The third year of the Centre for Quantum Computer Technology has seen the achievement of significant milestones in the fabrication and measurement of solid-state quantum computer prototypes, together with important new conceptual advances for solid-state quantum computation.

The Centre’s research is a nationally coordinated experimental and theoretical effort directed towards the realisation of scalable solid-state quantum computer prototype system based on a precisely controlled array of phosphorus donors embedded in a silicon crystal. This goal is being approached using two parallel strategies, a ‘top-down’ approach in which single phosphorus donors are implanted into the device through masks and a ‘bottom-up’ approach that offers the ultimate in atomically precise construction. Here the devices are built atomic layer by atomic layer, using scanned probe lithography and epitaxial semiconductor growth.

During 2002 the Centre achieved the following important milestones in solid-state device fabrication and measurement.

• We have demonstrated controlled incorporation of single phosphorus atoms in silicon at pre-defined sites—the first step towards the realisation of atomic-scale doping of a semi-conductor.
• A fully configured device was fabricated using the top-down ion implantation route with the low keV implant energies required for a functioning prototype.
• The ability to read-out the charge state of these devices at microsecond timescales was demonstrated using a radio-frequency (rf) twin-SET simulation device.
• A new charge-based scheme for silicon quan-tum computing was conceived and patented.

Bottom-up STM fabricated devices
Previously we have demonstrated a strategy to adsorb single phosphine molecules to a silicon surface in precise locations using a hydrogen resist. We have now managed to incorporate the phosphorus atoms from these individual molecules into the silicon surface at atomically precise locations using a critical anneal, see Figure 1.

Figure 1. Left: A hydrogen terminated silicon surface with a nanometer sized area of bare silicon (circled) produced by hydrogen desorption with an STM tip. Right: A single P atom has been incorporated into the surface, to form a P-Si-H ‘hydrided heterodimer’, at the precise position defined by the STM.

As a result the P atoms are now ‘locked in’ to the surface, preventing them from desorbing or diffusing across the surface during subsequent silicon encapsulation. This figure represents an important step towards the ability to perform single atom doping in silicon.

Secondly we have addressed the key problem of minimising P segregation during silicon encapsulation. The influence of the Si growth temperature on P segregation was investigated using Secondary Ion Mass Spectrometry (SIMS) and STM measurements. SIMS depth profiling confirmed that phosphorus segregated less than ~ 5 nm during Si encapsulation at 250°C (below the detection limit of SIMS). Using the STM itself as a characterisation tool we were able to show that we could reduce this by a factor of ~5 by encapsulating the phosphorus atoms at room temperature.

Expansion of the Atomic Fabrication Facility
In 2002 the Atomic Fabrication Facility was significantly expanded with the addition of a $3.1 million multi-chamber scanning tunnelling microscope and molecular beam epitaxy system across two rooms of the AFF laboratory — see figure 2. This new system contains several unique design features that provide the necessary registration capability and high purity silicon growth required for the fabrication of the final multi-qubit device, and importantly for realising atomic-scale devices in silicon.
Top-Down Ion-Implanted Devices

Figure 2 New combined UHV MBE/STM/SEM system, commissioned in November 2002, which is custom-designed to enable bottom-up fabrication of buried P arrays with registration to surface control gates and SETs.

The first implanted donor test devices were constructed during 2002. These devices were designed to demonstrate controlled single electron transfer events between implanted donors. The structures incorporate control gates for manipulating electron transfer, together with integrated single electron transistors (SETs) for sensing charge motion (see Figure 3). Their electrical characterisation is currently underway and represents an exciting moment for the Centre. During 2002 researchers from a number of programs within the Centre also conceived a novel charge-based qubit scheme that acts as a intermediate stage for spin qubit development.

Figure 3 (a - c) SEM and optical images of complete buried atom device showing critical dimensions. (d) Nanoapertures 24nm in dimension for ion implantation. (e) Completed device design with optmised SETs

Quantum Device Measurement
Amazing results were achieved by the quantum measurement team with the development and demonstration of the twin radio frequency single electron transistor (rf-SET), a key device for the fast and efficient readout of solid-state qubits. They established the ability to sense charge motion on sub-microsecond timescales (see figure 4) thereby creating a world-class capability in efficient charge readout and moving us significantly closer to the ultimate goal of single spin detection.

Figure 4. Measurements of a rf twin single electron transistor device at mK temperatures, showing periodic “Coulomb Blockade” oscillations in the conductance(left). Radio frequency measurements of the power reflected from the rf-SET (right) allow changes in the the device conductance to be monitored on sub-ms timescales, providing high speed detection of single electron motion.

In addition to the work on the rf-SET, 2002 saw several other key projects achieve success. This included the fabrication of MOSFET devices for the characterization of interface trap densities and the development of a fast pulse-gate capability aimed at delivering voltages pulses to on-chip qubit gates with rise/fall times of the order of 40ps. Finally, congratulations are due to the team whose results on spin manipulation in ultra-low-disorder quantum wires were published in Physical Review Letters.

During 2002 the Centre published 80 papers with 5 Physical Review Letters and 28 Physics Review A,B or E. It also lodged 5 patents and gave over 100 conference presentations, 27 of which were invited. During the year the Centre was involved in a nationally competitive bid to reposition itself within the Australian Research Council’s newly announced Centres of Excelllence (COE). The main thrust of our COE application is to enable the Centre to address key issues of fault tolerant quantum computing. In December 2002 it was announced that our COE application was successful, making us one of the eight COEs awarded nationally.

Michelle Simmons, Alex Hamilton and Robert Clark