Quantum tunneling devices and circuits with lattice-mismatched semiconductor structures

Information

  • Patent Grant
  • 8216951
  • Patent Number
    8,216,951
  • Date Filed
    Monday, December 20, 2010
    14 years ago
  • Date Issued
    Tuesday, July 10, 2012
    12 years ago
Abstract
Structures include a tunneling device disposed over first and second lattice-mismatched semiconductor materials. Process embodiments include forming tunneling devices over lattice-mismatched materials.
Description
FIELD OF THE INVENTION

This invention relates to methods and materials for formation of structures including tunneling devices built on alternative materials (i.e., non-silicon (Si)). More particularly, it relates to resonant tunneling devices and circuits using lattice-mismatched semiconductor heterostructures.


BACKGROUND

Electronic devices utilizing quantum tunneling mechanism are potential candidates to replace conventional Si-based devices as a new type of logic device, or to enhance the performance of conventional Si-based devices and circuits when used together with the conventional Si devices. Quantum tunneling devices inherently have very high speeds and also can enable more functionality. Their negative differential resistance (NDR) characteristics result in an inherent latching mechanism that enables very compact circuit configurations. They may be utilized to construct high speed and low power dissipation logic and memory circuits. A tunneling device-based circuit also has a much smaller footprint since fewer devices are needed in constructing a functional circuit, in comparison to a conventional circuit.


SUMMARY

The embodiments discussed in detail below allow integration of quantum tunneling devices into a variety of technologies and materials. For example, using III-V materials for resonant tunneling diodes (RTDs) in conjunction with circuits built over Si-based substrates allows the implementation of a wide variety of applications with increased performance and higher circuit densities. Si-based applications that benefit from III-V based RTD integration include static random access memory (SRAM), dynamic RAM (DRAM), digital signal processing (DSP) and field programmable gate array (FPGA). These represent a few examples of chips and applications using just one type of tunneling device.


Examples of various quantum tunneling devices include: Esaki diodes (also called tunnel diodes); single-barrier tunnel diodes; resonant tunneling diodes (RTD), which are typically use intraband tunneling with double-barriers; triple-barrier or multiple-barrier resonant tunneling diodes; resonant interband tunneling diodes (RITD); single-barrier interband-tunneling diodes; resonant tunneling transistors (RTT); resonant tunneling field-effect transistors (RTFET), double electron layer tunnel transistors (DELTT); quantum-well-based resonant tunneling transistors (QWBRTT); resonant tunneling bipolar transistors (RTBT or RBT); resonant tunneling hot-electron transistors (RHET), etc.


Of these devices, the RTD is one of the most basic device structures. Referring to FIG. 1, the RTD is a conventional two-terminal diode 20, as described in U.S. Pat. No. 5,825,049. The illustrated RTD includes a cathode 22, a bulk emitter 24, an undoped layer 26, a two-dimensional resonant electron layer 27, thin barriers 28, another undoped layer 29, a bulk collector 30, and an anode 32, all disposed over a substrate 34. The RTD is typically realized in III-V semiconductor material system, where heterojunctions create a quantum well, switched by two barrier layers that enable resonant tunneling. Their current-voltage relationship exhibits a characteristic “N” shape, and thus a characteristic NDR region. This characteristic originates from abrupt changes in the probability of quantum mechanical tunneling through a potential barrier with applied bias over the diode. A variation of the RTD is an RITD. A three-terminal RTT may be constructed by adding a control terminal to an RTD structure. A resonant tunneling structure and mechanism may be integrated into a three-terminal transistor device in many other ways.


Referring to FIG. 2, a DELTT transistor, including a thin tunnel barrier 40, is positioned between two two-dimensional quantum wells (QW) 45, 50. In operation, electrons quantum mechanically tunnel from one well to the other through the barrier. Electrical contact is made by a top layer contact 52 to the top QW 45 only by the source, and by a bottom layer contact 54 to the bottom QW 50 only by the drain, in both cases accomplished by using the top and back depletion gates 55, 60 to remove electrons from the QW that one does not wish to contact. Tunneling between the two QWs occurs only when electrons in both wells have substantially identical momentum and energy; this may be controlled by the top control gate. A control gate 65 is disposed over the structure. A total thickness of the semiconductor layers is kept less than 2 microns, allowing the back depletion gate to be brought close to the QWs and the entire device to be made small. The illustrated device includes two gallium arsenide layers, each 150 angstroms thick, separated by a 125 angstrom aluminum gallium arsenide barrier. At left is shown the energy band diagram 65 of the structure, containing the two quantum wells separated by the thin tunnel barrier 40.


The resonant tunneling device is one of the potential emerging logic devices to continue functional scaling of electronics integrated circuits beyond the end of Si-based CMOS technology roadmap. This class of devices may include RTT and hybrid tunneling devices or circuit units that include both tunneling devices and conventional devices such as MOSFETs or HEMTs, etc. For example an RTD-FET includes several RTDs and one or more conventional FETs.


A hybrid system can potentially provide more computational power and/or functionality since it requires fewer devices in comparison to the conventional logic circuit and it is inherently faster due to the RTD's intrinsic high switching speed and negative differential resistance property, and yet it uses less power than conventional devices. Referring to FIG. 3, for example, by connecting a conventional FET with two RTDs in series, one forms a static random access memory (SRAM) structure. The illustrated tunnel diode SRAM has two stable operating points 300, 300′ that may be used to represent “0” and “1” in memory operation. This device has the potential to reduce the cell size in comparison to the conventional 6-transistor SRAM (6-T SRAM). It is refresh-free and offers a low power consumption advantage. When multiple RTDs are connected together, the SRAM cell offers multi-value memory operation.


Another example is a comparator circuit. A conventional latched comparator may include 12 high electron mobility transistors (HEMTs) and 6 Schottky diodes. An RTD-based latched comparator has been fabricated with 2 RTDs and 2 HEMTs, with latching behavior being inherent to RTD devices. The unit area is reduced to one sixth in comparison to the conventional comparator.


One of the challenges of tunneling devices is that they are conventionally fabricated in III-V material systems, and thus typically are difficult to integrate into Si-based systems. Also the cost is typically high, which limits their widespread application.


Fabricating resonant tunneling devices on a Si-based substrate is highly desirable. Recently, several group IV-based, i.e., Si/SiGe-based, tunneling devices have been developed that enable the integration of the tunneling devices into Si systems, including a tri-state logic devices fabricated in SiGe, a Si-based field-induced band-to-band tunneling transistors, Si/SiGe RITDs. However, there are new technical challenges with those structures. For example, Si/SiGe RITD devices typically require epitaxial growth in low temperature molecular beam epitaxy (MBE) systems, instead of metal-organic chemical vapor deposition (MOCVD) systems that may be unable to achieve the preferred high n-type phosphorus doping and also suffer from phosphorus poisoning issues. Moreover, the Si/SiGe RITD also typically has lower performance than an optimized III-V RITD device.


There is thus a need to develop new methods and structures to fabricate tunneling devices on Si substrates with improved material systems, such as III-V and II-VI material. There is also a need to develop new methods to fabricate alternative semiconductor materials for tunneling devices on Si substrates using improved material growth techniques, such as, for example, CVD. There is also a need to develop new structures and methods to incorporate tunneling devices in circuits on Si substrates for various functions.


Embodiments of the present invention include methods and structures for fabricating tunneling devices on a first type of substrate, e.g., group IV, with improved material systems, e.g., III-V and II-VI materials. The manufacturing process may be compatible with conventional CMOS processes utilizing Si substrates. The process flow may include an aspect ratio trapping (ART) epitaxial technique.


In another embodiment, manufacturing methods use epitaxy techniques, including MBE and CVD systems, for ART. CVD has a number of advantages, including low defect density, high quality regrowth capability, rapidly varying alloy compositions, and improved manufacturability due to CVD's higher throughput, relatively short downtimes, and scalability to very large reactors


Embodiments of the invention also include new structures and methods for functional circuits incorporating both tunneling devices and conventional Si-based devices. Such circuit units may function as, e.g., a logic circuit, an SRAM or a DRAM.


In an aspect, the invention includes a structure. The structure has a non-crystalline material defining an opening having a sidewall disposed above a surface of a substrate that comprises a first crystalline semiconductor material. The opening has a non-crystalline sidewall. A second crystalline semiconductor material, lattice-mismatched to the first crystalline semiconductor material, is disposed in the opening. A tunneling device is disposed over at least a portion of the second crystalline semiconductor material.


One or more of the following features may be included. The tunneling device may be, e.g., an Esaki diode (tunnel diode), a single-barrier tunnel diode, an RTD, a triple-barrier or multiple-barrier resonant tunneling diode, an RITD, a single-barrier interband-tunneling diode, an RTT, an RTFET, a DELTT, a QWBRTT, an RTBT or RBT, or an RHET.


The structure may comprise a memory cell, with a phase change layer or a ferroelectric material disposed over the tunneling device. The structure may include memory circuitry, such as SRAM circuitry, DRAM circuitry, or non-volatile memory circuitry, that includes the tunneling device.


The first crystalline semiconductor material may include a group IV element or compound, a II-VI compound, and/or a III-V compound, and the second crystalline material may include a group IV element or compound, a II-VI compound, and/or a III-V compound. The first crystalline semiconductor material and the second crystalline semiconductor material may include materials selected from different groups.


A majority of defects arising from the lattice-mismatch of the first and second crystalline semiconductor materials may terminate within the opening, e.g., below a height of the opening. The width of the opening may be less than the height of the opening. The tunneling device may be in contact with at least a portion of the second crystalline material. A circuit component, such as a MOSFET, a MISFET, a HEMT, a capacitor, or a resistor, may be coupled to the tunneling device and disposed above a region of the substrate. The tunneling device may include a tunneling diode disposed above a transistor. The circuit component may include a gate array, a memory circuit, and/or a processor circuit.


The tunneling diode may be disposed above a source or drain of a transistor. A second tunneling diode may be disposed over the source or drain of the transistor.


In another aspect, the invention features a method for forming a structure. The method includes the formation of an opening above a region of a substrate comprising a first crystalline semiconductor material. The opening is defined by a dielectric sidewall and an exposed surface of the substrate. A second crystalline semiconductor material lattice-mismatched to the first crystalline semiconductor material is formed in the opening. A tunneling device is defined over at least a portion of the second crystalline semiconductor material.


One or more of the following features may be included. The tunneling device may be, e.g., an Esaki diode (tunnel diode), a single-barrier tunnel diode, an RTD, a triple-barrier or multiple-barrier resonant tunneling diode, an RITD, a single-barrier interband-tunneling diode, an RTT, an RTFET, a DELTT, a QWBRTT, an RTBT or RBT, or an RHET. The second crystalline material may be formed in the window by selective epitaxy. The opening may be formed above a region of a transistor. The second crystalline semiconductor material may include a II-VI compound and/or a III-V compound.


In still another aspect, the invention includes a structure integrating a tunneling diode and a transistor using lattice-mismatched materials. The structure includes the transistor formed proximate a substrate including a first crystalline semiconductor material. The tunneling diode is coupled to one of a source or a drain of the transistor, and the tunneling diode is formed using a second crystalline semiconductor material lattice-mismatched to the first crystalline semiconductor material. At least a portion of the second semiconductor material is disposed within an opening disposed above the substrate. The opening has a non-crystalline sidewall and configured to trap a majority of defects arising from the lattice mismatch between the first and second semiconductor materials.


In another aspect, the invention includes a method of reducing SRAM circuit surface area. The method includes the formation of an RTD-based static random access memory (SRAM) circuit. The RTD-based SRAM circuit has a perimeter defining an RTD-based SRAM cell area, the cell area including circuit components defining an SRAM memory cell. A transistor having a gate length is formed above a substrate including a first crystalline material. At least one RTD is formed proximate an opening disposed above the substrate, the opening comprising a non-crystalline sidewall. The RTD is coupled to a source or a drain of the transistor and includes crystalline material lattice-mismatched to the first crystalline material. The RTD-based SRAM cell area is at least 33% smaller than a surface area of an SRAM circuit constructed using six transistors having the same gate length as the transistor in the RTD-based SRAM cell area.


In still another aspect, the invention features a method of forming a transistor structure including a tunneling structure. The method includes the formation of an opening above a surface of a substrate comprising a first crystalline semiconductor material. The opening has a non-crystalline sidewall. A second crystalline semiconductor material is formed in the opening. The second crystalline semiconductor material is lattice-mismatched to the first crystalline semiconductor material. A majority of defects arising from lattice-mismatch between the first and second semiconductor materials terminate within the opening. A tunneling structure is formed over and in contact with at least a portion of the second crystalline semiconductor material. First, second, and third terminals of a transistor are formed proximate the tunneling structure.


In an aspect, the invention features a structure. The structure includes a substrate comprising a first crystalline semiconductor material, and a window defined in a dielectric layer disposed over the first semiconductor substrate. Disposed in the window is a second crystalline semiconductor that is lattice-mismatched to the first crystalline semiconductor material. A tunneling device is disposed over and in contact with at least a portion of the second crystalline semiconductor material.


In some embodiments, the tunneling device may be an RTD, a triple-barrier or multiple-barrier resonant tunneling diode, an RITD, an RTT, an RTFET, a DELTT, a quantum-well-base resonant tunneling transistor (QEBRTT), an RTBT or RBT, or an RHET. The first crystalline semiconductor material comprises or consists essentially of at least one of a group IV element, a II-VI compound, or a III-V compound. The second crystalline semiconductor material comprises a II-VI compound and/or a III-V compound. A phase change layer is disposed over the tunneling device, and the structure includes a memory cell. A ferroelectric material is disposed over the tunneling device, and the structure includes a memory cell. Memory circuitry, such as SRAM circuitry, DRAM circuitry, or non-volatile memory circuitry, may include the tunneling device.


In another aspect, the invention features a method for forming a structure. The method includes the formation of a masking layer over a substrate comprising a first crystalline semiconductor material. A window is defined in the masking layer. The window is filled with a second crystalline semiconductor material that is lattice-mismatched to the first crystalline semiconductor material. A tunneling device is disposed over and in contact with at least a portion of the second crystalline semiconductor material.


The tunneling device may be a DELTT, an RTD, or an RITD. The window may be filled by selective epitaxy.


In another aspect, the invention features a structure. The structure includes a substrate comprising a first crystalline semiconductor material, and an RTD comprising a second crystalline semiconductor material disposed over the substrate. The first crystalline semiconductor material includes a group IV element or compound and the second crystalline semiconductor material includes a III-V or a II-VI compound.


In another aspect, the invention features a structure. The structure includes a substrate including a first crystalline semiconductor material comprising a group IV element; An RITD is disposed over the substrate. The RITD includes a second crystalline semiconductor material lattice-mismatched to the first crystalline semiconductor material, the second crystalline semiconductor material including at least one of a III-V compound or a II-VI compound. The RITD includes a bottom injector layer, a bottom spacer, a top spacer, a top injector layer separated by an offset from the bottom injector, the bottom injector and top injector layers forming the ends of a p-i-n junction, and a material inserted between the bottom injector and top injector which serves as a tunnel barrier, and where i in the p-i-n junction represents at least one material provided between the bottom and top injectors.


In another aspect, the invention features a structure. The structure includes a substrate comprising a first crystalline semiconductor material comprising a group IV element and a DELTT device disposed over the substrate. The DELTT device includes a second crystalline semiconductor material lattice-mismatched to the first crystalline semiconductor material, the second crystalline semiconductor material including at least one of a III-V compound or a II-VI compound. The DELTT device includes an emitter contact, a collector contact, a two dimensional quantum well emitter layer electrically contacting the emitter contact, and a two dimensional quantum well resonant tunneling layer. The DELTT device also includes a collector layer disposed opposite the resonant tunneling layer from the emitter layer, the collector layer electrically contacting the collector contact, a first tunneling barrier layer disposed between the emitter layer and the resonant tunneling layer, a second tunneling barrier layer disposed between the resonant tunneling layer and the collector layer; and a voltage drop layer between the resonant tunneling layer and the collector.


In another aspect the invention features a structure. The structure includes a substrate comprising a first crystalline semiconductor material, and a magnetic resonant tunneling diode comprising a second crystalline semiconductor material disposed over the substrate. The first crystalline semiconductor material may comprises or consist essentially of a group IV element or compound and the second crystalline semiconductor material may comprise or consist essentially of a III-V or a II-VI compound.


In another aspect, the invention features a structure. The structure includes a substrate comprising a first crystalline semiconductor material. A first window is defined in a dielectric layer disposed over the first semiconductor substrate. A first device is disposed over the first window. A second window is defined in a dielectric layer disposed over the first semiconductor substrate. A second crystalline semiconductor is disposed in the second window, the second crystalline semiconductor material being lattice-mismatched to the first crystalline semiconductor material. A tunneling device is disposed over and in contact with at least a portion of the second crystalline semiconductor material. The first device and the tunneling device are interconnected to form a circuit unit.


In yet another aspect, the invention features a structure including a substrate that comprises a first crystalline semiconductor material. A dielectric layer is disposed over at least a portion of the substrate and has a window defined therein. A second crystalline semiconductor material is disposed in the window and over the first crystalline semiconductor material, and is lattice-mismatched to the first crystalline semiconductor material. A tunneling device is disposed over and in contact with at least a portion of the second crystalline semiconductor material over a first region of the substrate. A circuit component is formed over a second region of the substrate. The first crystalline semiconductor material includes a group IV element or compound, a II-VI compound, and/or a III-V compound, the second crystalline material includes a group IV element or compound, a II-VI compound and/or a III-V compound, and the first crystalline semiconductor material and the second crystalline semiconductor material include materials selected from different groups.


The following feature may be included. The circuit component may be a MOSFET, a MISFET, a HEMT, a capacitor, or a resistor.


In another aspect, a structure may include a substrate comprising a first crystalline semiconductor material. A dielectric layer is disposed over at least a portion of the substrate and has a window defined therein. A second crystalline semiconductor material is disposed in the window and over the first crystalline semiconductor material, the second crystalline semiconductor material being lattice-mismatched to the first crystalline semiconductor material. An amplifier includes (i) a tunneling diode disposed over and in contact with at least a portion of the second crystalline semiconductor material, and (ii) a resistor. The first crystalline semiconductor material includes a group IV element or compound, a II-VI compound, and/or a III-V compound, the second crystalline material includes a group IV element or compound, a II-VI compound and/or a III-V compound, and the first crystalline semiconductor material and the second crystalline semiconductor material include materials selected from different groups.


In another aspect, the invention features a structure including a substrate comprising a first crystalline semiconductor material. A dielectric layer is disposed over at least a portion of the substrate and has a window defined therein. A second crystalline semiconductor material is disposed in the window and over the first crystalline semiconductor material, the second crystalline semiconductor material being lattice-mismatched to the first crystalline semiconductor material. A memory cell includes a tunneling diode is disposed over and in contact with at least a portion of the second crystalline semiconductor material. The first crystalline semiconductor material includes a group IV element or compound, a II-VI compound, and/or a III-V compound, the second crystalline material includes a group IV element or compound, a II-VI compound and/or a III-V compound, and the first crystalline semiconductor material and the second crystalline semiconductor material include materials selected from different groups.


In another aspect, the invention features a structure including a substrate comprising a first crystalline semiconductor material. A window is defined in a dielectric layer disposed over the substrate. A second crystalline semiconductor material is disposed in the window, the second crystalline semiconductor material being lattice-mismatched to the first crystalline semiconductor material. A logic inverter includes (i) a tunneling device disposed over and in contact with at least a portion of the second crystalline semiconductor material, and (ii) a MOSFET. The first crystalline semiconductor material includes a group IV element or compound, a II-VI compound, and/or a III-V compound, the second crystalline material includes a group IV element or compound, a II-VI compound and/or a III-V compound, and the first crystalline semiconductor material and the second crystalline semiconductor material include materials selected from different groups.


In another aspect, the invention features a structure including a substrate comprising a first crystalline semiconductor material. A dielectric layer is disposed over at least a portion of the substrate and has a first window and a second window defined therein. A second crystalline semiconductor material is disposed in each of the first and second windows and over the first crystalline semiconductor material. The second crystalline semiconductor material is lattice-mismatched to the first crystalline semiconductor material. An SRAM includes (i) a first resonant tunneling diode disposed over and in contact with at least a portion of the second crystalline semiconductor material disposed in the first window, (ii) a second resonant tunneling diode disposed over and in contact with at least a portion of the second crystalline semiconductor material disposed in the second window, and (iii) a MOSFET. The first crystalline semiconductor material includes a group IV element or compound, a II-VI compound, and/or a III-V compound, the second crystalline material includes a group IV element or compound, a II-VI compound and/or a III-V compound, and the first crystalline semiconductor material and the second crystalline semiconductor material include materials selected from different groups.


One or more of the following features may be included. Each of the first and second resonant tunneling diodes may be in electrical communication with a drain of the MOSFET. Each of the first and second resonant tunneling diodes may be in electrical communication with a gate of the MOSFET.


In another aspect, the invention features a structure including a substrate comprising a first crystalline semiconductor material. A dielectric layer is disposed over at least a portion of the substrate and has plurality of windows defined therein. A second crystalline semiconductor material is disposed in each of the plurality of windows and over the first crystalline semiconductor material, the second crystalline semiconductor material being lattice-mismatched to the first crystalline semiconductor material. A field programmable array includes a plurality of SRAMs, each SRAM comprising (i) a first resonant tunneling diode disposed over and in contact with at least a portion of the second crystalline semiconductor material disposed in one of the plurality of windows, (ii) a second resonant tunneling diode disposed over and in contact with at least a portion of the second crystalline semiconductor material disposed in another of the plurality of windows, and (iii) a MOSFET. The first crystalline semiconductor material includes a group IV element or compound, a II-VI compound, and/or a III-V compound, the second crystalline material includes a group IV element or compound, a II-VI compound and/or a III-V compound, and the first crystalline semiconductor material and the second crystalline semiconductor material include materials selected from different groups.


In another aspect, the invention features a structure that includes a substrate comprising a first crystalline semiconductor material. A dielectric layer is disposed over at least a portion of the substrate and has a plurality of windows defined therein. A second crystalline semiconductor material is disposed in each of the windows and over the first crystalline semiconductor material, the second crystalline semiconductor material being lattice-mismatched to the first crystalline semiconductor material. A microprocessor includes (i) arithmetic logic units (ALU) adapted for executing arithmetic and logical functions; (ii) storage locations and memory units adapted to store data, addresses, instructions and control programs; and (iii) control systems and interconnection systems adapted to provide links among different components, accept and generate external control signals, and to provide timing signals. At least one of the ALU, storage locations and memory units, and control and interconnection systems comprises a tunneling device disposed over and in contact with the second crystalline semiconductor material. The first crystalline semiconductor material includes a group IV element or compound, a II-VI compound, and/or a III-V compound, the second crystalline material includes a group IV element or compound, a II-VI compound and/or a III-V compound, and the first crystalline semiconductor material and the second crystalline semiconductor material include materials selected from different groups.


In another aspect, the invention features a method for forming a structure. The method includes providing a substrate comprising a first crystalline semiconductor material and a dielectric layer disposed thereover. A CMOS device is formed in a first region of the substrate. A window is defined in the dielectric layer to expose at least a portion of a surface of the first semiconductor material. The window is filled with a second crystalline semiconductor material lattice-mismatched to the first crystalline semiconductor material. A tunneling device is defined over and in contact with at least a portion of the second crystalline semiconductor material. An interconnection is formed between the tunneling device and the CMOS device.


In yet another aspect, the invention features a method for forming a structure, the method including providing a substrate comprising a first crystalline semiconductor material. A masking layer is formed over the substrate. A window is defined in the masking layer. The window is filled with a second crystalline semiconductor material lattice-mismatched to the first crystalline semiconductor material, the second semiconductor material having a total thickness higher than a height of the masking layer. The second semiconductor material is planarized to define a planar surface of the second semiconductor material. A tunneling device is defined over and in contact with at least a portion of the second crystalline semiconductor material. Planarizing the second semiconductor material may include chemical-mechanical polishing utilizing the masking layer as a polishing stop.





BRIEF DESCRIPTION OF THE FIGURES


FIG. 1 is a cross-sectional view of an RTD or RITD device structure, prior art;



FIG. 2 is a perspective view of a double electron layer tunnel transistor (DELTT), prior art;



FIG. 3 illustrates a circuit of a tunnel diode SRAM and its associated IV curve, prior art;



FIG. 4 is a cross-sectional view of a GaAs/AlGaAs RTD formed on a Si substrate with the use of ART in accordance with an embodiment of the invention;



FIG. 5 is a cross-sectional view of an RTD formed with the use of ART and including III-V materials in accordance with an embodiment of the invention;



FIGS. 6
a-6b is a cross-sectional view of an RTD formed with the use of ART and including contacts thereto in accordance with an embodiment of the invention;



FIGS. 7
a-7b and 8a-8c are cross-sectional views of device structures with non-flat top surfaces formed with the use ART in accordance with embodiments of the invention;



FIG. 9 is a cross-sectional view of a triple-barrier RTD formed with the use of ART in accordance with an embodiment of the invention;



FIGS. 10
a-10e are cross-sectional views of five different RITD structures formed with the use of ART in accordance with embodiments of the invention;



FIG. 11 is a cross-sectional view of a magnetic resonant tunneling diode as a spin selector for spintronic applications in accordance with an embodiment of the invention;



FIG. 12 is a pair of cross-sectional views illustrating a process for forming an RTFET in accordance with an embodiment of the invention;



FIG. 13 is a cross-sectional view of a QWBRTT formed with the use of ART in accordance with an embodiment of the invention;



FIGS. 14
a-14c are cross-sectional views of tunneling quantum structures integrated within a bipolar transistor, formed with the use of ART in accordance with embodiments of the invention, as well as the corresponding energy-band diagrams;



FIG. 15 is a cross-sectional view of a three-terminal bipolar quantum resonant tunneling transistor formed over GaAs grown by ART on a Si substrate in accordance with an embodiment of the invention;



FIG. 16
a is a cross-sectional view of a multi-state RTBT formed by ART in accordance with an embodiment of the invention; FIG. 16b is an energy-band diagram of the device of FIG. 16a;



FIG. 17 is a cross-sectional view of an RHET formed by ART in accordance with an embodiment of the invention;



FIG. 18 is a cross-sectional view of a typical DELTT structure disposed over GaAs material formed by ART in accordance with an embodiment of the invention;



FIG. 19 is a cross-sectional view of a building block for various logic circuit units in accordance with an embodiment of the invention;



FIGS. 20
a-20c and 21 are diagrams illustrating circuit units incorporating RTD formed by ART in accordance with embodiments of the invention;



FIG. 22 is a cross-sectional view of an SRAM structure including two RTDs and an n-FET in accordance with another embodiment of the invention;



FIGS. 23
a-23e, 24a-24e, and 25 are diagrams illustrating SRAMs incorporating RTD devices formed in accordance with embodiments of the invention;



FIG. 26 is a cross-sectional view of a non-volatile memory cell including an RTD and one memory storage material (such as phase change material or ferroelectric material) in accordance with another embodiment of the invention;



FIG. 27 is a cross-sectional view of a non-volatile magnetic random access memory including an RITD and a magnetic tunneling structure in accordance with another embodiment of the invention;



FIG. 28 is a diagram of a monostable-bistable transition logic element including an RTD formed by RTD in accordance with an embodiment of the invention;



FIG. 29 is a diagram of an analog-to-digital converter circuit employing multi-state RTBT devices formed by ART in accordance with an embodiment of the invention;



FIG. 30 is a diagram of a RTD/HEMT comparator incorporating an RTD into a HEMT formed by ART in accordance with an embodiment of the invention;



FIG. 31 is a diagram of an RTD-based clock generator formed by ART in accordance with an embodiment of the invention;



FIG. 32 is a diagram of a field programmable gate array incorporating RTD-based SRAMS formed by ART in accordance with an embodiment of the invention;



FIGS. 33 and 34 are block diagrams illustrating RTD-based or RTT-based microprocessors formed by ART in accordance with embodiments of the invention;



FIG. 35 is a block diagram of an RTD-based digital signal processor formed by ART in accordance with an embodiment of the invention;



FIG. 36 is a block diagram illustrating an RTD-based graphics processing unit formed by ART in accordance with an embodiment of the invention; and



FIG. 37 is a cross-sectional view of a device including two RTDs connected to a drain of a MOSFET, formed by ART in accordance with an embodiment of the invention.





DETAILED DESCRIPTION OF THE INVENTION

Although the following examples are illustrated with particular III-V material systems for tunneling device on Si substrates, embodiments of this invention may include other materials systems, such as SiGe, Ge, III-V and/or II-VI. The described structures may be fabricated on various semiconductor substrates, such as Si substrates, Ge substrates, semiconductor-on-insulator (SOI) substrates, strained-semiconductor-on-insulator (SSOI) substrates, and other substrate systems. Embodiments of the invention also include other similar tunneling device structures and circuit units. Structures may include various tunneling devices and various circuit units; methods are described herein for producing such structures or circuit units on a first type of substrate, e.g., from group IV such as a Si or Ge substrate, with improved material systems, e.g., III-V and II-VI epitaxial materials. The III-V or II-VI epitaxial layers may be grown selectively on conventional substrates such as Si, SOI, or SSOI substrates in selected areas (such as the device active area). Such layers may be formed by a heteroepitaxy technique, e.g., by ART heteroepitaxial techniques, as described in pending U.S. patent application Ser. Nos. 11/436,198 and 11/436,062.


Cross-sectional TEM micrographs were taken of an experimental sample of lattice-mismatched semiconductor material grown within dielectric windows on a Si substrate by the ART heteroepitaxial technique. The micrographs demonstrated that defects, e.g., dislocations, are trapped by the dielectric sidewalls, resulting in high quality material suitable for device fabrication.


I. Individual Devices

Referring to FIG. 4, the use of an ART technique enables the integration of low-defect III-V or II-VI materials on a Si substrate for tunneling device 400 fabrication. An exemplary RTD device structure is illustrated in FIG. 5. After a window is opened in a dielectric layer over a substrate 405, ART is used to grow a low-defect high-quality semiconductor material 410 (e.g., GaAs) on a Si surface, such that defects 415, e.g., dislocations and stacking faults, are trapped by the dielectric side walls 420, resulting a low-defect upper region of GaAs material. Next, multiple layers 425 of GaAs and AlGaAs are grown on the low-defect GaAs material by epitaxy. These layers may be used to fabricate an RTD device. This RTD device may be contacted via first contact 430 and second contact 435. In some embodiments, the first contact 430 may be the anode terminal and the second contact 435 may be the cathode terminal of tunneling device 400, and in other embodiments, the first contact 430 may be the cathode terminal and the second contact 435 may be the anode terminal.


A detailed preferred fabrication process is as follows. The substrate 405 may be, for example, a bulk silicon wafer, a bulk germanium wafer, a semiconductor-on-insulator (SOI) substrate, or a strained semiconductor-on-insulator (SSOI) substrate. The substrate 405 may include or consist essentially of a first semiconductor material, such as a group IV element, e.g., germanium or silicon. In an embodiment, the substrate 405 includes or consists essentially of (100) silicon.


A non-crystalline material, e.g., a dielectric layer 440, is formed over the semiconductor substrate 405. The dielectric layer 440 may include a dielectric material, such as silicon nitride or silicon dioxide. The dielectric layer 440 may be formed by a method known to one of skill in the art, e.g., thermal oxidation or plasma-enhanced chemical vapor deposition. As discussed below, the dielectric layer 440 may have a thickness t1 corresponding to a desired height of crystalline material to be deposited in a window 445 formed through the dielectric layer 440. In some embodiments, the thickness t1 of the dielectric layer 440 may be selected from a range of, e.g., 20-50000 nm.


A mask (not shown), such as a photoresist mask, is formed over the substrate 405 and the dielectric layer 440. The mask is patterned to expose at least a portion of the dielectric layer 440. The exposed portion of the dielectric layer 440 is removed by, e.g., reactive ion etching (RIE) to define the window 445. The window 445 extends to a surface of the substrate 405 and may be defined by at least one sidewall 420. The sidewall 420 is formed from the dielectric layer 440 and is, therefore, non-crystalline. The sidewall 420 may have a height h at least equal to a predetermined distance H from the surface of the substrate 405. It has been observed experimentally that dislocations 415 in a mismatched cubic semiconductor grown on a Si (100) surface in the near-vicinity (e.g., within approximately 500 nm or less) of a vertical dielectric sidewall 420 surface bend toward that surface at approximately 30 degrees through 60 degrees. For example, the dislocations 415 may bend toward that surface at approximately a 45 degree angle to that surface. Based on this relationship, one may typically expect the predetermined distance H necessary to trap defects 415 to be approximately equal to a width between ½ w and 2w, where w is the width of the window 445.


The window 445 may be substantially rectangular in terms of cross-sectional profile, a top view, or both, and have a width w that is smaller than a length l (not shown) of the window. For example, the width w of the window may be less than about 5000 nm, e.g., about 20-1000 nm. In some embodiments, the width of the window is about 150 nm. A length l of the window may exceed each of w and H. A ratio of the height h of the window to the width w of the window may be ≧1, preferably between about 1 and about 50.


A second crystalline semiconductor material 410 is formed in the window. The second crystalline semiconductor material 410 may include or consist essentially of a group IV element or compound, a III-V compound, or a II-VI compound. Examples of suitable group IV elements or compounds include germanium, silicon germanium, and silicon carbide. Examples of suitable III-V compounds include gallium arsenide, gallium nitride, indium arsenide, indium antimonide, indium aluminum antimonide, indium aluminum arsenide, indium phosphide, and indium gallium arsenide. Examples of suitable II-VI compounds include zinc selenide and zinc oxide.


The second crystalline semiconductor material 410 may be formed by selective epitaxial growth in any suitable epitaxial deposition system, including, but not limited to, metal-organic chemical vapor deposition (MOCVD), atmospheric-pressure CVD (APCVD), low- (or reduced-) pressure CVD (LPCVD), ultra-high-vacuum CVD (UHCVD), molecular beam epitaxy (MBE), or atomic layer deposition (ALD). In the CVD process, selective epitaxial growth typically includes introducing a source gas into the chamber. The source gas may include at least one precursor gas and a carrier gas, such as, for example, hydrogen. The reactor chamber may be heated by, for example, RF-heating. The growth temperature in the chamber may range from about 300° C. to about 900° C., depending on the composition of the epitaxial region. The growth system may also utilize low-energy plasma to enhance the layer growth kinetics. CVD has a number of advantages, including the capability for depositing films with low defect densities and rapidly varying alloy compositions, as well as high quality regrowth capability. CVD may also provide improved manufacturability due to relatively higher throughput, relatively short downtimes, and scalability to very large reactors.


The epitaxial growth system may be a single-wafer or multiple-wafer-batch reactor. Suitable CVD systems commonly used for volume epitaxy in manufacturing applications include, for example, an Aixtron 2600 multi-wafer system available from Aixtron, based in Aachen, Germany; an EPI CENTURA single-wafer multi-chamber systems available from Applied Materials of Santa Clara, Calif.; or EPSILON single-wafer epitaxial reactors available from ASM International based in Bilthoven, The Netherlands.


Dislocation defects 415 in the second crystalline semiconductor material 410 reach and terminate at the sidewalls 420 of the window in the dielectric material 440 at or below a vertical predetermined distance H from the surface of the substrate, such that dislocations 415 in the second crystalline semiconductor material 410 decrease in density with increasing distance from the bottom portion of the window. Accordingly, the upper portion of the crystalline material 410 is substantially exhausted of dislocation defects 415. Various dislocation defects 415 such as threading dislocations, stacking faults, twin boundaries, or anti-phase boundaries may thus be substantially eliminated from the upper portion of the epitaxial region. A density of such dislocation defects 415 may be less than, for example, 106/cm2, preferably less than 103/cm2. The second crystalline semiconductor material 410 may be either substantially relaxed or strained.


A high-quality tunneling device 400 may be fabricated on top of the low-defect second crystalline semiconductor material 410. The tunneling device may be, for example, an RTD including GaAs and AlGaAs. The RTD device 400 may include multiple thin semiconductor layers 425 such as, from bottom to top, a relatively thick n-type layer 446 (e.g., n-type gallium arsenide with a thickness of, e.g., 10-25000 nm); an undoped layer 447 (e.g., undoped gallium arsenide with a thickness of, e.g., 0-50 nm); a thin barrier layer with a large bandgap 448 (e.g., undoped aluminum gallium arsenide with a thickness of, e.g., 0-50 nm); a thin two-dimensional resonant electron layer 449 (e.g., undoped gallium arsenide with a thickness of, e.g., 0-50 nm); another thin barrier layer with large bandgap 450 (e.g., undoped aluminum gallium arsenide with a thickness of, e.g., 0-50 nm); another undoped layer 451 (e.g., undoped gallium arsenide with a thickness of, e.g., 0-50 nm); and another relatively thick n-type layer 452 (e.g., n-type gallium arsenide, 10-25000 nm). All of these layers 425 may be grown epitaxially in sequence in any suitable epitaxial deposition system, including, but not limited to, MOCVD, APCVD, LPCVD, UHCVD, MBE, or ALD. This exemplary RTD device has two external contacts: a first contact 430 contacts the n-type layer 446, and a second contact 435 contacts the n-type layer 452. Both contacts 430 and 435 typically include metal contacts. The first contact 430 may be made by etching a small hole (not shown) on one side of the device to reach the layer 446. Methods and variations for the fabrication of this RTD 400 device will be apparent to one of skill in the art.


In the above examples, the layers 425 of the RTD include or consist essentially of AlGaAs and GaAs. Thus, the substrate may include a group IV element such as Si, and the RTD may include a III-V compound. In some embodiments, the RTD may include a II-VI compound.


Although only one exemplary RTD is illustrated in FIG. 4, embodiments of the invention also include other varieties of resonant tunneling device structures with a variety of semiconductor material combinations, layer structures, layer thicknesses, doping levels in each layer, epitaxy growth conditions (e.g., precursors, growth temperature, etc), and contact configurations.


For example, referring to FIG. 5, the layers 425 of RTD may include various combinations of pairs of III-V materials. The layers may include a heavily doped materials 1 layer 546, an undoped materials 1 (spacer) layer 547, an undoped materials 2 barrier layer 548, an undoped materials 1 (2D resonant electron layer) layer 549, a second undoped materials 2 barrier layer 550, an undoped material 1 spacer layer 551, and a heavily doped materials 1 layer 552, where materials 1/materials 2 combinations may be, e.g., GaAs/AlGaAs, GaAs/InGaAs, GaInAs/AlInAs, GaAs/AlAs, InP/InGaAs, InGaAs/AlAs, InAs/AlSb, or InSb/AlInSb. Other material systems may also be used. In addition, the two doped layers 546 and 552 may include different materials. The 2D resonant electron layer 549 may be undoped and may have a thickness of 5 nm, and the barrier layers 548, 550 may be undoped and may have a thickness of 1 to 5 nm. The doped layers 546 and 552 may be, e.g., n-type with a doping level of 2×1018 to 1×1019/cm3.


Referring to FIGS. 6a and 6b, a contact to a tunneling device 400 may be formed by the following approach. The steps in this approach are also applicable to other tunneling devices and structures of other embodiments of the inventions described herein. In a first step, III-V heteroepitaxial layers 410 are grown selectively in a window 445 defined by a dielectric layer 440 disposed over a semiconductor substrate 405. Such dielectric windows may be utilized in some heteroepitaxial growth technologies such as ART technology. In a second step, the dielectric 440 is etched back, e.g., approximately halfway back to form a recess structure. The etch-back of the dielectric 440 partially exposes the layers 425 within the dielectric window, allowing, e.g., a first contact 430 to also be made to a bottom layer disposed in the dielectric window (e.g., to the doped layer 546 in FIG. 6b).


In most of the drawings discussed herein, the semiconductor epitaxial layers of the tunneling device structures are depicted with flat layers, for the sake of simplicity. Nevertheless, those drawings should also be considered to encompass structures with non-flat layers. One experimental sample of Ge material grown in silicon oxide windows on a Si substrate by ART technique illustrates that the surface of a semiconductor epitaxial layer is typically non-flat, due to faceting during the epitaxial growth. Therefore, the simplified illustrations such as FIG. 5 should be interpreted as including variations having non-flat epitaxial layers, as depicted in FIG. 7a. Another approach is depicted in FIG. 7b, where some layers 425 of the device structure are grown within the dielectric windows and are non-flat, while other layers are grown above the dielectric layer 440.


Referring to FIGS. 8a-8c, another embodiment of an epitaxial layer and device structure is formed as follows. In the first step, FIG. 8a, III-V material 410 is grown within and above dielectric windows 445 by ART technique, this III-V material 410 having a non-flat surface due to faceting. In the second step, FIG. 8b, the structure is planarized to achieve a flat surface, for example by CMP using the dielectric layer 440 as a polishing stop. In the third step (FIG. 8c), epitaxial device layers 425 are grown on the flat surface. Contacts to both doped layers 546 and 552 may be made, by, e.g., first and second contacts 430 and 435, as shown in FIG. 8c, as both doped layers 546 and 552 are disposed above the dielectric layer 440 and are at least partially exposed.


Although only one exemplary device structure is illustrated in both FIGS. 7 and 8a-8c, these approaches and methods are also applicable to other tunneling devices and structures of other embodiments of the inventions described herein.


In addition to the basic RTD structures discussed above, some other exemplary tunneling devices are further described below. In some embodiments, the structures described below may be fabricated on III-V or II-VI epitaxial layers that are grown selectively on conventional substrates such as Si, SOI, or SSOI substrates in selected areas (such as the device active area). Such layers may be formed by heteroepitaxy, such as with ART techniques.


Referring to FIG. 9, a triple-barrier tunneling diode includes three barriers 548, 550, 550′ and two 2D resonant electron layers 549, 549′. This structure gives rise to multiple current peaks on I-V characteristics. A multiple-barrier tunneling diode may be formed by incorporating more than three barrier layers separating multiple 2D resonant electron layers (not shown).


Typical RTDs are often intra-band tunneling diodes. A variation of a tunneling diode is the RITD device. FIGS. 10a-10e illustrate five different RITD structures 1000. In each example, the III-V epitaxial layers 410, e.g., n+-InAs, are grown on a semiconductor substrate 405, e.g., an Si wafer, in the device area by heteroepitaxy. Referring to FIG. 10a, an RITD 1000 may include an ohmic contact 1005 and an injecting junction 1010 that is a polytype heterostructure, i.e., including a barrier layer 1020, 1020′ (e.g., i-AlSb layer having a thickness selected from a range of about 1.5 nm to 3 nm) inserted in a type-III (or so-called broken-gap) heterojunction, including e.g., a p-GaSb layer 1030 and an n-InAs layer 1040. Referring to FIG. 10b, an RITD 1000 may include a type-III (or broken-gap) heterojunction including, e.g. a p-GaSb layer 1030 and an n-InAs layer 1040.


Referring to FIG. 10c, an RITD 1000 may include two delta-doped layers δn+, and δp+1050, 1060. The delta-doped layer sheet charges are typically on the order of 1×1013/cm2. With delta-doped layers 1050, 1060, a single material may be used to make the diode, e.g., intrinsic 1061 and p+ doped 1062. Referring to FIG. 10(d), an RITD 1000 may include three delta-doped layers, e.g., δn+, δp+, and δn+1050, 1060, 1050′, between two intrinsic layers 1061 and an n+ layer 1063. Referring to FIG. 10(e), an RITD 1000 may include a double-quantum-well heterostructure with layers including, e.g., n+-InAlAs 1070, i-InGaAs 1075, i-InAlAs 1080, i-InGaAs 1085, and p+-InAlAs 1090.


One advantage of RITDs is a high peak-to-valley current ratio (PVCR). Among the illustrated devices, the RITD 1000 with double-quantum-well heterostructure (FIG. 10(e)) exhibits the highest PVCR.


Referring to FIG. 11, an example of a magnetic resonant tunneling diode 1100 as a spin selector for spintronics applications is illustrated. It is fabricated in two phases. The first phase includes heteroepitaxy growth of a buffer layer 410 of either GaAs or ZnBeSe material using ART techniques to achieve high-quality low defect material. The second phase includes fabricating a typical magnetic resonant tunneling diode 1100. The magnetic resonant tunneling diode 1100 includes layer structure 425 that may include, for example, Zn0.97Be0.03Se layer (300 nm thick, n-type doping at 8×1018/cm3) 1105, a first ZnSe layer (100 nm thick, n-type doping at 1.5×1019/cm3) 1110, Zn0.97Be0.03Se layer (10 nm thick, n-type doping at 1×1018/cm3) layer 1115, a second ZnSe (10 nm, intrinsic) layer 1120, Zn0.7Be0.03Se layer (5 nm thick, intrinsic) 1125, Zn0.96Mn0.04Se layer (9 nm thick, intrinsic) 1130, Zn0.7Be0.3Se layer (5 nm thick, intrinsic) 1135, a third ZnSe layer (10 nm thick, intrinsic) 1140, Zn0.97Be0.03Se layer (15 nm thick, n-type doping at 1×1018/cm3) 1145, and a fifth ZnSe layer (30 nm thick, n-type doping at 1.5×1019/cm3) 1150. Metal contacts 1160 are formed to the first ZnSe layer 1110 and the fifth ZnSe layer 1150. With magnetic Mn incorporated into the quantum well material in the magnetic resonant tunneling diode 1100, a strong splitting of well resonance is obtained as a function of external magnetic field. The device 400 can thus be used as a spin selector.


Referring to FIGS. 12 to 17, besides various 2-terminal tunneling diodes, a number of 3-terminal transistor devices incorporate quantum tunneling structures. The advantages of 3-terminal tunneling devices over the 2-terminal tunneling diodes may include: (i) providing a high-speed tunable negative differential resistance/transconductance, and (ii) providing isolation between the input and output.


Referring to FIGS. 12a-12b, an exemplary fabrication process for forming an RTFET on ART semiconductor material is as follows. Multiple layers 425 are grown by heteroepitaxy, e.g., by an ART process, to form a quantum tunneling device structure 400. The fabrication process and materials used may be similar to those discussed with respect to FIG. 5, and include a source 1200 and a drain 1205. The dielectric layer 440 is at least partially etched back, e.g., halfway, and then a gate structure 1210 is formed to control the device. The gate 1210 is typically either a metal contact (Schottky diode), or a p-n junction. The function of the gate is to control the depletion width to vary the net cross-section area of the diode.


Devices may have one of two types of gate structures. In a two-sided gate structure, the gate material may form on two opposite sides of the diode, resulting in a 1-D resonant-tunneling device, or quantum wire device. In a four-sided gate structure, the gate material may form around all sides of the diode, resulting in a 0-D resonant-tunneling device, or quantum dot (or single-electron) transistor.


Referring to FIGS. 13a-13b, a QWBRTT is an ultrafast device in which carrier conduction from an emitter 1300 to a collector 1305 takes place via a tunneling structure. It is essentially an RTD device with a third terminal (a base) 1310 that contacts the quantum well to control its potential. Two types of QWBRTT include a bipolar device with p-type base that has a doping type (e.g., p-type) opposite to that of the emitter 1300 and the collector 1305 (e.g., n-type) (FIG. 13a), and a unipolar device with an n-type base that has a doping type (e.g., n-type) that is the same as the emitter 1300 and collector 1305 (e.g., n-type) (FIG. 13b). Terminals for an emitter contact 1320 and a base contact 1325 are provided.


Referring to FIGS. 14a-14c, an RTBT or RBT integrates a tunneling quantum structure within a bipolar transistor. Three exemplary structures vary by the location of where the quantum tunneling structure is incorporated.


Referring to FIG. 14a, an RTBT structure may include a tunneling quantum structure in the base 1310 region. Its energy band diagram is shown. As illustrated, a GaAs material system may be used; another exemplary material system is GaInAs/AsInAs. Referring to FIG. 14b, an RTBT structure may include a tunneling quantum structure at the base-emitter junction. Referring to FIG. 14c, an RTBT structure may include a tunneling quantum structure in the emitter 1300 region.


Referring to FIG. 15, a three-terminal bipolar quantum RTT may be disposed over lattice-mismatched material 410, e.g., GaAs, formed by ART growth on a substrate 405. This bipolar quantum resonant tunneling transistor operates in a manner similar to a conventional bipolar transistor; the tunneling current may be controlled by modulating the potential inside a quantum well that functions as a base. The fabrication process has two phases. In the first phase, a second semiconductor material 410 (e.g., GaAs) is grown in a window 445 on a Si substrate 405 using ART. In the second phase, multiple layers 425 of GaAs and AlGaAs quantum wells are grown epitaxially on this structure. These QWs may be formed by typical processes familiar to those of skill in the art, and may include, e.g., the following layers: GaAs buffer 410 formed on Si substrate 405 by ART, undoped GaAs 1510 (1 μm thick); contact layer 1515 of GaAs (2 μm thick, Si-doped to 2×1018/cm3); supperlattice collector 1520 (50 periods of 8 nm GaAs and 2 nm of AlGaAs Si-doped to 2×108/cm3), 3 periods of the same superlattice 1525 (8 nm GaAs and 2 nm AlGaAs, undoped), tunnel barrier 1530 (5 nm undoped AlGaAs), quantum well 1535 (15 nm with center 5 nm Be-doped to 1×1019/cm3), tunnel barrier 1540 (5 nm undoped AlGaAs), 3 periods of the same superlattice 1545 (8 nm GaAs and 2 nm AlGaAs, undoped), supperlattice collector 1550 (50 periods of 8 nm GaAs and 2 nm of AlGaAs Si-doped to 2×1011/cm3), and contact layer 1555 (2 μm GaAs, Si-doped to 2×1018/cm3).


Referring to FIG. 16a, a multi-state RTBT 1600 may include a stack of two Al0.48In0.52As (5 nm)/Ga0.47In0.53As (5 nm)/Al0.48In0.52As (5 nm) resonant tunneling double barrier structures incorporated into the emitter 1300 region of a Ga0.47In0.53As n-p-n transistor. Its corresponding energy band diagram is shown in FIG. 16b. This multi-state RTBT exhibits multiple negative differential resistance and transconductance characteristics.


The emitter 1300 of RTBT 1600 includes n+ layer 1610 (GaInAs, 5000 angstroms), n-type layer 1615 (GaInAs, 3000 angstroms), p+ layer 1620 (GaInAs, 3000 angstroms), n-type layer 1625 (GaInAs, 500 angstroms), undoped layer 1630 (AlInAs, 50 angstroms), undoped layer 1635 (GaInAs, 50 angstroms), undoped layer 1640 (AlInAs, 50 angstroms), n+ layer 1645 (GaInAs, 1000 angstroms), undoped layer 1650 (AlInAs, 50 angstroms), undoped layer 1655 (GaInAs, 50 angstroms), undoped layer 1660 (AlInAs, 50 angstroms), and n+ layer 1665 (GAInAs, 5000 angstroms). Undoped layers 1660, 1655 and 1650 form one tunneling structure, and undoped layers 1640, 1635 and 1630 form a second tunneling structure.


Referring to FIG. 17, an RHET 1700 incorporates a tunneling diode structure between the emitter 1300 and the base 1310 regions. This figure shows an exemplary RHET on a heteroepitaxial layer 410 grown on Si, and its energy-band diagram. One distinct characteristic is that the RHET has heavily doped emitter 1300 and base 1310 regions, except for the AlGaAs/GaAs/AlGaAs tunneling diode region layers in the lower portion of emitter 1300. It combines resonant tunneling with hot-electron ballistic transport in the base, resulting in super high speed. More specifically, the RHET includes a GaAs buffer layer 410 formed by heteroepitaxy, a collector 1305 including n+-GaAs 1710 and i-AlGaAs 1720, a base 1310 including n+-GaAs 1725, and an emitter 1300 including i-AlGaAs 1730, i-GaAs 1735, i-AlGaAs 1740, and n+-GaAs 1745.


Referring to FIG. 18 a DELTT structure 1800 may be disposed over lattice-mismatched material, e.g., GaAs region 410. The GaAs region 410 is formed by ART growth on a substrate 405. An AlGaAs region 1850 is formed above the GaAs region 410. In this example, GaAs is used as narrow bandgap material and AlGaAs is used as wide bandgap material. Other narrow/wide bandgap material systems may be also used to fabricate DELTT devices in a similar manner.


In an exemplary fabrication process, a dielectric layer over a semiconductor substrate is patterned to define a window for the DELTT region. Other devices, such as CMOS devices, may be fabricated on another region of the substrate. In the DELTT region, undoped GaAs is epitaxially grown by ART techniques, such that a top portion of the GaAs is substantially defect-free. Next a DELTT device is fabricated on top of the GaAs by, for example, a typical fabrication process such as that described in U.S. Pat. No. 5,825,049, FIGS. 6A-8D. The DELTT device may include the following features: a metal emitter contact 1842, a metal collector contact 1854, a two dimensional quantum well emitter layer 1844 electrically contacting the emitter contact 1842, a two dimensional quantum well resonant tunneling layer 1846 (e.g., about 7 nm of GaAs), a collector layer 1848 disposed opposite the resonant tunneling layer 1846 from the emitter layer 1844 (e.g., a 15 nm thick, 2-dimensional emitter layer for electrons), the collector layer 1848 electrically contacting the collector contact 1854, a first barrier layer 1845 (e.g., including two layers including a 20-30 nm GaAs voltage drop layer over the collector layer 1848 and a 5 nm AlGaAs wide bandgap barrier layer under the resonant tunneling layer 1846) disposed between the collector layer 1848 and the resonant tunneling layer 1846, a second thin tunneling barrier layer 1843 (e.g., a 7.5 nm AlGaAs layer) disposed between the resonant tunneling layer 1846 and the emitter layer 1844. The device may also include a back gate 1852 formed by implantation, an insulator 1860 (60 nm), an opening 1856 for insulation, and a metal control gate 1858. At least some of these features may also include GaAs. Thus, the substrate may include a group IV element such as Si, and the DELTT device may include a III-V compound. In some embodiments, the DELTT device may include a II-VI compound.


The various transistor structures illustrated in FIGS. 13a-18 show three terminal devices (i.e., a transistor with terminals for the base, emitter, and collector) that include a tunneling structure and in which ART techniques are used to trap defects.


II. Integrated Circuit Units with Tunneling Devices and Conventional Devices

Functional circuit units may incorporate both quantum tunneling devices, such as those described above, and conventional Si-based devices such as a MOSFET, MESFET, or conventional III-V-based device such as a HEMT, etc., to form a hybrid system. Such circuit units may function as, e.g., a logic circuit, an amplifier, an SRAM or a DRAM, a microprocessor, etc.


A generic integrated circuit building block is disclosed in FIG. 19, a structure where tunneling devices are a building component for a variety of logic circuitry units. The tunneling device and/or the conventional device may be formed over desirable III-V or II-VI semiconductor material structures grown on common semiconductor substrates by ART techniques.


Referring to FIG. 19, a structure including tunneling devices may be used as a building block for a variety of logic circuitry units. An exemplary generic logic circuit building block includes at least one tunneling device(s) 400 (e.g., RTD, RTT, RITD, RTFET, QEBRTT, RTBT, RHET, DELTT, etc.) and at least one conventional device(s) 1900 (e.g., MOSFET, MISFET, HEMT, etc.). Either or both of the tunneling device 400 and the conventional device 1900 may be formed over semiconductor material 410 grown on Si substrate 405 by ART techniques. The tunneling devices 400 may be fabricated in accordance with the fabrication process described above, or other exemplary devices discussed above. The conventional device may be fabricated in accordance with typical processes. An exemplary fabrication process may include fabricating a quantum tunneling device structure after the typical front-end process for a conventional Si-based CMOS process, but before the typical back-end process of the conventional Si-based CMOS process. Other similar processes may be also used for device and circuit fabrication.


Such monolithic integration of conventional devices with high-speed tunneling devices enable the formation of high-speed logic circuits by increasing output drive current levels, logic fan-out, and output-to-input isolation, etc, while reducing both the power consumption and the number of devices used in the circuit. This hybrid circuit unit, fabricated on Si or other common substrates using ART techniques, may be used as a building block for various functional logic circuitry for various applications, such as digital logic, mixed signal, analog, etc. Several examples are illustrated below. Embodiments of the invention are not limited to those exemplary circuits, and may include other types of circuit units or circuit systems.


Example 1

Referring to FIGS. 20a-20c, basic circuit units may incorporate an RTD fabricated on III-V epitaxy layers grown on Si. Alternatively, other types of tunneling diodes may also be used, such an RITD. Referring to FIG. 20a, an exemplary circuit unit 2000 includes an amplifier that provides an AC voltage gain. Referring to FIG. 20b, another exemplary circuit unit 2010 includes a simple memory cell comprising one RTD. The negative differential resistance of an RTD gives rise to two bi-stable operating states that enable a memory function. Referring to FIG. 20c, yet another exemplary circuit unit 2020 includes a logic inverter.


Example 2

Referring to FIG. 21, two RTDs 400 and one MOSFET 1900 may be integrated to form an SRAM (TSRAM) 2100 with heteroepitaxial layers on Si. This circuit unit SRAM 2100 may be constructed with the structure illustrated in FIG. 22, where two RTD (or RITD) devices are integrated on a semiconductor substrate 405 with a conventional n-MOSFET 1900 to construct an SRAM memory cell unit. The RTD or RITD devices may be fabricated in a process flow using ART techniques. The two RTD devices may be connected in series by an underlying doped region 2205, e.g., a p+ Si region defined by implantation. Both RTD devices may be disposed over an n-well 2210 that is electrically isolated from an adjacent p-well 2215 underlying the n-FET. Interconnections (not shown) may be defined between the n-FET and the two RTD devices to form SRAM memory cell unit. The resulting structure illustrated in FIG. 22 show two tunneling structures (in the two RTDs) formed via ART and coupled to a MOSFET 1900, which as illustrated in FIGS. 23, 24 and 25 can be coupled to a terminal (gate, source, or drain) of the transistor.


Such a SRAM memory cell has two stable operation points that are substantially the same as those illustrated in FIG. 3. One may arrange multiple SRAM memory cells together within one chip to form a large scale SRAM array.


Because fewer devices are used in the SRAM memory cell unit (see FIG. 21) in comparison to conventional six-transistor SRAM units, and also due to the small footprint of RTD devices that can be made much smaller than a MOSFET, this novel hybrid SRAM unit 2100 may provide a 3 to 5 times area reduction compared to the conventional six-transistor SRAM. It is refresh-free and has fast read and write. Its low-voltage operation and low leakage current enable it to serve as a low power solution. It may also be used to replace the conventional DRAM function.


In summary, this exemplary RTD-based SRAM has the following features:

    • about 3-5× area reduction in comparison to a conventional 6-transistor SRAM (the size of RTD can be much smaller than that of a conventional transistor);
    • refresh free;
    • fast read and write; and
    • low voltage operation.


In the example of FIG. 21, a load 2110 of the memory cell is another RTD device. Alternatively, a resistor or a conventional MOSFET can serve as a load as well. Various exemplary alternative binary T-SRAM configurations 2300 are illustrated in FIGS. 23a-23e.


In the examples of FIG. 21 and FIGS. 23a-23e, the SRAM logic is binary logic. RTDs can also be used to construct multi-valued SRAM unit as well. Referring to FIGS. 24a-24e, various exemplary alternative multi-valued T-SRAM cell configurations 2400 include two or more RTDs connected in series to realize multi-value logic.


In comparison to conventional SRAM circuits that typically use six transistors, when forming transistors of a given gate length, SRAM circuits according to FIGS. 21-24, i.e., incorporating RTD(s) with transistors may be implemented in at least about a 33% smaller cell area and typically in a cell area at least 50% smaller than the area required for conventional SRAM circuits. In other words, the RTD-based SRAM cell area is at least 33% smaller than a surface area of an SRAM circuit constructed using six transistors having the same gate length as the transistor in the RTD-based SRAM cell area.


Referring to FIG. 25, as a variation of the memory cell of FIG. 21, two RTDs 400 are connected to the gate 2500 of the MOSFET 1900, instead of connected to the drain as in the (case of FIG. 21. This circuit also functions as a memory cell.


Referring to FIG. 26, a memory cell unit may include a tunneling device 400 with a memory storage film 2600 disposed thereover. The memory storage film may be a phase change material film, i.e., a phase change layer. In an alternative embodiment, the memory storage film may include a ferroelectric material film.


Referring to FIG. 27, a non-volatile RITD/magnetic random access memory (RITD/MRAM) may include (1) an RITD device 400 fabricated in a window in a dielectric layer (not shown), and (2) a tunneling structure or magnetic tunnel junction (MTJ) 2700 disposed directly on top of the RITD. MTJ 2700 has, from bottom to top, an anti-ferromagnetic exchange bias layer 2710, a magnetic pinned layer 2715, a Ru spacer layer 2720, another magnetic pinned layer 2725, a tunnel barrier layer 2730, and a magnetic free layer 2735. The MTJ 2700 structure may include a conventional MRAM memory cell structure, and may have different variations By stacking more than one MTJ in series in this structure, the memory cell can have multi-value memory function.


Another memory example includes a non-volatile memory cell including an RITD and a phase change chalcogenide resistor memory storage unit. By connecting a chalcogenide resistor as the load for the RITD, this memory cell may be used to realize a smaller multi-valued RAM memory. A typical chalcogenide resistor material is GeSbTe.


In the manner described above, various memory circuitry, such as SRAM circuitry, DRAM circuitry, and non-volatile memory circuitry, may be fabricated incorporating RTDs.


Example 3

Referring to FIG. 28, another basic circuit unit, a monostable-bistable transition logic element (MOBILE) 2800, has an edge-trigger and latching characteristics and can be used to build circuit units. For example, it can be used as an inverter. This circuit unit uses either two three-terminal resonant tunneling transistors 400, 400′ or one two-terminal RTD and one three-terminal resonant tunneling transistor 400, 400′. For example, a first RTD 400 may have a controlled gate, such as a RTFET, an RTBT/RBT or a RHET. The second RTD 400′ may or may not have a control gate. For detailed structure of three-terminal resonant tunneling transistor, refer to, for example, structures in I (Individual Devices).


Example 4

Referring to FIG. 29, an exemplary analog-to-digital converter circuit 2900 with an analog input 2910 and a binary output 2920 employs multi-state RTBT devices 1600, rather than standard RTBT devices, with an analog input 2910. For more details regarding the structure of three-terminal multi-state RTBT devices, refer to FIG. 16.


Example 5

Referring to FIG. 30, an exemplary RTD/HEMT comparator 3000 incorporates an RTD 400 into a HEMT 1900. This comparator can significantly reduce the number of devices used in a circuit, reduce power consumption, and increase speed, compared to a conventional HEMT-based comparator.


Example 6

Referring to FIG. 31, an exemplary RTD-based clock generator 3100 allows a great reduction in the number of devices used in the circuit by employing RTDs 400.


Example 7

Referring to FIG. 32, an exemplary field programmable gate array (FPGA) 3200 incorporates a large number of RTD-based SRAMs fabricated on III-V epitaxial layer structures on Si, e.g., by using ART techniques.


In a typical SRAM-based FPGA system, SRAM cells throughout the FPGA determine the functionality of the device. It typically includes:


(a) configurable logic block (CLB) 3210. The fixed arrays of CLBs are connectable by a system of pass transistors, driven by SRAM cells. Each CLB typically has two lookup tables (LUT) and two registers;


(b) switch matrix 3220; and


(c) input/output block 3230.


SRAM cells are primarily used for three purposes:

    • As look-up tables (LUTs) for implementing logic (i.e., as a truth-table).
    • As embedded block RAM blocks (for buffer storage, etc.).
    • As control to routing and configuration switches.


For the LUT, in addition to their use as general logic “gates,” LUTs may alternatively be used as general purpose RAM. Each combination of four LUTs can become a 16×1-bit RAM array.


Example 8

Referring to FIGS. 33 and 34, an exemplary RTD-based or RTT-based microprocessor 3300 is a VLSI device that carries out a range of instructions on data transferred from a register array. Similarly to the previous examples of RTD-based circuitry, the use of RTD or RTT devices significantly reduces the number of devices needed in a circuit, reduces power consumption, and increases speed, in comparison to conventional FET-based microprocessors. The microprocessor 3300 comprises (1) arithmetic logic units (ALUs) that execute a variety of arithmetic and logical function; (2) storage locations (called registers) and memory units which are used to store data, addresses, instructions and control programs, etc; and (3) control systems and interconnection systems that provide links among different components, accept and generate external control signals (such as read and write), and provide timing signals for the entire system, etc.


Using tunneling devices such as RTDs or RTTs for all of three of these components of the microprocessor can provide performance benefits including increased speed and reduced power consumption, thereby improving the basic functionalities of a microprocessor, which include:

    • retrieving data from memory;
    • manipulating data; and
    • storing data in memory.

      FIG. 33 shows an exemplary block diagram of a modern microprocessor 3300, and FIG. 34 shows its micro-architecture.


Example 9

Referring to FIG. 35, an RTD-based digital signal processor (DSP) 3500 is a specialized microprocessor designed specifically for processing signal in digital form, generally in real-time. DSPs have at least four major subfields: audio signal processing, control engineering, digital image processing, and speech processing. The baseline DSP includes one arithmetic logic unit (ALU), dual memory interfaces, and a control unit (instruction decoder, branch control, task control). An analog signal 3510 enters the analog-to-digital converter (ADC) 3520. The analog signal is converted to a digital signal 3530 and is processed by a signal processor 3540. The digital signal is transmitted to a digital-analog converter 3550, and an analog signal 3560 is transmitted to an audio amplifier 3570.


A DSP typically has many advantages over its counterpart analog processing system. It is able to provide far better levels of signal processing than is possible with analog hardware alone. It is able to perform mathematical operations that enable many of the spurious effects of the analog components to be overcome. In addition to this, it is possible to easily update a digital signal processor by downloading new software.


Example 10

Referring to FIG. 36, an RTD-based graphics processing unit (GPU) 3600 (also called a visual processing unit (VPU)) is a dedicated graphics rendering device for a personal computer or game console that is very efficient at manipulating and displaying computer graphics. Modern GPUs also have support for 3D computer graphics, and typically also include digital video-related functions. GPU 3600 components include an EDRAM 3605, unified memory 3610, texture/vertex fetch 3615, unified shader 3620, primitive setup clipper rasterizer hierarchical Z/S 3625, index stream generator tessellator 3630, and an output buffer 3635, with memory export from the output buffer 3635 to the unified memory 3610. The vertex pipeline 3640, pixel pipeline 3645, and display pixels 3650 are illustrated


III. Structures and Processes for Integrating Tunneling Devices and Conventional Devices into Circuit Units on Si or Common Substrates

In various integrated circuit units described in II (Integrated Circuit Units with Tunneling Devices and Conventional Devices), the tunneling devices may be integrated with conventional devices by several different structures and processes, using ART techniques in accordance to one aspect of this invention.


In one embodiment, for example referring to FIG. 22, two RTD (or RITD) devices 400 may be integrated on a semiconductor substrate 405 with a conventional n-FET 1900 to construct an SRAM memory cell unit 2100. The two RTD devices 400 may be connected in series by an underlying doped region 2205, e.g., a p+ Si region defined by implantation. Both RTD devices 400 may be disposed over an n-well 2210 that is electrically isolated from an adjacent p-well 2215 underlying the n-FET 1900. Interconnections (not shown) may be defined between the n-FET 1900 and the two RTD devices 400 to form SRAM memory cell unit 2100.


A typical fabrication process for forming the SRAM memory cell unit 2100 may be summarized as follows. Substrate 405, including a first semiconductor material, is patterned to define an n-FET region and RTD (or RITD) regions. The n-FET device 1900 is fabricated first using the conventional front-end Si CMOS processes. Then, windows in a dielectric 440 layer are defined in RTD/RITD regions. The windows are filled with a second crystalline semiconductor material 410, e.g., a III-V or a II-VI compound, employing ART heteroepitaxy. RTD device layers 425 are formed over the second crystalline semiconductor material 410. Finally, back-end CMOS processing is used to define interconnections between the RTD devices 400 and the n-FET 1900. Issues regarding the thermal budget are mitigated by performing the CMOS front-end processing before the formation of the RTD devices. This sequence is preferable because front-end n-FET fabrication involves high process temperatures, while RTD material may only sustain a low thermal budget. This process is also highly compatible with current Si process.


Referring still to FIG. 22, details of an exemplary fabrication process are now given. The fabrication process may be divided into three phases. n-FETs 1900 are fabricated during the first phase, which essentially is similar to a conventional front-end process of Si CMOS processing; RTDs or RITDs 400 are fabricated in the second phase; and interconnections are made during the third phase, which essentially is similar to the conventional back-end Si CMOS process. As used herein, the terms “front-end processing” and “back-end processing” have the meanings generally understood for these terms. Front-end processing generally refers to all the steps up to and including the formation of a gate and source/drain regions. Back-end generally refers to all the steps performed thereafter to fabricate a CMOS device, i.e., formation of inter-level dielectrics and conductive interconnects.


During the first phase, the substrate is cleaned using a standard RCA clean. A pad oxide of 50 nm is grown on the surface, followed by deposition of a 150 nm layer of silicon nitride. Then windows are etched through the silicon nitride and oxide layers to expose the Si surface in the area of n-well 2210 on which the RTD devices will be defined in third phase. These areas are implanted to form n-well 2210 by, for example, phosphorus implantation to a dose of 5.5E12/cm2 at an implantation energy of 100 keV.


Next, a p-well 2215 is similarly formed in the area where the n-FET 1900 will be fabricated. To form a p-well, boron implantation may be used, for example with a dose of 6E12/cm2 at 55 keV.


After both n- and p-wells are implanted, a well drive-in anneal is carried out, for example at 1100 degrees C. in a nitrogen ambient. The remaining oxide is etched away.


Next, another pad oxide of 50 nm and another silicon nitride layer 150 nm are deposited over the substrate. A window is etched through the pad oxide and silicon nitride layers to expose the underlying silicon surface only in the n-FET area. In this window, an n-FET 1900 is fabricated. Fabrication of the n-FET includes forming a silicon channel region with appropriate channel implantation, source/drain regions with appropriate source/drain implantation, a gate dielectric layer, and a gate electrode layer with appropriate gate implantation. At least one high temperature annual is conducted to activate the dopant, for example at 1000° C. for 30 minutes.


In the second phase in which the RTD devices 400 are fabricated, and in the third phase in which interconnections are defined, the processes typically have a much lower thermal budget than the first phase.


In the second phase, in an embodiment, the RTD devices 400 are grown after a FET 1900 is defined and after a pre-metal dielectric (PMD) layer is formed, but before a first metal layer is defined. Windows are etched through the dielectric layers on the areas in which RTD devices will be formed to expose the underlying silicon surface. In these windows, a second crystalline semiconductor (e.g., GaAs) 410 and then RTD devices 400 are grown epitaxially using essentially the methods described above. A relatively low temperature anneal may be carried out, for example 700 degrees C. for 1 minute.


In phase three, a back-end process similar to conventional CMOS processing is performed. First, contact windows are etched through the dielectric layer to reach the contact layers of the RTD devices. Another set of contact windows are also made to reach the source/drain/gate contacts of the n-FET. A metal layer, such as Al or Cu, is deposited over the whole surface and into the contact windows. The metal layer is patterned and selectively removed by conventional lithographic and etch processes, leaving behind a pre-determined metal interconnection network that connects the n-FET 1900 and RTD devices 400 into a functional circuit system.


In another embodiment, referring to FIG. 37, two RTD (or RITD) 400 devices may be integrated on a semiconductor substrate 405 with a conventional n-FET 1900 to construct an SRAM memory cell unit 2100. In this structure, the two RTDs 400 are directly connected with the drain region 3700 of the conventional MOSFET 1900. Alternatively, one or both of the RTDs 400 may be coupled to a source 3720 region of the MOSFET 1900. The other process steps may be substantially similar to that in FIG. 22, i.e., a MOSFET 1900 fabricated in the first phase; an RTD 400 fabricated in the second phase; and metal interconnection in the last phase. The resulting structure illustrated in FIG. 37 includes two tunneling structures (in RTDs 400) formed via ART above the drain region 3700 of MOSFET 1900, with the MOSFET bound by a shallow trench isolation region 3730.


In the configuration illustrated in FIG. 37, the size of the RTD 400 is similar to the size of traditional contact holes 3710 for the source and drain. In a typical MOSFET layout, which may have multiple contact holes 3710 formed through a PMD layer 3705 over the source 3720 and/or drain 3700 areas, incorporating the RTDs does not require an additional significant increase in source/drain area.


In yet another embodiment, the tunneling devices 400 and other devices may be fabricated on top of each other, for example, as in the structures in FIGS. 26 and 27. In those structures, the RTDs are fabricated first, before the fabrication of other device structures on the top.


The invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The foregoing embodiments are therefore to be considered in all respects illustrative rather than limiting on the invention described herein. Scope of the invention is thus indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are intended to be embraced therein.

Claims
  • 1. A method for forming a structure, the method comprising: forming an opening above a region of a substrate comprising a first crystalline semiconductor material, the opening being defined by a dielectric sidewall and an exposed surface of the substrate, the opening having a width, the dielectric sidewall having a height equal or greater than a predetermined height, the predetermined height being between half of the width and twice the width;forming in the opening a second crystalline semiconductor material lattice-mismatched to the first crystalline semiconductor material, the second crystalline semiconductor material having a thickness at least the predetermined height; anddefining a tunneling device over at least a portion of the second crystalline semiconductor material.
  • 2. The method of claim 1, wherein the tunneling device is selected from the group consisting of an Esaki diode (tunnel diode), a single-barrier tunnel diode, resonant tunneling diode (RTD), a triple-barrier or multiple-barrier resonant tunneling diode, a resonant interband tunneling diode (RITD), a single-barrier interband-tunneling, diode, a resonant tunneling transistor (RTT), a resonant tunneling field-effect transistor (RTFET), a double electron layer tunnel transistor (DELTT), a quantum-well-based resonant tunneling transistor (QWBRTT), a resonant tunneling bipolar transistor (RTBT or RBT), and a resonant tunneling hot-electron transistor (RHET).
  • 3. The method of claim 1, wherein the second crystalline material is formed in the opening by selective epitaxy.
  • 4. The method of claim 1, wherein the opening is formed above a region of a transistor.
  • 5. The method of claim 1, wherein the second crystalline semiconductor material comprises at least one of a II-VI compound or a III-V compound.
  • 6. The method of claim 1, wherein a majority of defects arising from the lattice-mismatch of the first and second crystalline semiconductor materials terminate below the height of the opening.
  • 7. The method of claim 1, wherein the width of the opening is less than the height of the opening.
  • 8. A method of forming a transistor structure, the method comprising: forming an opening above a surface of a substrate comprising a first crystalline semiconductor material, the opening having a non-crystalline sidewall;forming a second crystalline semiconductor material in the opening, the second crystalline semiconductor material being lattice-mismatched to the first crystalline semiconductor material, a majority of defects arising from lattice-mismatch between the first and second semiconductor materials terminating within the opening;forming a tunneling structure over and in contact with at least a portion of the second crystalline semiconductor material; andforming first, second, and third terminals of a transistor proximate the tunneling structure.
  • 9. The method of claim 8, wherein forming the first, the second, and the third terminals comprises removing a portion of the non-crystalline sidewall.
  • 10. The method of claim 8, wherein forming the tunneling structure comprises epitaxially growing multiple layers of semiconductor material that form quantum wells.
  • 11. The method of claim 8, wherein forming the tunneling structure comprises forming a superlattice.
  • 12. The method of claim 8, wherein the second crystalline semiconductor material comprises at least one of a II-VI compound or a III-V compound.
  • 13. The method of claim 8, wherein the opening having the non-crystalline sidewall is formed in a dielectric layer.
RELATED APPLICATIONS

This application is a divisional of U.S. patent application Ser. No. 11/862,850, filed on Sep. 27, 2007, entitled “Quantum Tunneling Devices and Circuits with Lattice-Mismatched Semiconductor Structures,” which claims the benefit of and priority to U.S. Provisional Application Ser. No. 60/848,037 filed Sep. 27, 2006, and U.S. Provisional Application Ser. No. 60/923,838 filed Apr. 17, 2007; the disclosures of these applications are hereby incorporated by reference in their entireties.

US Referenced Citations (431)
Number Name Date Kind
4307510 Sawyer et al. Dec 1981 A
4322253 Pankove et al. Mar 1982 A
4370510 Stirn Jan 1983 A
4545109 Reichert Oct 1985 A
4551394 Betsch et al. Nov 1985 A
4651179 Reichert Mar 1987 A
4727047 Bozler et al. Feb 1988 A
4774205 Choi et al. Sep 1988 A
4789643 Kajikawa Dec 1988 A
4826784 Salerno et al. May 1989 A
4860081 Cogan Aug 1989 A
4876210 Barnett et al. Oct 1989 A
4948456 Schubert Aug 1990 A
4963508 Umeno et al. Oct 1990 A
5032893 Fitzgerald, Jr. et al. Jul 1991 A
5034337 Mosher et al. Jul 1991 A
5061644 Yue et al. Oct 1991 A
5079616 Yacobi et al. Jan 1992 A
5091333 Fan et al. Feb 1992 A
5091767 Bean et al. Feb 1992 A
5093699 Weichold et al. Mar 1992 A
5098850 Nishida et al. Mar 1992 A
5105247 Cavanaugh Apr 1992 A
5108947 Demeester et al. Apr 1992 A
5156995 Fitzgerald, Jr. et al. Oct 1992 A
5159413 Calviello et al. Oct 1992 A
5164359 Calviello et al. Nov 1992 A
5166767 Kapoor et al. Nov 1992 A
5223043 Olson et al. Jun 1993 A
5236546 Mizutani Aug 1993 A
5238869 Shichijo et al. Aug 1993 A
5256594 Wu et al. Oct 1993 A
5269852 Nishida Dec 1993 A
5269876 Mizutani Dec 1993 A
5272105 Yacobi et al. Dec 1993 A
5281283 Tokunaga et al. Jan 1994 A
5285086 Fitzgerald, Jr. Feb 1994 A
5295150 Vangieson et al. Mar 1994 A
5356831 Calviello et al. Oct 1994 A
5403751 Nishida et al. Apr 1995 A
5405453 Ho et al. Apr 1995 A
5407491 Freundlich et al. Apr 1995 A
5410167 Saito Apr 1995 A
5417180 Nakamura May 1995 A
5427976 Koh et al. Jun 1995 A
5432120 Meister et al. Jul 1995 A
5438018 Mori et al. Aug 1995 A
5461243 Ek et al. Oct 1995 A
5518953 Takasu May 1996 A
5528209 MacDonald et al. Jun 1996 A
5545586 Koh Aug 1996 A
5548129 Kubena Aug 1996 A
5589696 Baba Dec 1996 A
5621227 Joshi Apr 1997 A
5622891 Saito Apr 1997 A
5640022 Inai Jun 1997 A
5710436 Tanamoto et al. Jan 1998 A
5717709 Sasaki et al. Feb 1998 A
5792679 Nakato Aug 1998 A
5825049 Simmons et al. Oct 1998 A
5825240 Geis et al. Oct 1998 A
5849077 Kenney Dec 1998 A
5853497 Lillington et al. Dec 1998 A
5869845 Vander Wagt et al. Feb 1999 A
5883549 De Los Santos Mar 1999 A
5886385 Arisumi et al. Mar 1999 A
5903170 Kulkarni et al. May 1999 A
5953361 Borchert et al. Sep 1999 A
5959308 Shichijo et al. Sep 1999 A
5966620 Sakaguchi et al. Oct 1999 A
5998781 Vawter et al. Dec 1999 A
6011271 Sakuma et al. Jan 2000 A
6015979 Sugiura et al. Jan 2000 A
6049098 Sato Apr 2000 A
6083598 Ohkubo et al. Jul 2000 A
6100106 Yamaguchi et al. Aug 2000 A
6110813 Ota et al. Aug 2000 A
6111288 Watanabe et al. Aug 2000 A
6121542 Shiotsuka et al. Sep 2000 A
6150242 Van de Wagt et al. Nov 2000 A
6153010 Kiyoku et al. Nov 2000 A
6191432 Sugiyama et al. Feb 2001 B1
6225650 Tadatomo et al. May 2001 B1
6228691 Doyle May 2001 B1
6229153 Botez et al. May 2001 B1
6235547 Sakuma et al. May 2001 B1
6252261 Usui et al. Jun 2001 B1
6252287 Kurtz et al. Jun 2001 B1
6271551 Schmitz et al. Aug 2001 B1
6274889 Ota et al. Aug 2001 B1
6300650 Sato Oct 2001 B1
6320220 Watanabe et al. Nov 2001 B1
6325850 Beaumont et al. Dec 2001 B1
6339232 Takagi Jan 2002 B1
6342404 Shibata et al. Jan 2002 B1
6348096 Sunakawa et al. Feb 2002 B1
6352942 Luan et al. Mar 2002 B1
6362071 Nguyen et al. Mar 2002 B1
6380051 Yuasa et al. Apr 2002 B1
6380590 Yu Apr 2002 B1
6403451 Linthicum et al. Jun 2002 B1
6407425 Babcock et al. Jun 2002 B1
6456214 van de Wagt Sep 2002 B1
6458614 Nanishi et al. Oct 2002 B1
6475869 Yu Nov 2002 B1
6492216 Yeo et al. Dec 2002 B1
6500257 Wang et al. Dec 2002 B1
6503610 Hiramatsu et al. Jan 2003 B2
6512252 Takagi et al. Jan 2003 B1
6521514 Gehrke et al. Feb 2003 B1
6552259 Hosomi et al. Apr 2003 B1
6566284 Thomas, III et al. May 2003 B2
6576532 Jones et al. Jun 2003 B1
6579463 Winningham et al. Jun 2003 B1
6603172 Segawa et al. Aug 2003 B1
6606335 Kuramata et al. Aug 2003 B1
6617643 Goodwin-Johansson Sep 2003 B1
6635110 Luan et al. Oct 2003 B1
6645295 Koike et al. Nov 2003 B1
6645797 Buynoski et al. Nov 2003 B1
6686245 Mathew et al. Feb 2004 B1
6703253 Koide Mar 2004 B2
6709982 Buynoski et al. Mar 2004 B1
6710368 Fisher et al. Mar 2004 B2
6720196 Kunisato et al. Apr 2004 B2
6727523 Morita Apr 2004 B2
6753555 Takagi et al. Jun 2004 B2
6756611 Kiyoku et al. Jun 2004 B2
6762483 Krivokapic et al. Jul 2004 B1
6767793 Clark et al. Jul 2004 B2
6784074 Shchukin et al. Aug 2004 B2
6787864 Paton et al. Sep 2004 B2
6794718 Nowak et al. Sep 2004 B2
6800910 Lin et al. Oct 2004 B2
6803598 Berger et al. Oct 2004 B1
6809351 Kuramoto et al. Oct 2004 B2
6812053 Kong et al. Nov 2004 B1
6812495 Wada et al. Nov 2004 B2
6815241 Wang Nov 2004 B2
6815738 Rim Nov 2004 B2
6825534 Chen et al. Nov 2004 B2
6831350 Liu et al. Dec 2004 B1
6835246 Zaidi Dec 2004 B2
6835618 Dakshina-Murthy et al. Dec 2004 B1
6838322 Pham et al. Jan 2005 B2
6841410 Sasaoka Jan 2005 B2
6841808 Shibata et al. Jan 2005 B2
6849077 Ricci Feb 2005 B2
6849487 Taylor, Jr. et al. Feb 2005 B2
6849884 Clark et al. Feb 2005 B2
6855583 Krivokapic et al. Feb 2005 B1
6855982 Xiang et al. Feb 2005 B1
6855990 Yeo et al. Feb 2005 B2
6867433 Yeo et al. Mar 2005 B2
6873009 Hisamoto et al. Mar 2005 B2
6882051 Majumdar et al. Apr 2005 B2
6887773 Gunn, III et al. May 2005 B2
6888181 Liao et al. May 2005 B1
6900070 Craven et al. May 2005 B2
6900502 Ge et al. May 2005 B2
6902965 Ge et al. Jun 2005 B2
6902991 Xiang et al. Jun 2005 B2
6909186 Chu Jun 2005 B2
6917068 Krivokapic Jul 2005 B1
6919258 Grant et al. Jul 2005 B2
6920159 Sidorin et al. Jul 2005 B2
6921673 Kobayashi et al. Jul 2005 B2
6921963 Krivokapic et al. Jul 2005 B2
6921982 Joshi et al. Jul 2005 B2
6936875 Sugii et al. Aug 2005 B2
6943407 Ouyang et al. Sep 2005 B2
6946683 Sano et al. Sep 2005 B2
6949769 Hu et al. Sep 2005 B2
6951819 Iles et al. Oct 2005 B2
6955969 Djomehri et al. Oct 2005 B2
6955977 Kong et al. Oct 2005 B2
6958254 Seifert Oct 2005 B2
6960781 Currie et al. Nov 2005 B2
6974733 Boyanov et al. Dec 2005 B2
6977194 Belyansky et al. Dec 2005 B2
6982204 Saxler et al. Jan 2006 B2
6982435 Shibata et al. Jan 2006 B2
6984571 Enquist Jan 2006 B1
6991998 Bedell et al. Jan 2006 B2
6994751 Hata et al. Feb 2006 B2
6995430 Langdo et al. Feb 2006 B2
6995456 Nowak Feb 2006 B2
6996147 Majumdar et al. Feb 2006 B2
6998684 Anderson et al. Feb 2006 B2
7001804 Dietz et al. Feb 2006 B2
7002175 Singh et al. Feb 2006 B1
7012298 Krivokapic Mar 2006 B1
7012314 Bude et al. Mar 2006 B2
7015497 Berger Mar 2006 B1
7015517 Grant et al. Mar 2006 B2
7033436 Biwa et al. Apr 2006 B2
7033936 Green Apr 2006 B1
7041178 Tong et al. May 2006 B2
7045401 Lee et al. May 2006 B2
7049627 Vineis et al. May 2006 B2
7061065 Horng et al. Jun 2006 B2
7074623 Lochtefeld et al. Jul 2006 B2
7078299 Maszara et al. Jul 2006 B2
7078731 D'Evelyn et al. Jul 2006 B2
7084051 Ueda Aug 2006 B2
7084441 Saxler Aug 2006 B2
7087965 Chan et al. Aug 2006 B2
7088143 Ding et al. Aug 2006 B2
7091561 Matsushita et al. Aug 2006 B2
7095043 Oda et al. Aug 2006 B2
7098508 Ieong et al. Aug 2006 B2
7101444 Shchukin et al. Sep 2006 B2
7109516 Langdo et al. Sep 2006 B2
7118987 Fu et al. Oct 2006 B2
7119402 Kinoshita et al. Oct 2006 B2
7122733 Narayanan et al. Oct 2006 B2
7125785 Cohen et al. Oct 2006 B2
7128846 Nishijima et al. Oct 2006 B2
7132691 Tanabe et al. Nov 2006 B1
7138292 Mirabedini et al. Nov 2006 B2
7138302 Xiang et al. Nov 2006 B2
7145167 Chu Dec 2006 B1
7154118 Lindert et al. Dec 2006 B2
7160753 Williams, Jr. Jan 2007 B2
7164183 Sakaguchi et al. Jan 2007 B2
7176522 Cheng et al. Feb 2007 B2
7179727 Capewell et al. Feb 2007 B2
7180134 Yang et al. Feb 2007 B2
7195993 Zheleva et al. Mar 2007 B2
7198995 Chidambarrao et al. Apr 2007 B2
7205586 Takagi et al. Apr 2007 B2
7205604 Ouyang et al. Apr 2007 B2
7211864 Seliskar May 2007 B2
7217882 Walukiewicz et al. May 2007 B2
7224033 Zhu et al. May 2007 B2
7244958 Shang et al. Jul 2007 B2
7247534 Chidambarrao et al. Jul 2007 B2
7247912 Zhu et al. Jul 2007 B2
7250359 Fitzgerald Jul 2007 B2
7262117 Gunn, III et al. Aug 2007 B1
7268058 Chau et al. Sep 2007 B2
7297569 Bude et al. Nov 2007 B2
7344942 Korber Mar 2008 B2
7361576 Imer et al. Apr 2008 B2
7372066 Sato et al. May 2008 B2
7420201 Langdo et al. Sep 2008 B2
7449379 Ochimizu et al. Nov 2008 B2
7582498 D'Evelyn et al. Sep 2009 B2
7626246 Lochtefeld et al. Dec 2009 B2
7638842 Currie et al. Dec 2009 B2
7655960 Nakahata et al. Feb 2010 B2
7777250 Lochtefeld Aug 2010 B2
7799592 Lochtefeld Sep 2010 B2
7825328 Li Nov 2010 B2
7875958 Cheng et al. Jan 2011 B2
8034697 Fiorenza et al. Oct 2011 B2
20010006249 Fitzgerald Jul 2001 A1
20010045604 Oda et al. Nov 2001 A1
20020011612 Hieda Jan 2002 A1
20020017642 Mizushima et al. Feb 2002 A1
20020022290 Kong et al. Feb 2002 A1
20020030246 Eisenbeiser et al. Mar 2002 A1
20020036290 Inaba et al. Mar 2002 A1
20020046693 Kiyoku et al. Apr 2002 A1
20020047155 Babcock et al. Apr 2002 A1
20020066403 Sunakawa et al. Jun 2002 A1
20020070383 Shibata et al. Jun 2002 A1
20020084000 Fitzgerald Jul 2002 A1
20020127427 Young et al. Sep 2002 A1
20020168802 Hsu et al. Nov 2002 A1
20020168844 Kuramoto et al. Nov 2002 A1
20020179005 Koike et al. Dec 2002 A1
20030030117 Iwasaki et al. Feb 2003 A1
20030045017 Hiramatsu et al. Mar 2003 A1
20030057486 Gambino et al. Mar 2003 A1
20030064535 Kub et al. Apr 2003 A1
20030070707 King et al. Apr 2003 A1
20030087462 Koide et al. May 2003 A1
20030089899 Lieber et al. May 2003 A1
20030155586 Koide et al. Aug 2003 A1
20030168002 Zaidi Sep 2003 A1
20030178677 Clark et al. Sep 2003 A1
20030178681 Clark et al. Sep 2003 A1
20030183827 Kawaguchi et al. Oct 2003 A1
20030203531 Shchukin et al. Oct 2003 A1
20030207518 Kong et al. Nov 2003 A1
20030227036 Sugiyama et al. Dec 2003 A1
20030230759 Thomas, III et al. Dec 2003 A1
20040005740 Lochtefeld et al. Jan 2004 A1
20040012037 Venkatesan et al. Jan 2004 A1
20040016921 Botez et al. Jan 2004 A1
20040031979 Lochtefeld et al. Feb 2004 A1
20040041932 Chao et al. Mar 2004 A1
20040043584 Thomas et al. Mar 2004 A1
20040072410 Motoki et al. Apr 2004 A1
20040075105 Leitz et al. Apr 2004 A1
20040075464 Samuelson et al. Apr 2004 A1
20040082150 Kong et al. Apr 2004 A1
20040087051 Furuya et al. May 2004 A1
20040092060 Gambino et al. May 2004 A1
20040118451 Walukiewicz et al. Jun 2004 A1
20040121507 Bude et al. Jun 2004 A1
20040123796 Nagai et al. Jul 2004 A1
20040142503 Lee et al. Jul 2004 A1
20040150001 Shchukin et al. Aug 2004 A1
20040155249 Narui et al. Aug 2004 A1
20040173812 Currie et al. Sep 2004 A1
20040183078 Wang Sep 2004 A1
20040185665 Kishimoto et al. Sep 2004 A1
20040188791 Horng et al. Sep 2004 A1
20040195624 Liu et al. Oct 2004 A1
20040227187 Cheng et al. Nov 2004 A1
20040247218 Ironside et al. Dec 2004 A1
20040256613 Oda et al. Dec 2004 A1
20040256647 Lee et al. Dec 2004 A1
20040262617 Hahm et al. Dec 2004 A1
20050001216 Adkisson et al. Jan 2005 A1
20050003572 Hahn et al. Jan 2005 A1
20050009304 Zheleva et al. Jan 2005 A1
20050017351 Ravi Jan 2005 A1
20050035410 Yeo et al. Feb 2005 A1
20050040444 Cohen Feb 2005 A1
20050045983 Noda et al. Mar 2005 A1
20050054164 Xiang Mar 2005 A1
20050054180 Han et al. Mar 2005 A1
20050056827 Li et al. Mar 2005 A1
20050056892 Seliskar Mar 2005 A1
20050072995 Anthony Apr 2005 A1
20050073028 Grant et al. Apr 2005 A1
20050093021 Ouyang et al. May 2005 A1
20050093154 Kottantharayil et al. May 2005 A1
20050104152 Snyder et al. May 2005 A1
20050104156 Wasshuber May 2005 A1
20050118793 Snyder et al. Jun 2005 A1
20050118825 Nishijima et al. Jun 2005 A1
20050121688 Nagai et al. Jun 2005 A1
20050127451 Tsuchiya et al. Jun 2005 A1
20050136626 Morse Jun 2005 A1
20050139860 Snyder et al. Jun 2005 A1
20050145941 Bedell et al. Jul 2005 A1
20050145954 Zhu et al. Jul 2005 A1
20050148161 Chen et al. Jul 2005 A1
20050156169 Chu Jul 2005 A1
20050156202 Rhee et al. Jul 2005 A1
20050161711 Chu Jul 2005 A1
20050164475 Peckerar et al. Jul 2005 A1
20050181549 Barr et al. Aug 2005 A1
20050184302 Kobayashi et al. Aug 2005 A1
20050205859 Currie et al. Sep 2005 A1
20050205932 Cohen Sep 2005 A1
20050211291 Bianchi Sep 2005 A1
20050212051 Jozwiak et al. Sep 2005 A1
20050217565 Lahreche et al. Oct 2005 A1
20050245095 Haskell et al. Nov 2005 A1
20050263751 Hall et al. Dec 2005 A1
20050274409 Fetzer et al. Dec 2005 A1
20050280103 Langdo et al. Dec 2005 A1
20060009012 Leitz et al. Jan 2006 A1
20060019462 Cheng et al. Jan 2006 A1
20060049409 Rafferty et al. Mar 2006 A1
20060057825 Bude et al. Mar 2006 A1
20060073681 Han Apr 2006 A1
20060105533 Chong et al. May 2006 A1
20060112986 Atwater, Jr. et al. Jun 2006 A1
20060113603 Currie Jun 2006 A1
20060128124 Haskell et al. Jun 2006 A1
20060131606 Cheng Jun 2006 A1
20060144435 Wanlass Jul 2006 A1
20060145264 Chidambarrao et al. Jul 2006 A1
20060160291 Lee et al. Jul 2006 A1
20060162768 Wanlass et al. Jul 2006 A1
20060166437 Korber Jul 2006 A1
20060169987 Miura et al. Aug 2006 A1
20060175601 Lieber et al. Aug 2006 A1
20060186510 Lochtefeld et al. Aug 2006 A1
20060189056 Ko et al. Aug 2006 A1
20060197123 Lochtefeld et al. Sep 2006 A1
20060197124 Lochtefeld et al. Sep 2006 A1
20060197126 Lochtefeld et al. Sep 2006 A1
20060202276 Kato Sep 2006 A1
20060205197 Yi et al. Sep 2006 A1
20060211210 Bhat et al. Sep 2006 A1
20060266281 Beaumont et al. Nov 2006 A1
20060267047 Murayama Nov 2006 A1
20060292719 Lochtefeld et al. Dec 2006 A1
20070025670 Pan et al. Feb 2007 A1
20070029643 Johnson et al. Feb 2007 A1
20070054465 Currie et al. Mar 2007 A1
20070054467 Currie et al. Mar 2007 A1
20070099315 Maa et al. May 2007 A1
20070099329 Maa et al. May 2007 A1
20070102721 DenBaars et al. May 2007 A1
20070105256 Fitzgerald May 2007 A1
20070105274 Fitzgerald May 2007 A1
20070105335 Fitzgerald May 2007 A1
20070181977 Lochtefeld et al. Aug 2007 A1
20070187668 Noguchi et al. Aug 2007 A1
20070187796 Rafferty et al. Aug 2007 A1
20070196987 Chidambarrao et al. Aug 2007 A1
20070248132 Kikuchi et al. Oct 2007 A1
20070267722 Lochtefeld et al. Nov 2007 A1
20080001169 Lochtefeld Jan 2008 A1
20080070355 Lochtefeld et al. Mar 2008 A1
20080073641 Cheng et al. Mar 2008 A1
20080073667 Lochtefeld Mar 2008 A1
20080093622 Li et al. Apr 2008 A1
20080099785 Bai et al. May 2008 A1
20080154197 Derrico et al. Jun 2008 A1
20080187018 Li Aug 2008 A1
20080194078 Akiyama et al. Aug 2008 A1
20080245400 Li Oct 2008 A1
20080257409 Li et al. Oct 2008 A1
20080286957 Lee et al. Nov 2008 A1
20090039361 Li et al. Feb 2009 A1
20090042344 Ye et al. Feb 2009 A1
20090065047 Fiorenza et al. Mar 2009 A1
20090072284 King et al. Mar 2009 A1
20090110898 Levy et al. Apr 2009 A1
20090321882 Park Dec 2009 A1
20100012976 Hydrick et al. Jan 2010 A1
20100025683 Cheng Feb 2010 A1
20100072515 Park et al. Mar 2010 A1
20100078680 Cheng et al. Apr 2010 A1
20100176371 Lochtefeld Jul 2010 A1
20100176375 Lochtefeld Jul 2010 A1
20100213511 Lochtefeld Aug 2010 A1
20100216277 Fiorenza et al. Aug 2010 A1
20100252861 Lochtefeld Oct 2010 A1
20100308376 Takada et al. Dec 2010 A1
20110011438 Li Jan 2011 A1
20110049568 Lochtefeld et al. Mar 2011 A1
Foreign Referenced Citations (50)
Number Date Country
2550906 May 2003 CN
10017137 Oct 2000 DE
10320160 Aug 2004 DE
0352472 Jun 1989 EP
0600276 Jun 1994 EP
0817096 Jan 1998 EP
1551063 Jul 2005 EP
1796180 Jun 2007 EP
2215514 Sep 1989 GB
2062090 Mar 1990 JP
7230952 Aug 1995 JP
10126010 May 1998 JP
10284436 Oct 1998 JP
10284507 Oct 1998 JP
11251684 Sep 1999 JP
11307866 Nov 1999 JP
2000021789 Jan 2000 JP
2000216432 Aug 2000 JP
2000286449 Oct 2000 JP
2000299532 Oct 2000 JP
2001007447 Jan 2001 JP
2001102678 Apr 2001 JP
3202223 Aug 2001 JP
2001257351 Sep 2001 JP
2002118255 Apr 2002 JP
2002141553 May 2002 JP
2002241192 Aug 2002 JP
2002293698 Oct 2002 JP
2003163370 Jun 2003 JP
3515974 Apr 2004 JP
2004200375 Jul 2004 JP
2009177167 Aug 2009 JP
20030065631 Aug 2003 KR
20090010284 Jan 2009 KR
544930 Aug 2003 TW
WO0072383 Nov 2000 WO
WO0101465 Jan 2001 WO
WO0209187 Jan 2002 WO
WO02086952 Oct 2002 WO
WO02088834 Nov 2002 WO
WO03073517 Sep 2003 WO
WO2004004927 Jan 2004 WO
WO2004023536 Mar 2004 WO
WO2005013375 Feb 2005 WO
WO2005048330 May 2005 WO
WO2005098963 Oct 2005 WO
WO2005122267 Dec 2005 WO
WO2006025407 Mar 2006 WO
WO2006125040 Nov 2006 WO
WO2008124154 Oct 2008 WO
Related Publications (1)
Number Date Country
20110086498 A1 Apr 2011 US
Provisional Applications (2)
Number Date Country
60923838 Apr 2007 US
60848037 Sep 2006 US
Divisions (1)
Number Date Country
Parent 11862850 Sep 2007 US
Child 12973616 US