Here we made an analysis of the principles of a semiconductor NMR quantum computer and its developments. The known variant of an individual-access computer (B. Kane) and alternative solid-state bulk-ensemble approach versions allowing to avoid some difficulties in implementing the first variant are considered.
As an ensemble scheme of solid-state NMR quantum computers the extension of Kane's many-qubits silicon scheme based on the array of 31 P donor atoms are spaced lengthwise of the strip gates is considered. The possible planar topology of such ensemble quantum computer is suggested. The estimation of the output NMR signal was performed and it was shown that for the number N>=10^5 of ensemble elements involving L~10^3 qubits each, the standard NMR methods are usable. As main mechanisms of decoherence for low temperature (<0.1K), the adiabatic processes of random modulation of qubit resonance frequency determined by secular part of nuclear spin hyperfine interaction with electron magnetic moment of basic atom and dipole-dipole interaction with nuclear moments of neighboring impurity atoms was considered, It was made estimations of allowed concentrations of magnetic impurities and of spin temperature whereby the required decoherence suppression is obtained. Semiclassical decoherence model of two qubit entangled states is also presented. As another variant of the solid-state ensemble quantum computer, the gateless architecture of cellular-automaton with antiferromagnetically ordered electron spins is also discussed here.
Problems in realization of silicon-based solid-state NMR quantum computer with ensemble addressing to qubits are considered. It is presented the extension of Kane's scheme to ensemble approach version with strip gates. For the initialization of nuclear quantum states it is proposed to use the solid-state effect in ENDOR technique whereby the nuclear spins can be practically fully polarized or, that is the same, indirect cooled to the spin temperature less than ~ 1 mK. It is suggested the possible planar silicon topology of such ensemble quantum computer and shown that the measurement with standard NMR methods signal of L ~ 10^3 qubit system may be achieved for a number of ensemble components N >= 10^5. As another variant of ensemble silicon quantum computer the gateless architecture of cellular-automaton is also considered. The decoherence of quantum states in the ensemble quantum computers and ways of its suppression is also discussed.
The computer industry has followed Moore's Law closely and faithfully over the past few decades. However, transistors cannot continue to shrink at their current rate forever, and new methods of computation must be explored. Quantum computation is one such method that has received much attention over the past few years and will heavily rely on technological advances in the smart electronics and nanotechnology arena. In this review, we will present some of the problems facing classical computers and why quantum computers may be a viable alternative. We will briefly describe some of the "killer" quantum applications, such as Deutsch's, Shor's and Grover's algorithms that demonstrate the computational powers of quantum computation. Kane's solid state quantum computer in silicon promises to demonstrate some of these applications. However there remain many significant technological difficulties which will need to be overcome if we are to see a useful quantum computer. The main technological challenges, for Kane's solid-state computer, of interest to the smart materials and structures community, will be highlighted.