Ground-breaking research on semi-conductor quantum dots has demonstrated that quantum dots could potentially be suitable as a building block (qubit) for quantum data and calculations.
Quantum dots are quasi-zero-dimensional nanostructures which can restrain single electrons, whose spin or charge degree of freedom can subsequently be utilized to represent quantum bits (qubits).
Quantum computing approaches based on semiconductors can expand upon mature micro and Nano-fabrication technologies. This will be instrumental in scaling up to a greater quantity of reproducible qubits with practical yields, and integrated electronics.
Within the traditional information technology industry, CMOS transistors are now being developed with small enough feature sizes that quantum effects can start to take great effect. This stimulates the investigation of quantum effects in such transistors created using CMOS processes.
Triton and Nanonis Tramea
A Triton 200 Dilution Refrigerator, Nanonis Tramea with lock-in module, and SR 830 lock-in amplifier were used in this study.
The experiments detailed below were carried out on quantum dots created within CMOS fin Field Effect Transistor devices. Such transistors are made up of a number of fins atop an insulating oxide layer.
These share a single source and drain, as well as a wraparound top-gate through which conduction in the fins is controlled. Additional control is also possible by the bulk silicon substrate which can act as a back-gate.
As a result of its shape, the field effect of the top-gate is strongest in the corners of the fin cross-section. This results in the development of quantum dots when operating the finFET subthreshold, which can be seen at cryogenic temperatures (refer to transfer curves at 300 K and 30 mK in Figure 1).
Figure 1. Measurement setup for RF reflectometry and finFET device.
Rather than using measurements based on movement through the FET, it is possible to examine, with precision, the charge transitions of quantum dots, even at a time when no signal in transport is identified.
This is managed through the use of the top-gate as a dispersive sensor, detecting minor adaptations in the capacitance of the device as a result of the tunneling of single electrons, identified through the response of an LC resonator at radio-frequency (RF). This measurement is well-suited to examining the transistor in a system where transport is pinched off, because of its bandwidth and high sensitivity.
In this type of gate-based RF-reflectometry, low-level RF modulation is applied to the top-gate through a bias tee, close to the resonance frequency of an LC circuit, made up of a surface mount inductor and the parasitic capacitance of the PCB and device itself (Figure 1).
The reflected RF signal is augmented at numerous stages, followed by IQ-demodulation to gain the amplitude and stage of the reflected signal. A variation in phase of the reflected signal is linked to an actual alteration in device capacitance.
To obtain transport measurements of a quantum dot, it is necessary that electrons can tunnel on and off the dot. In order to detect such activity, the ability to measure extremely small currents, on the order of picoamps, is required.
Additionally, as tunnel barriers increase in opacity, it becomes increasingly difficult to obtain transport measurements, leading to a need for substantial integration to reach appropriate signal-to-noise.
Contrastingly, RF-reflectometry allows quick and precise measurements, even when tunnel barriers are opaque. Nonetheless, such measurements are often more difficult to read, as the signal stems from any alteration in device capacitance.
Figure 2. Measurement setup for RF reflectometry and finFET device.
The Nanonis Tramea system has the capacity to enhance current measurement processes in numerous ways. The integrated solution of extremely quick logic, data acquisition and storage, united with low-noise inputs, allow for great improvement in the speed of transport measurements, for instance, by lessening the communication time present when distinct instruments (e.g. lock-in amplifiers) are used.
Figure 3. Transport measurement using Nanonis Tramea system.
Additionally, transport and reflectometry measurements can be completed concurrently, allowing for instant comparison between these processes. This can be especially useful when investigating a novel device where it is required to identify precise operating regions via large scans within the parameter space.
Figure 4. Reflectometry measurement using Nanonis Tramea system.
It could take a number of hours for a large gate map from a device examined in transport, taken with a GPIB-interfaced lock-in amplifier controlled using a measurement computer. An illustration of this can be found in Figure 2 which demonstrates Coulomb oscillations as a function of bulk and top-gate voltage.
Figure 5. Coulomb diamonds in transport using Nanonis Tramea system.
A comparable scan at the same resolution can be achieved in just ten minutes with the Tramea system. Moreover, during direct comparison between the two hours and ten minute scans, a drift or instability can be identified in the slower scan, which is not apparent in the fast scan.
Figure 6. Coulomb diamonds in reflectometry using Nanonis Tramea system.
Steady and exact voltage sources are other important elements for such measurements, as measurements can last several days, and quantum devices have low tolerance levels. By inputting the I and Q outputs from the IQ-demodulator into two additional Tramea inputs, it is possible to gain a reflectometry scan at the same time as the direct transport measurement.
Many additional oscillations at low gate voltage, where tunnel barriers have lost transparency, are observed when contrasting the reflectometry with the transport scan.
For additional quantum dot characterization, a scan is taken following adjustments to the source-drain and top-gate voltage. Coulomb diamonds compatible with multiple quantum dots in transport and reflectometry are subsequently observed.
To conclude, this study focused on a CMOS fin-FET device. At cryogenic temperatures, signatures consistent with numerous quantum dots were noted. The Nanonis Tramea system allowed researchers to make the characterization process considerably faster and at the same time, gain transport and RF-reflectometry data, which can be valuable in exploring the parameter space of a new device.
References and Further Reading
- “Spins in a few-electron quantum dots”, R. Hanson, L. P. Kouwenhoven, J. R. Petta, S. Tarucha and L. M. K. Vandersypen – Reviews of Modern Physics 79 (2007)
- “Fast single charge sensing with an rf Quantum Point Contact”, D. J. Reilly, C. M. Marcus, M. P. Hanson and A. C. Gossard – Appl. Phys. Lett. 91 (2007)
- “Dispersive Readout of a Few-Electron Double Quantum Dot with Fast rf Gate Sensors”, J. I. Colless, A. C. Mahoney, J. M. Hornibrook, A. C. Doherty, H. Lu, A. C. Gossard and D. J. Reilly – Phys. Rev. Lett. 110 (2013)
- “Charge Dynamics and Spin Blockade in a Hybrid Double Quantum Dot in Silicon”, M. Urdampilleta, A. Chatterjee, C. C. Lo, T. Kobayashi, J. Mansir, S. Barraud, A. C. Betz, S. Rogge, M. F. Gonzalez-Zalba, and John J. L. Morton – Phys. Rev. X 5 (2015)
- “Probing the limits of gate-based charge sensing”, M. F. Gonzalez-Zalba, S. Barraud, A. J. Ferguson & A. C. Betz – Nature Communications 6 (2015)
- “A CMOS silicon spin qubit”, R. Maurand, X. Jehl, D. Kotekar-Patil, A. Corna, H. Bohuslavskyi, R. Lavieville, L. Hutin, S. Barraud, M. Vinet, M. Sanquer & S. De Franceschi – Nature Communications 7 (2016)
This information has been sourced, reviewed and adapted from materials provided by Oxford Instruments Nanoscience.
For more information on this source, please visit Oxford Instruments Nanoscience.