box in the presence of quasiparticle poisoning. The model uses a master equation approach to flnd the probabilities for the box to be in the even or odd state. The transition rates between the two states are calculated assuming a flxed number of non-equilibrium quasiparticles in the leads and island of the box. We fabricated Al/AlOx/Al devices with a Cooper pair box capacitively coupled to an SET and measured the charge on the box for SET bias currents ranging from about 1 pA to 1 nA. We flnd good agreement between the theory and measurements in the temperature range from 60 mK to 300 mK. For large SET bias the poisoning in the Cooper pair box increases and the charge staircase develops additional features. Our model is capable of qualitatively reproducing the features induced by the measuring SET.
We report on the effect of the backaction of a single Cooper pair transistor electrometer (
E) on the charge state of a superconducting box (
B). The charge is
e periodic in the gate bias of
B when
E is operated near voltages $2\ensuremath{\Delta}/e$ or $4\ensuremath{\Delta}/e$. We show that this is due to quasiparticle poisoning of
B at a rate proportional to the number of quasiparticle tunneling events in
E per second. We are able to eliminate this backaction and recover
2e-charge periodicity using a new measurement method based on switching-current modulation of
E.
I report an investigation into using an rf Single Electron Transistor (rf-SET) to implement single shot readout of the quantum state of a Cooper pair box. I fabricated Al/AlOx/Al SETs and Cooper pair boxes using e-beam lithography and double-angle evaporation. The devices had typical charging energies of about 1 K and were measured at 30 mK in a dilution refrigerator. I built rf-SETs with an LC resonant circuit integrated on the chip and operating frequencies of 650 MHz and 1.5 GHz. The rf-SETs had a typical bandwidth of 30 MHz and the best achieved charge sensitivity was 8X10-6 erms/(Hz)1/2 at 1.5 GHz. The rf-SETs were used to measure the Coulomb staircase on the Cooper pair box. Measurements of the staircase show features due to quasiparticle poisoning that are dependent on temperature and SET bias. The data was compared to a theory I developed to extend existing theories of quasiparticle poisoning. My model is based on the calculation of tunneling rates to explain observations of parity effects in the Cooper pair box over a wide temperature range. Quantum state control of the Cooper pair box was demonstrated by using fast voltage pulses applied to the gate. Temporal oscillations of charge states and spectroscopy measurements reveal a dephasing time T2=200 ps at 30 mK, for a drive frequency of 30 GHz. The energy relaxation time T1=140 ns was determined from time domain measurements of the decay of the Cooper pair box charge. The fidelity of single shot measurements and the feasibility of this quantum readout scheme based on the results are discussed.