nanotubes are used as one example of small wires for use in the invention and, in preferred embodiments, devices of the invention include wires of scale commensurate with nanotubes. A “wire” refers to any material having a conductivity of any semiconductor or any metal.
One significant aspect of the invention involves electronic memory elements in which crossbar junctions (the intersections of crossed wires) are defined by at least one nanoscopic-scale wire. These crossbar junctions themselves can serve as both diode or memory elements of the device and as elements for switching between different states (“on” and “off” states). That is, a crossbar array device is provided in which one or more, preferably all, elements are free of auxiliary circuitry defining the elements. Current silicon based devices require a transistor and capacitor to store a bit in dynamic RAM, or four to six transistors are required to store a bit in static RAM. The present device is free of auxiliary circuitry, such as transistors and capacitors, addressing the elements to switch them between states. This is described more fully below. Of course, “free of auxiliary circuitry defining the elements” does not encompass electrodes and power sources or other macroscopic elements required to operate the device which are outside the nanoscopic memory element array.
In all of the exemplary embodiments described herein, circuitry including at least one nanoscopic-scale wire is described. Nanotubes, in particular carbon nanotubes, are primarily described and shown as examples of nanoscopic-scale wires. While nanotubes, and in particular carbon nanotubes, are preferred for use in the invention, any nanoscopic-scale wires can be used, including nanoscopic wires such as nanorods, nanowires, organic and inorganic conductive and semiconducting polymers, and the like. Other conductive or semiconducting elements that may not be molecular wires, but are of various small nanoscopic-scale dimension, also can be used in some instances (described more fully below), e.g. inorganic structures such as main group and metal atom-based wire-like silicon, transition metal-containing wires, gallium arsenide, germanium, cadmium selenide structures, thin metal wires formed by lithographic techniques, and the like. A wide variety of these and other nanoscopic-scale wires can be grown on and/or applied to surfaces in patterns useful for electronic devices in a manner similar to techniques described herein involving nanotubes, without undue experimentation. The criteria for selection of nanoscopic-wires and other conductors or semiconductors for use in the invention are based, in some instances, mainly upon dimension to achieve high-density memory devices, and in other instances based on dimension and robustness. The wires should be able to be formed of at least one micron, preferably at least three microns, more preferably at least five microns, and more preferably still at least ten or twenty microns in length, and preferably are less than 100 nanometers, more preferably less than 75 nanometers, and more preferably less than about 50 nanometers, and more preferably still less than about 25 nanometers in thickness (height and width). The wires should have an aspect ratio (length to thickness) of at least about 10:1, preferably greater than about 1000:1. Selection of suitable conductors or semiconductors, including nanoscopic wires, will be apparent and readily reproducible by those of ordinary skill in the art with the benefit of the present disclosure.
A preferred nanoscopic-scale wire (i.e. nanoscopic wire) for use in devices of the invention can be either a nanotube or a nanowire. Nanotubes (e.g. carbon nanotubes) are hollow. Nanowires (e.g. silicon nanowires) are solid. Whether nanotubes or nanowires or other nanoscopic scale wires are selected, it is preferred that they be selected among those that are robust.
The invention facilitates reliable connection of input/output electronic lines to nanoscopic devices in a highly integrated system such that the state of the device, or individual junctions, is readable in the presence of many other connected elements. System architectures of the invention are defect and/or fault tolerant. Single wall nanotubes (SWNTs) are particularly preferred as they exhibit unique structural, electronic, and chemical properties that are particularly suitable for molecular electronics. Structurally, SWNTs are formed of a single graphene sheet rolled into a seamless tube with a diameter on the order of 0.5-5 nm and a length that can exceed 10 microns. Depending on diameter and helicity, SWNTs can behave as one-dimensional metals or semiconductor and are currently available as a mixture of metallic and semiconducting nanotubes. Methods of manufacture of nanotubes, including SWNTs, and characterization are known. Methods of selective functionalization on the ends and/or sides of nanotubes also are known, and the present invention makes use of these capabilities for molecular electronics. The basic structural/electronic properties of nanotubes can be used to create connections or input/output signals, and nanotubes have a size consistent with molecular scale architecture.
Structural and/or chemical changes associated with preferred nanoscopic-scale wires used in the invention can modulate the properties of the wires and create electronic devices of a variety of types. In addition, covalent and noncovalent chemistry can be used to direct, with molecular precision, the assembly of these wires into structures on a massively parallel scale.
Preferred nanoscopic wires of the present invention also are individual nanotubes, preferably individual SWNTs. As used herein, “individual nanotube” means a nanotube free of contact with another nanotube (but not excluding contact of the type desired between individual nanotubes in a crossbar array or other electronic device, as described herein). Typical individual nanotubes can have a thickness as small as 0.5 nm. This is in contrast to nanotubes produced primarily by laser vaporization techniques that produce high-quality materials, but materials formed as ropes having diameters of 2-50 nanometers or more and containing many individual nanotubes (see, for example, Thess, et al., “Crystalline Ropes of Metallic Carbon Nanotubes” Science 273, 483-486 (1996), incorporated herein by reference). While nanotube ropes can be used in the invention, individual nanotubes are preferred.
The invention preferably utilizes metal-catalyzed CVD to synthesize high quality individual nanoscopic-scale wires such as nanotubes for molecular electronics. CVD synthetic procedures needed to prepare individual wires directly on surfaces and in bulk form are known, and can readily be carried out by those of ordinary skill in the art. See, for example, Kong, et al., “Synthesis of Individual Single-Walled Carbon Nanotubes on Patterned Silicon Wafers”, Nature 395, 878-881 (1998); Kong, et al., “Chemical Vapor Deposition of Methane for Single-Walled Carbon Nanotubes” Chem. Phys. Lett. 292, 567-574 (1998), both incorporated herein by reference.
One aspect of the present invention provides an article comprising an electrical crossbar array comprising at least two crossed wires. The array can comprise an n×m array in which n and m can be the same or different and each is an integer greater than 0. Thus, at least two crossed wires can comprise a 1×1 array or a much larger array. At least one of the crossed wires is a nanoscopic wire. For ultra-dense arrays, it is preferred that all of the wires in the array are nanoscopic wires to provide small device sizes. Other devices, however, may not require such ultradense arrays and to reduce costs, non-nanoscopic wires can be used in conjunction with nanoscopic wires. Alternatively, nanoropes can be used exclusively or in conjunction with nanotubes.
At least two crossed wires can define a junction, e.g., wire 201 is disposed adjacent wire 203 at junction 206. “Disposed adjacent” does not require that wires 201 and 203 be in contact with each other, but rather that wires 201 and 203 are positioned relative to each other such that they are capable of contacting each other. The wires can be free of contact with no other material intermediate the wires, or spacing between the wires can be provided with the aid of intermediary material (for example, see
In one embodiment, wires 203 and 201 are disposed adjacent each other by positioning wire 203 directly on substrate 214. Wire 203 (and 204) is referred herein as an example of a “bound” wire. Wire 203 is intermediate substrate 214 and wire 201. Wire 201 is supported above wire 203, relative to substrate 214, by supports 210. Wire 201 (and 202) is referred herein as an example of a “supported” wire. Supports 210 can be spaced periodically on either side of each bound wire (e.g., wires 203 and 204). This arrangement of supports 210 allows each junction (e.g., junctions 206 and 207) to be addressed independently of each other.
In another embodiment, wires 201 and 203 can be disposed adjacent each other by positioning wire 203 in a trench. Referring to
Referring back to
In contrast, wires 201 and 204 are in contact with each other at junction 207. “Contact” can refer to electrical contact, in which there is a conduction pathway between wires 201 and 204, or in van der Waals contact with each other. Often, for conducting wires, van der Waals contact provides electrical contact. To achieve such contact, wire 201 is capable of deformation localized at a region around junction 207. Thus, the wire of the present invention should have a sufficient Young's modulus such that the wire is capable of deformable van der Waals contact with the crossed wire at junction 207. In addition, this deformation occurs without affecting neighboring junctions, i.e., wires at neighboring junctions can independently undergo deformation or remain free of contact with a crossed wire.
It is another feature of the array of the present invention that deformation occurs via application of a stimulus. Typically, the stimulus is an electrical stimulus as provided through the at least one electrode 212. A stimulus can comprise biasing the two crossed wires with opposite polarities such that electrical attraction causes wire 201 to deform towards wire 207. In addition, wires 201 and 204 preferably are selected to have sufficient adhesion energy to maintain deformable van der Waals contact upon removal of the stimulus. Thus, no applied energy is required in preferred embodiments to maintain wires 201 and 204 in contact with each other. The benefits of this capability are discussed below with respect to the potential for the crossbar array of the present invention to provide bistable (i.e. non-volatile) elements.
It can be seen that bound wires 203 and 204 do not require deformation of any type. Supported wires 201 and 202, however, should have sufficient stiffness to remain supported above bound wires, yet have sufficient flexibility to enable deformation to contact the bound wires without breaking and have sufficient adhesion energy to maintain van der Waals contact. Furthermore, supported wires 201 and 202 should be of sufficient robustness to withstand the strain of several deformation processes. Carbon nanotubes possess all these properties and are a preferred material for supported wires of the crossbar array. Of course, carbon nanotubes can be used as the bound wires. In addition, the supported wires should have enough friction to prevent slippage on supports, particularly during deformation processes. The interaction with the support could be further enhanced through chemical modification, if required, although this is not always necessary.
A device that is capable of maintaining at least two different states, e.g., contact between crossed wires or lack of contact between crossed wires without the need for applied energy to maintain either state, is referred to as a “bistable” device. As an example, bistable device elements provided by the invention at junctions of crossbar arrays will now be described. Bistable elements of the invention include those in which each of the two states is stable or metastable. These bistable elements have, as a common feature, nanoscopic scale electromechanical switching between “on” and “off” states where the “on” state involves nanoscopic-scale wires in electrical contact at the crossbar junction and in the “off” state the nanoscopic-scale wires are free of electrical contact (separated). Reading the “on” or “off” state of any cross junction is easily accomplished by measuring the resistance/conductance of the junction. The two states in this device architecture can be easily distinguished independent of fluctuations in the “on” state resistance. The “off” state conductance is lower than a conductance of the “on” state to an extent limited only by detection sensitivity, for typical bistable structure parameters. For example, for carbon nanotubes calculations suggest a change in “on” to “off” state resistance by a factor of 103 to 106 for typical device parameters (i.e. for typical nanotube separations in the “off” state). The device architecture and function is thus highly robust.
The operation of bistable switching elements of the invention now will be described with reference to
Nanoscopic-scale devices of the invention can exhibit switching times on the order of 6-30 picoseconds, thus they can operate at a frequency of as high as about 200 GHz. Thus, the invention encompasses low frequency devices also. Moreover, the energy to switch an individual device is extremely small. Much lower power is required than for silicon DRAMs at similar frequencies. Additionally, nanotubes have extremely good thermal conductivity along their tube axes and thus are an ideal structure for removing heat from a highly integrated device.
In addition, the nanoscopic-scale device of the present invention displays stability at a wide range of temperatures, limited only by the stability of the metal contacts. For example, memory elements provided from carbon nanotubes exhibit an intrinsic thermal stability up to or higher than 1000° C. The memory elements also exhibit radiation stability and non-volatility of the stored information.
The “off” and “on” states can be read easily by measuring a resistance of the junctions. In one embodiment, a difference between the resistance between two crossed wires must be sufficient to differentiate between the “on” and “off” position.
One aspect of the invention involves the spatially-controlled formation of nanoscopic-scale wires into well-defined crossbar arrays. Two general techniques for spatially-controlled formation are described herein, one involving directed assembly of pre-formed nanoscopic-scale wires onto a surface in a desired pattern, and the second involving growth of nanoscopic-scale wires on a surface in a desired pattern directed, for example, by electric fields, highly-directional precursor flow rates (“precursor wind”), template-directed growth (growth within host channels; mechanically-directed growth), growth and assembly within host materials such as self-assembled block copolymers, inorganic host structures, and the like. Relatively simple (1×1) crossbar arrays are provided in addition to more highly integrated arrays, for example, 1×8, 8×8, 16×16 arrays, etc. The arrays can serve as memory devices, where junctions of the arrays can function as memory elements. Alternatively, the junctions can serve as diodes, for overcoming problems such as cross-talk, as described below.
Referring now to
Next, open areas in the patterned surface (the SAM-free linear region between linear SAM 14) is filled with an amino-terminated SAM that interacts in a highly specific manner with a nanoscopic wire such as a nanotube. The result is a patterned SAM, on substrate 10, including linear SAM portions 14 of methyl terminated material (methyl groups exposed at the surface) separated by a line 16 of amino-terminated SAM material. Of course, any desired pattern can be formed where regions of the amino-terminated SAM material corresponds to regions at which wire deposition is desired. The patterned surface then is dipped into a suspension of wires, e.g. nanotubes, and rinsed to create an array in which wires 18 are located at regions 16 of the SAM. Where nanotubes are used, an organic solvent such as dimethyl formamide can be used to create the suspension of nanotubes. Suspension and deposition of other nanoscopic-scale wires is achievable with easily selected solvents.
Any of a variety of substrates and SAM-forming material can be used along with microcontact printing techniques, such as those described in international patent publication WO 96/29629 of Whitesides, et al., published Jun. 26, 1996 and incorporated herein by reference. Patterned SAM surfaces can be used to direct a variety of nanoscopic wires or nanoscopic-scale electronic elements. SAM-forming material can be selected, with suitable exposed chemical functionality, to direct assembly of a variety of electronic elements. Electronic elements, including nanotubes, can be chemically tailored to be attracted specifically to specific, predetermined areas of a patterned SAM surface. Nanotubes are particularly suitable for chemical functionalization on their exterior surfaces, as is well known.
Chemically patterned surfaces other than SAM-derivatized surfaces can be used, and many techniques for chemically patterning surfaces are known. Suitable exemplary chemistries and techniques for chemically patterning surfaces are described in, among other places, international patent publication serial no. WO 97/34025 of Hidber, et al, entitled, “Microcontact Printing of Catalytic Colloids”, and U.S. Pat. Nos. 3,873,359; 3,873,360; and 3,900,614, each by Lando, all of these documents incorporated herein by reference. Another example of a chemically patterned surface is a micro-phase separated block copolymer structure. These structures provide a stack of dense lamellar phases. A cut through these phases reveals a series of “lanes” wherein each lane represents a single layer. The block copolymer is typically an alternating block and can provide varying domains by which to dictate growth and assembly of a nanoscopic wire.
Described above are techniques for assembly of pre-formed nanoscopic-scale electronic elements onto a surface. Another general technique included in the invention for creation of nanoscopic-scale electronic devices involves directed growth of nanoscopic-scale elements on a surface in a predetermined pattern.
Re-orientation of the structure such that a series of catalytic sites 44 are diametrically opposed in a direction perpendicular to catalytic sites 34, with creation of an electrical field in a direction perpendicular to nanotubes 42, and exposure to chemical vapor deposition conditions results in deposition of nanotubes 46 aligned in crossbar array orientation perpendicular to nanotubes 42.
Catalyst sites for growing nanoscopic wires can be patterned by lithographic techniques, self-assembly with or without host phases, or patterning via proximal probes (e.g. dip pen lithography).
As mentioned, only one catalyst site per wire is required. With reference to
A combination of techniques to form crossbar arrays of the invention can be used. With reference to
Nanoscopic wires can also be formed on a surface in a pattern dictated by a mechanically patterned surface. Mechanically patterned surfaces can be provided by walls, channels, trenches, capillaries or the like. Referring back to
Another example of wire growth along a substrate involves an article having a plurality of indentations and protrusions. By positioning protrusions on the substrate surfaces, cavities are formed and defined by the substrate surface and the plurality of indentations. These indentations can have the shape of capillaries. The capillary walls can be used to mechanically direct growth originating from a catalytic site or can be used to chemically pattern a surface which in turn directs growth of the nanowire.
Nanoscopic wires can also be formed on a surface in a pattern dictated by gas flow. For example, disposing a series of catalyst nanoparticles (i.e., catalyst sites) and subjecting these nanoparticles to a gas flow of high pressure can direct growth of a nanowire in the direction of gas flow. Particularly, this method provides a novel growth technique where the gas comprises a reactant for the nanowire. For example, carbon precursor gases such as ethylene, acetylene, methane or carbon monoxide can be used to grow carbon nanotubes whereas silane can be used to grow silicon nanowires. Other precursors are also well known in the art.
Referring again to
Using the techniques of the invention it is possible to achieve an active element two-dimensional density of as high as about 1012 memory elements/cm2. Thus, low or high density arrays are envisioned. These element can be used to construct non-volatile random access memory (RAM), read-only memory bits and configurable logic, among other applications. This is facilitated where an array of nanoscopic wires 42 (
As mentioned above, the invention provides a series of memory elements in a crossbar array that can be written and read (switched between “on” and “off” settings) without auxiliary circuitry addressing each element. For example,
Referring now to
Another aspect of the invention provides an article comprising an electrical crossbar array comprising at least two crossed wires defining a diode. The device is free of auxiliary circuitry other than the at least two crossed wires defining the diode. For example, if bound wires are semiconducting and support wires are metallic conductors, contact between the two wires will provide a metal/semiconductor junction, which functions as a rectifying Schottky diode at the contact junction. Thus, an on junction provides intrinsic rectifying behavior. Currently, nanotubes exist as a mixture of semiconductor and metallic conductor nanotubes. It has not been possible to separate the mixture into purely semiconducing nanotubes and metallic nanotubes. Rectifying junctions can also be provided by providing semiconductor nanowires, preferably n-type. For metallic nanotubes, a rectifying Schottky junction will result in the “on” state while for semiconducting nanotubes (which are intrinsically p-type) a rectifying p/n diode junction will result if the semiconducting nanowire is n-type doped.
The provision of diode junctions can eliminate the problem of cross talk in crossbar arrays.
As mentioned previously, providing pure semiconductor nanotubes or metallic nanotubes has not been possible to this point. Thus, another aspect of the present invention provides a method involving a mixture of metallic nanotubes and semiconducting nanotubes. The method involves separating the metallic nanotubes from the semiconducting nanotubes. It is known that metallic nanotubes have a greater polarizability than semiconducting nanotubes by about three orders of magnitude. Thus, in one embodiment, the step of separating the mixture involves subjecting the mixture to an electric field of intensity sufficient to selectively orient metallic nanotubes. Depending on the intensity of electric field, there is a critical length below which only metallic nanotubes will be aligned. Thus, the intensity can be adjusted to be in accord with a field below the critical length. Above this critical length, both types of nanotubes can be aligned by the field. Below this length, the electric field is of an intensity such that the semiconducting nanotubes remain unoriented with respect to the electric field. Once selective alignment is achieved, separation is routine.
As noted above, it is a significant advantage of the invention that junction elements can be switched between “on” and “off” states remotely, i.e., by simply addressing at least one end of each of the wires that cross to form the memory element (wires 18 and 30 in
One feature of the invention is that, with reference to
Referring now to
Another aspect of the present invention provides an article comprising an electrical crossbar array comprising at least two crossed nanoscopic wires defining a memory element capable of being switched irreversibly between at least two readable states. Thus, the device is “read-only”. In another embodiment, the article comprises an electrical crossbar array comprising at least two crossed nanoscopic wires defining a memory element capable of being switched reversibly. Thus, the devices can be used for a volatile RAM (e.g. require a power supply to keep the stored information) or a non-volatile RAM.
To quantify the bistability and switching behavior of the device element a total energy, ET, can be calculated:
where Evdw is the van der Waals (vdW) energy, Eelas is the elastic energy and Eelec is the electrostatic energy for the device. The first two terms in (1), which define the static potential, can be evaluated to assess the range of parameters that yield bistable devices.
where B is the product of the nanotube elastic modulus and geometric moment of inertia, k is the elastic modulus of the support, L is the length of the suspended nanotube,
There are several important points that can be drawn from these calculations. Foremost, is the wide range of parameters that yield a bistable potential for the device configuration. The robustness of the on/off states suggests strongly that this architecture will be tolerant of variations in structure that inevitably arise during fabrication by, for example, self assembly. Second, the differences in separation between nanotubes in the on and off states will produce large differences in resistance (i.e., I˜exp(−kd), where I is the current, k is a decay constant of order 2 Å−1 and d is the tube-tube separation in angstroms), and thus should enable reliable reading of the on/off states independent of variations in cross contact resistance. Third, the range of mechanical strains required to achieve bistability in
The effectiveness of switching the suspended nanotube devices between on and off states has been assessed by evaluating the voltage-dependent contribution of the electrostatic energy to the total energy. In this calculation, the boundary element method was used to numerically solve the Laplace equation for the complex three-dimensional geometry of the crossed nanotube device. Calculations of ET for switching a 20 nm device on and off (
This example provides a demonstration of electrostatic switching of the nanotube device.
Administration of reversible switching and the ability of the device to function as a non-volatile RAM is provided in this Example. Properties of suspended, crossed nanotube devices made from SWNT ropes were studied by mechanical manipulation (
Those skilled in the art would readily appreciate that all parameters listed herein are meant to be exemplary and that actual parameters will depend upon the specific application for which the methods and apparatus of the present invention are used. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, the invention may be practiced otherwise than as specifically described.