© 1999 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Nanomedicine, Volume I: Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999

10.2.4.2 Quantum Computers

Digital computers manipulate discrete units of information, with data strings stored as binary digits, or bits. Any two-state physical system (e.g., high or low voltages in semiconductor circuits) can represent these bits. Similarly, a quantum computer would use quantum states of atomic or molecular systems to store data. For example, the states of a hydrogen atom could be assigned such that a wavefunction corresponding to a state of hydrogen would represent a bit of data. Shining laser light on the atoms would trigger processing the data by inducing transitions between electronic states.

Quantum systems, however, exhibit a peculiar phenomenon known as superposition, in which several discrete states can be processed by a single physical system at once, crudely analogous to the musical effect of a single note being made up of many different harmonics.¹²⁵⁹ Superposition gives quantum systems the potential to be enormously more powerful than conventional computers. Since more than one state can be supported, quantum bits, or qubits, may store a mixture of many bits of data at the same time. A superposition of N qubits can store 2^N binary digits. An operation on these mixtures is massively parallel, effectively performing many calculations simultaneously. For example, photons interacting with an atom in a superposition of states drive all the states in the superposition, producing another superposition corresponding to all solutions of the original states. The advantage over conventional computers is not merely linear, but exponential: A conventional processor operates sequentially on 64-bit numbers, but a quantum computer with 64 qubits would simultaneously operate on the full set of 2⁶⁴ (~10¹⁹) binary values. An N-qubit quantum computer could model an N-body system in real time⁶; in contrast, the number of operations required by a conventional computer to perform such a simulation increases exponentially with the number of bodies.

Since the early 1980s, theoretical interest in quantum computing has developed rapidly.^{6,1996-2004,2013} Many groups have begun designing quantum computational architectures and constructing physical quantum gates and devices. In 1995, one group at NIST made a controlled NOT gate of two qubits from a cooled beryllium ion in a radio-frequency trap,²⁰⁰⁵ while a French group used single photons trapped in quantum cavities to control cesium atom states.²⁰⁰⁶ In 1999, experiments began on an analog quantum computer using electrons floating on liquid helium.³²⁷⁶ Others work with NMR techniques,^{2007,2008,2014} storing data not in electron states but in nuclear spins, which are less susceptible to perturbations.²⁰⁰² Resonance effects between proton and electron spins in hydrogen atoms can make AND and NOT gates, hence NAND gates, from which all Boolean computers can be made. It has been suggested that by encoding the amplitudes of a couple of thousand electron states in a superposition, large amounts of data could be written onto a single atom (see Section 10.2.4.3). The discovery of certain error-correcting procedures,²⁰⁰³ since verified experimentally,²¹⁴¹ was another important breakthrough. Unlike conventional devices, it is not possible to simply check that qubits are in the right states, because the act of measuring destroys the coherence. But Shor²⁰⁰⁹ proved that information can be reliably stored and read from a several qubit system even if one of the bits is corrupted, and it has been shown that error correction can be applied during the computation process itself.²⁰¹⁰ This opens the door to many-qubit systems; by 1998, two groups had already performed 4-bit sums, and one group had demonstrated a 2-qubit device.²³²⁹ Quantum game theory has also been investigated.²⁸⁶⁷

Last updated on 24 February 2003