Please note that JavaScript and style sheet are used in this website,
Due to unadaptability of the style sheet with the browser used in your computer, pages may not look as original.
Even in such a case, however, the contents can be used safely.
While there are several conditions that a qubit must satisfy, first and foremost is the ability to freely create a superposition state, which is the most outstanding difference with bits in existing computers. Next, given that a superposition state can be created using one qubit, some kind of interrelationship (correlation) must be formed between one qubit and another. The reason for this is that if we simply prepare N independent qubits, it will not be possible to operate on 2N pieces of information. (Refer to "entangled state" described later.) To be more specific, it must be possible to achieve logical operations that two qubits can participate in.
In a 1999 experiment, we examined the operation of a single qubit formed using a solid-state device designed to achieve a superposition state, the first of the above two requirements.

To begin with, we constructed the solid-state device shown in the figure to the left. This device consisted of aluminum thin film having a thickness of about 50 nanometers. The tunnel barriers on the device were fabricated with aluminum-oxide film. The Cooper-pair box that a Cooper-pair (two coupled electrons) enters and leaves through the tunnel junction is likewise fabricated by aluminum thin film. The box is about 50×700 nanometers in size, which is appropriate for controlling Cooper pairs on a one-by-one basis.

We also utilized a special type of refrigerator called a "dilution refrigerator" to cool this circuit to an extremely low temperature environment of about 30 millikelvin on the absolute temperature scale (272.85°C below zero). Although superconducting phenomena can still occur at low temperatures above this temperature, extremely low temperatures are needed to remove all heat and noise effects on a qubit.
Now, with these preparations completed, we began the experiment by placing the box into a state in which no excess Cooper-pair is in the box (0 state).
We next applied a high-speed voltage pulse from the pulse-voltage gate to the Cooper-pair box. During the time that this pulse is being applied, a Cooper-pair tunnel is enabled between the Cooper-pair box and Reservoir, and the box changes from a state with no excess Cooper-pair (0 state) to a state with an excess Cooper-pair (1 state) only to reverses states again and repeat this cycle continuously. This coming and going of the 0 and 1 states is called "quantum oscillation" or "coherent oscillation."
A superposition of these 0 and 1 states appeared under the quantum oscillation with a Cooper-pair entering and leaving the box.
We repeated this experiment a number of times while adjusting the voltage pulse by various means such as by varying pulse width (voltage application time) and by fine tuning the pulse voltage. In this way, we succeeded in creating the superposition state that we anticipated.
This achievement was portrayed on the cover of the April 29, 1999 issue of the British journal Nature. For technical details, please see Nature, Vol. 398, No. 6730, p. 786.
The tunnel junctions shown in the figure. The oxidized film between two layers of aluminum becomes an energy barrier (tunnel barrier) with respect to the movement of a Cooper-pair. The expression "tunnel junction" indicates a sandwich structure of metal/tunnel barrier/metal.
How are these states observed?
Quantum oscillation is observed by connecting the probe electrode to an ammeter installed outside the qubit solid-state device. When there is an excess Cooper-pair inside the Cooper-pair box (1 state), the Cooper-pair breaks down after a certain lifetime becoming two electrons that flow into the probe electrode. These electrons can be observed by the ammeter as current.