Progress and Challenges for Semiconductor Spin Qubits

Since the first schemes for spin qubits in semiconductors were proposed [1,2] in 1998 they have generated wide interest because of intrinsically long spin coherence times combined with the prospect of employing semiconductor manufacturing to realize large numbers of integrated qubits. Initial demonstrations utilized quantum dot devices in gallium-arsenide to confine electron spins [3], and by 2012 the first silicon qubits appeared, using either quantum dots or donor potentials to confine electron [4,5] or nuclear spins [6]. Most research nowadays is focused on silicon (and SiGe) due to the very long coherence times (> 1 s) attainable [7] using isotopically-enriched ²⁸Si, which can be considered a spin vacuum. Two-qubit gates using exchange-coupled spins in neighbouring Si quantum dots have now been demonstrated [8], as has strong-coupling between a single spin in Si and a microwave frequency photon [9], opening a path to long-distance qubit coupling. Despite recent rapid progress, a number of challenges must be addressed to realize large-scale spin qubit systems with the fidelities required for fault-tolerant QC. Chief amongst these are materials-related issues, in particular disorder potentials that lead to variations between qubits, and charge noise that degrades fidelities. I will discuss these challenges, and possible solutions accessible with the aid of the existing semiconductor industry.

[1] D. Loss & D.P. DiVincenzo, Phys. Rev. A 57, 120 (1998).
[2] B. E. Kane, Nature 393, 133 (1998).
[3] J.R. Petta et al., Science 309, 2180 (2005).
[4] B.M. Maune et al., Nature 481, 344 (2012).
[5] J.J. Pla et al., Nature 489, 541 (2012).
[6] J.J. Pla et al., Nature 496, 334 (2013).
[7] J.T. Muhonen et al., Nature Nanotechnology 9, 986 (2014).
[8] M. Veldhorst et al., Nature 526, 410 (2015).
[9] X. Mi et al., arXiv:1710.03265.