Slow-light dynamics from electromagnetically-induced-transparency spectra

TLDR: In this article, the absolute delay of a slow-light pulse in an optically thick, power-broadened medium can be determined directly from static transmission spectra using only the measured EIT linewidth and off-resonant transmission.

Abstract: We show how slow-light pulse delays in realistic electromagnetically-induced-transparency (EIT) media can be determined directly from static transmission spectra. Using only the measured EIT linewidth and off-resonant transmission, the absolute delay of a slow-light pulse in an optically thick, power-broadened medium can be simply and accurately determined, while capturing more complex optical pumping behavior.

Figures

FIG. 1. Intensity spectrum with normalized contrast and background levels. Solid gray line is the data while the dashed dark line is a fit to a skew-Lorentzian. F: signal intensity above background away from EIT resonance the “floor” ; A: amplitude of EIT peak; C: difference between EIT peak and 100% transmission the “ceiling” . Parameters are normalized so that F+A+C=1. For data shown here, T=40 °C, in 25 torr N2 buffer gas, and total laser power 490 W.

FIG. 3. Measured EIT floor F, circles and peak transmission F+A, triangles level vs total laser intensity for experimental conditions described in the text. Uncertainties in the measured contrasts in this figure are predominantly systematic. Dotted curves are fits of a four-level model to the data with good agreement between fit parameters and experimental conditions see text .

FIG. 2. Schematic of the apparatus used for our EIT and slowlight measurements. An electro-optic modulator EOM frequency modulates the laser light to generate signal and reference field sidebands, while an acousto-optic modulator AOM regulates the light intensity. Reference and output signal fields are sent into photodetectors PD for EIT line shape and slow-light delay measurements. See text for details.

FIG. 4. Measured slow-light delays solid squares and predictions hollow squares based on accessible EIT line-shape parameters. Good agreement is shown over a large range of pulse delays and laser intensities.