Amorphous BaTiO3 layers deposited on SrTiO3 (001) substrates at room temperature were subsequently crystallized using solid phase epitaxy (SPE). Heating an initially amorphous BaTiO3 layer in air at 650 °C for 3 h resulted in crystallization with components in two distinct crystallographic orientation relationships with respect to the substrate. Part of the volume of the BaTiO3 layer crystallized in a cube-on-cube relationship with the substrate. Other volumes crystallized in four variants of a 70.5° rotation about ⟨110⟩, resulting in a ⟨221⟩ surface normal in each case. Each of these four variants forms a Σ = 3 coincident site lattice with respect to the SrTiO3 substrate and the cube-on-cube oriented BaTiO3. Heating for the same duration and temperature in a reducing gas atmosphere resulted in the formation of polycrystalline BaTiO3 with no preferred crystallographic orientation. The dependence on the gas atmosphere indicates that it may be possible to tune the annealing time, temperature, and atmosphere to produce a single crystalline BTO on STO by SPE or produce a desired distribution of orientations.
Optical illumination of quantum-dot qubit devices at cryogenic temperatures, while not well studied, is often used to recover operating conditions after undesired shocking events or charge injection. Here, we demonstrate systematic threshold voltage shifts in a dopant-free, Si/SiGe field effect transistor using a near infrared (780 nm) laser diode. We find that illumination under an applied gate voltage can be used to set a specific, stable, and reproducible threshold voltage that, over a wide range in gate bias, is equal to that gate bias. Outside this range, the threshold voltage can still be tuned, although the resulting threshold voltage is no longer equal to the applied gate bias during illumination. We present a simple and intuitive model that provides a mechanism for the tunability in gate bias. The model presented also explains why cryogenic illumination is successful at resetting quantum dot qubit devices after undesired charging events.
Valley splitting affects the energy dispersion of silicon quantum dot qubits, and occasionally manifests itself through some striking features. Here, the authors observe a strong correlation between unexpected ``sweet spots'' and ``hot spots'' in the coherence rates of a quantum-dot hybrid qubit and in some anomalous features in the energy dispersion. Through tight-binding simulations, they are able to attribute such effects to disorder at the quantum-well interface and speculate on the possibility of harnessing disorder to enhance qubit coherence times.
We investigate the lifetime of two-electron spin states in a few-electron Si/SiGe double dot. At the transition between the (1,1) and (0,2) charge occupations, Pauli spin blockade provides a readout mechanism for the spin state. We use the statistics of repeated single-shot measurements to extract the lifetimes of multiple states simultaneously. When the magnetic field is zero, we find that all three triplet states have equal lifetimes, as expected, and this time is ~10 ms. When the field is nonzero, the T(0) lifetime is unchanged, whereas the T- lifetime increases monotonically with the field, reaching 3 sec at 1 T.