Transcript Modeling - University of Melbourne

Modeling Read-Out for
Solid-State Quantum
Computers in Silicon
Vincent Conrad
Supervisors:
C.Pakes & L. Hollenberg
Introduction
Solid-State Quantum Computers in Silicon
Single Electron Transistors
Modeling Read-Out
Results & Conclusion
Further Work
Solid-State Quantum Computers
in Silicon
Scalable
Hard Qubits
Kane Quantum Computer
Spin-Qubit
Buried Donor Charge Qubit
Quantum Computer
Charge-Qubit
Kane Quantum Computer
Kane Quantum Computer
spin-qubit
Buried Donor Charge-Qubit
Quantum Computer
Charge-qubit
Single Electron Tunneling
Potential Barriers
Quantised energy levels
Fermi Level of Source is lower then
first unoccupied level of dot
{
Energy spacing must be greater
then thermal smearing
E  kBT
Single Electron Tunneling
Applying a potential shifts
energy
Fermi energy of source the
nowdot’s
higher
thenlevels.
dot’s 1st unoccupied energy level.
An electron can now occupy the dot.
Coulomb blockade prevents others.
S E T Transistor
Single-Electron
control
Including a control gate
allows us to manipulate
the island’s energy
levels.
source
dot (island)
drain
controlled single electron tunneling
Orthodox SET theory
The only quantized energy levels occur in the island.
The time of electron tunneling through the barrier is
assumed to be negligibly small.
Coherent quantum processes consisting of several
simultaneous tunneling events
("co-tunneling") are ignored.
(ne  qo )
n1eC2 n2 eC1
F


2C
C
C
2
Energy stored in a capacitor
Work done by tunneling events
SET Sensitivity
extremely
sensitive to
voltage
variations on
the island
conductance
control gate voltage
electron motion
Read-Out
drain
Single electron’s motion between dopants.
Vary potential on the
sourceisland (control gate).
island
Induced island charge.
Require induced charge > SET sensitivity.
electron
hole
Q = CV
Spin-Qubit Read-Out
q  Ce (Ve  Vi )  Ch (Vh  Vi )
Q = CV
Charge Qubit Read-Out
q  C (V  Vi )  Ch (Vh  Vi )
'
h
'
h
Results
N.B. For charge qubit q is difference between two points.
q = 2.49x10-2 e
q = 2.14x10-2 e
Conclusions
Induced island charge >> SET Sensitivity
2 x 10-2 e
>>
3.2 x 10-6 e
Need an answer before information loss
Electron-spin relaxation time (spin-qubit)
Charge dissipation time (charge-qubit)
Time given by shot-noise limit

6

2
Well
t min inside
 3.2 estimated
10 e / qtimes
for both information loss
mechanisms
Both qubit types should produce measurable
results using current technology made by the
SRCQCT
Further Work
More complete architecture simulations.
Full type3 simulation ISE-TCAD input files prepared.
Estimate 100 000 node points required.
Accounts and ISE-TCAD setup at HPC.
Beowulf in-house cluster under construction.
Matching simulations to experiment.
Convert type3 simulation to replicate macroscopic
charge-qubit experiment.
A circuitry interlude:
Nano-circuits
are pretty darn small.
Type3 Device
hole
electron
electron and hole (spin-qubit)
Integrated Systems Engineering –
I S E –Computer
T C A D Aided
Technology
Design
coarse
Poisson’s Equation

Software package
designed
for microchip
industry.

  (r )
2
User specifies mesh
  V (r )
spacing to vary
over  0 DESSIS
MESH
PICASSO
regions of interest.Graphical user interface for visual
analysisAC
of analysis
simulations.
Orthodox approach to single-electron tunneling.
fine
I  A V  j    C V
Extend ISE-TCAD to nanotech/mesoscopic devices.