Population Inversion in a Single InGaAs Quantum Dot Using the Method of Adiabatic Rapid Passage

Preparation of a specific quantum state is a required step for a variety of proposed quantum applications. We report an experimental demonstration of optical quantum state inversion in a single semiconductor quantum dot using adiabatic rapid passage. This method is insensitive to variation in the optical coupling in contrast with earlier work based on Rabi oscillations. We show that when the pulse power exceeds a threshold for inversion, the final state is independent of power. This provides a new tool for preparing quantum states in semiconductor dots and has a wide range of potential uses.

Figure 1

Schematic representation of the eigenenergies of the two-level quantum system in the rotating frame of the optical field in the (a) Rabi excitation regime with a transform-limited 2 ps pulse and (b) the ARP regime with a chirped pulse of 15 ps pulse width. The red (blue) color represent the pure ground (excited) state of the system and the purple color represents the mixing of the two states during the pulse. For a system initially in the ground state, the Bloch spheres show that the final state of the system depends on the pulse area, $\theta$, of the interacting field in (a), whereas in (b) the final inversion of the system is independent of the pulse energy.

Figure 2

(a) Schematic diagram of the sample structure and experimental setup. The linear chirp on the optical pulses is generated by a parallel grating pair with $1200\mathrm{grooves}/\mathrm{mm}$. The value of the chirp can be changed by varying the distance between the gratings. The laser light is focused by the objective down to $\sim 1\mu \mathrm{m}$ over a 200 nm aperture in a gold mask. The sample consists of a single layer of InGaAs quantum dots embedded in a Schottky diode structure. The band diagram of the diode shows the relative positions of each layer in the growth direction and in energy. (b) A spectrally resolved autocorrelation scan (top) of a transform-limited pulse 2 ps FWHM and the cross correlation (bottom) of the same pulse chirped to give 15 ps FWHM.

Figure 3

(a) A PL map of the emission from various excitonic states at different bias voltages. $X^{+}$ is the positively charged exciton, $X$ is the neutral exciton, $BX$ is the neutral biexciton, and $X^{-}$ is the negatively-charged exciton. (b) Photocurrent versus the wavelength of a continuous-wave excitation laser at $-1\mathrm{V}$ bias voltage. The FWHM of the line shape is $50\mu \mathrm{eV}$.

Figure 4

(a) Electrical readout signals of Rabi oscillation (solid circles) excited by the 2 ps transform-limited pulse and ARP (open squares) excited by the 15 ps chirped pulse. The curves are the difference between the on-resonant (1.343 eV) photocurrent and the corresponding off-resonant (1.341 eV) background photocurrent. The small Bloch spheres indicate the position of the Bloch vector at given excitation powers in each case. (b) Simulation of the corresponding electrical readout signals in (a) using the same optical pulse characterization and dot properties as the experiment assuming a recombination time of 1 ns and tunneling $T_2$ of 300 ps. (c) Simulation of electrical readout signals using recombination time of 1 ns, tunneling $T_2$ of 25 ps and a 15 ps transform-limited (chirped) pulse for the Rabi (ARP) simulation.