June 25, 1996, entitled Three dot computing elements; granted to
David P. DiVincenzo. This invention describes an invention logic and memory elements of atomic or near-atomic scale useful in computer central processing units. These elements consist of two quantum dots having opposite states and a third quantum dot situated between the two quantum dots and in physical contact with them. The third quantam dot is of a material which makes the opposite states of the first two quantum dots energetically favorable. In particular, there is provided by the invention a spin flip-flop suitable for use as electronic logic and memory in a quantum computer. The spin flip-flop is designed to have two highly stable states, encoded entirely in the arrangement of electronic spins in the structure. Switching between the two states is accomplished by fast electromagnetic pulsing generally and by optical pulsing in the case of the spin flip-flop. The two stable states are the up-down and the down-up spin states of two single electrons placed into two neighboring electronic quantum dots typically by doping or by a field effect. The operation of the device is facilitated and stabilized by the presence of a small particle or dot of an antiferromagnetic material placed between the two electronic dots, and in physical contact with both of them.
2. 5,768,297, patented June 16, 1998, entitled Method for reducing decoherence in quantum computer memory; granted to Peter W. Shor. This invention describes a methodology for reducing the rate of decoherence in quantum memory. A procedure is disclosed for storing an arbitrary state of n qubits using expanded groupings of these n qubits in a decoherence-resistant manner. Each qubit of the original n qubits is mapped into a grouping of qubits, and the process will reconstruct the original superposition simultaneously correcting both bit and phase errors if at most one qubit decoheres in each of these groups of qubits. In one preferred embodiment of the present invention, a method is disclosed for decoding a set of N'(2m+1) qubits, wherein the set of N'(2m+1) exposed qubits has undergone possible decoherence from a superposition of states due to exposure to an environment. The method comprises the steps of decoding the set of N'(2m+1) exposed qubits using a repetition decoder to yield a set of N' decoded qubits, applying a first transformation to each decoded qubit in the set of said N' decoded qubits to yield a set of N' transformed decoded qubits, wherein the first transformation is represented as a unitary matrix comprising matrix entries of equal absolute value, and decoding the set of N' transformed decoded qubits using an error correcting decoder to yield a set of n decoded qubits.
3. 5,7930,911, patented August 11, 1998, entitled Parallel architecture for quantum computers using ion trap arrays; granted to Ralph Godwin Devoe. This invention discloses a parallel architecture of quantum logic gates and quantum communication channels for a quantum computer, thereby achieving advantageous efficiency and computation speed. The architecture of the invention enables parallel memory operations on large quantum words, and permits the application, to the quantum case, of parallel algorithms for mathematical operations such as addition and multiplication. The invention also includes a novel apparatus for realizing parallel architecture using an array of miniature elliptical ion traps, with as many traps as there are bits in a quantum word. The ion trap array preferably uses an elliptical planar geometry, which can microfabricated by photolithography. Quantum information is transferred from one ion trap to another by either an optical coupling via a high finesse resonant cavity (photon coupling) or by electrostatic coupling of the ions' mechanical motion (phonon coupling).
4. 5,838,436, patented Nov. 17, 1998 entitled Multi-purpose quantum computing ; granted to Stephen P. Hotaling. This invention discloses a method and apparatus are provided for a general purpose photonic computer. A data signal is input through an encoder to encode such signal with an instruction. The encoded signal is transmitted by means of a laser beam to an input buffer where it interferes with a reference beam so as to form an interference pattern therein as a hollogram, IPH. A read beam is directed through the IPH and through a decoder which reads the instruction as having, e.g. an OP Code, data source and destination. The decoded instruction is forwarded on the read beam to ALU spin media which respond to the instruction by flipping spins between >2 energy levels, in one or more sequences of data patterns which are read or measured by one or more sensors. Such sensors can be one or more of RF, microwave or optical sensors, which sensors output radix >2 data signals for, e.g. storage, display or further processing as desired. Thus the present invention teaches a novel exploitation of photon-induced, quantum-mechanical spin transitions in spin media. The input signal can be from a keyboard, camera, bar code or other input source.
5. 5,917,322, patened June
29, 1999 entitled Method and apparatus for quantum information processing
; granted to Neil Gershenfeld. This invention discloses an approach to processing
quantum information which uses a bulk ensemble of a very large number of
identical entities as its source of quantum degrees of freedom. The information
is represented as the deviation from uniform population probability for at least one of the quantum states of the ensemble. Coherences between quantum states, created when the ensemble is modified in a way that removes it from thermal equilibrium can serve as effective degrees of freedom. A bulk thermal ensemble of nuclear spins in a static magnetic field is treated using nuclear magnetic resonance pulses for preparation of an initial pure state, and effecting arbitrary single-spin and coupled multi-spin rotations. Readout of the result is accomplished by observation of the magnetization of the ensemble.
6. 5,940,193 patened August 17, 1999 entitled General purpose quantum computing granted to Stephen P. Hotaling et al. This invention discloses a method and apparatus for a general purpose photonic computer. A data signal is input through an encoder to encode such signal with an instruction. The encoded signal is transmitted by means of a laser beam to an input buffer where it interferes with a reference beam so as to form an interference pattern therein as a hologram, IPH. A read beam is directed through the IPH and through a decoder which reads the instruction as having, e.g. an OP Code, data source and destination. The decoded instruction is forwarded on the read beam to ALU spin media which respond to the instruction by flipping spins between two energy levels, in one or more sequences of data patterns which are read or measured by one or more sensors. Such sensors can be RF, microwave or optical sensors, which sensors output Radix=2 or digital data signals for, e.g. storage, display or further processing as desired. Thus the present invention teaches a novel exploitation of photon-induced, quantum-mechanical spin transitions in spin media. The input signal can be from a keyboard, camera, bar code or other input source.
7. 6,081,882 patened June
27, 2000 entitled Quantum acceleration of conventional non-quantum computers
granted to Carroll P. Gossett. This invention discloses a A process
and apparatus for quantum acceleration of a conventional computer by coupling
a few quantum devices to the conventional computer. Initially, a first, second,
and third maximally entangled qubit are prepared in a Greenberger-Horne-Zeilinger
state. A fourth qubit is prepared in a perfect superposition of states which
is unentangled from the three qubits. The second qubit is then measured and
its measured value is input to the conventional computer. The conventional
computer operates on this measured input value and performs an inverse oracle
function. The second qubit is modified according to the output from the conventional
computer. This modified qubit is used as one of two control inputs
for controlling a quantum gate. The other control input is the fourth qubit.
The quantum gate phase inverts the third qubit according to the two control
inputs. A measurement of the complement of the first qubit is taken in order
to produce the necessary quantum interference of the third qubit. The third
qubit can now be measured to find the correct final solution. An N-bit quantum
accelerated computer can be constructed by implementing N numbers of 4-qubit
8. 6,256,767 patented July 3, 2001 entitled Demultiplexer for a molecular wire crossbar network granted to Philip J. Kuekes and R. Stanley Williams. A demultiplexer for a two-dimensional array of a plurality of nanometer-scale switches (molecular wire crossbar network) is disclosed. Each switch comprises a pair of crossed wires which form a junction where one wire crosses another and at least one connector species connecting said pair of crossed wires in said junction. The connector species comprises a bi-stable molecule. The demultiplexer comprises a plurality of address lines accessed by a first set of wires in the two-dimensional array by randomly forming contacts between each wire in the first set of wires to at least one of the address lines. The first set of wires crosses a second set of wires to form the junctions. The demultiplexer solves both the problems of data input and output to a molecular electronic system and also bridges the size gap between CMOS and molecules with an architecture that can scale up to extraordinarily large numbers of molecular devices. Further, the demultiplexer is very defect tolerant, and can work despite a large number of defects in the system. This appears to be the successful bridge between the micron scale electronics and practical nanocomputing.
9. 6,459,095 patented October 1, 2002 entitled Chemically synthesized and assembled electronics devices granted to James R. Heath, Philip J. Kuekes and R. Stanley Williams. A route to the fabrication of electronic devices is provided, in which the devices consist of two crossed wires sandwiching an electrically addressable molecular species. The approach is extremely simple and inexpensive to implement, and scales from wire dimensions of several micrometers down to nanometer-scale dimensions. The device of the present invention can be used to produce crossbar switch arrays, logic devices, memory devices, and communication and signal routing devices. The present invention enables construction of molecular electronic devices on a length scale than can range from micrometers to nanometers via a straightforward and inexpensive chemical assembly procedure. The device is either partially or completely chemically assembled, and the key to the scaling is that the location of the devices on the substrate are defined once the devices have been assembled, not prior to assembly.
Any of the above patents can be obtained on the web at either IBM or US Patent Office