Institut des
NanoSciences de Paris
insp
Accueil > Evénements > Séminaires > Archives 2012 > Towards the quantum (...)
insp
4.jpg

Séminaire général

Towards the quantum internet of the mid-21st C : building quantum bits in silicon

D. N. Jamieson

Our civilization is built on the band-gap of silicon. It is impossible to do anything in the western world without leaving deep digital footprints in the server farms that hold data about our web browsing habits, public transport use, credit card purchases, power consumption, telephone calls and many other everyday activities. Managing and securing these data presents near-term challenges, especially if we are to securely control dispersed, low emission technologies that may be adopted by the mid-21st C. To date we have been sustained by the extraordinary progress in the ever expanding capabilities of silicon nano-scale Complementary Metal-Oxide-Semiconductor (CMOS) field effect transistors. Present generation devices are now so small that the channel length in the transistors ( 20 nm) is comparable in size to the Bohr orbit of the donor electrons ( 1.22 nm for Si:P). The devices are sensitive to the distribution of the donor atoms themselvesand, when cooled, also to the quantum state of single donors. The first issue is flagged by the International semiconductor roadmap for 2011. We have exploited the second issue to open a pathway to the quantum internet of the mid-21st C that could address some of the near-term challenges and exploits the fact isotopically pure 28-Si is free of unpaired spins : a “semiconductor vacuum”. Our approach is to engineer silicon CMOS devices with single phosphorous atom implanted with a deterministic doping method that is cited by the 2011 roadmap. Our devices have nowproved the ability to perform single shot readout of the donor electron spin. We use electron spin resonance to drive Rabi oscillations to showa coherence time (T2) exceeding 200 µs suggesting a single electron spin can be used as a long-lived quantum bit. Further, the same device has allowed us to perform nuclear magnetic resonance on the single 31P nuclear spin by coupling the electron and nuclear spins and hence providing access to an even longer-lived nuclear qubit. This presentation reviews the remaining challenges of building a large scale silicon quantum device for computation and communication.