Spin qubits in semiconductor quantum dots are considered to be a promising platform for scalable
quantum computing. This is mostly attributed to their small size, the use of scalable VLSI fabrication
techniques, their long T1 time and the possibility to isotopically purify semiconductors such as silicon
or germanium to avoid hyperfine induced dephasing from nuclear spins. In the case of p-type devices,
the hole spin experiences a strong spin-orbit interaction that enables electrical qubit control simply by
applying an AC voltage to one of the gate electrodes. However, this coupling of the spin to its orbital
degree of freedom also makes such qubits susceptible to charge noise affecting their coherence. This
can be mitigated by selectively operating the qubit at sweet spots which potentially allow for fast
manipulation while “hiding” the qubit from its noise environment [1].
We have implemented hole spin qubits in two different semiconductor device platforms: One is based
on a silicon bulk fin field effect transistor (see (a) and [2,3]) while the second uses gate defined
quantum dots in a planar germanium quantum well (see (b) and [4]).
For the FinFET qubits I’ll discuss how we can control the number of holes down to the last one by
reading out the charge state with a lumped element resonator. This also highlights how disorder and
strain can impact MOS qubits. We have implemented both single and two qubit gates in these devices
and we find that the spin-orbit interaction leads to strong anisotropies in the g-tensor and in the
exchange coupling between the two qubits.
For the planar germanium platform, I will present our results on germanium quantum wells confining
a 2D hole gas (2DHG) with a low percolation density and state-of-the-art peak mobility. We also have
initial results on exchange coupled hole spin qubits with an integrated charge sensor where we can
compare g-tensor anisotropies with those of the FinFET devices. I’ll discuss the expected impact of
noise on hole spin qubit coherence, as well as strategies to mitigate this.
Finally, I’ll give an outlook on how the two platforms might be scaled to larger systems in the future
and what challenges may lie ahead on that path.

[1] N. W. Hendrickx and A. Fuhrer. “A spin qubit hiding from the noise.”, Nature Nanotechnology, 17, 1040,
(2022).
[2] S. Geyer, L. C. Camenzind, L. Czornomaz, V. Deshpande, A. Fuhrer, R. Warburton, D. M. Zumbühl, and A.
Kuhlmann. “Self-aligned gates for scalable silicon quantum computing.”, Appl. Phys. Lett., 118, 104004
(2021).
[3] L. C. Camenzind, S. Geyer, A. Fuhrer, R. J. Warburton, D. M. Zumbühl, and A. V. Kuhlmann. “A hole spin
qubit in a fin field-effect transistor above 4 kelvin.“, Nature Electronics, 5, 178, (2022).
[4] S. Geyer, B. Hetenyi, S. Bosco, L. C. Camenzind, R. S. Eggli, A. Fuhrer, D. Loss, R. J. Warburton, D. M. Zumbühl,
and A. V. Kuhlmann. “Two-qubit logic for holes in silicon fin field-effect transistors with spin-orbit induced
anisotropic exchange interaction.” arXiv: 2212.02308.
[5] N. W. Hendrickx, W. I. L. Lawrie, M. Russ, F. van Riggelen, S. L. de Snoo, R. N. Schouten, A. Sammak, G.
Scappucci, and M. Veldhorst. A four-qubit germanium quantum processor. Nature, 591, 580 (2021).