Slowlight: Home M. Zentile, L. Weller, P. Siddons, Charles Adams & Ifan Hughes.
The ability of slow-light media to dynamically control the propagation speed and polarisation state of light makes slow light a useful tool for quantum information processing and interferometry. Here in Durham we are mainly interested in the propagation of light through atomic media with a linear response to electric field. Such systems have a large frequency range (tens of GHz) over which slow-light effects occur, compared to nonlinear media which have sub-MHz bandwidth. The GHz bandwidth available to us means that nano or even picosecond optical pulses can be transmitted with low attenuation/distortion. Some applications of slow light include:
  • Tunable pulse delay for optical information processing and quantum computing operations.
  • Polarisation switching of narrow and broadband pulses.
  • Coherent state preparation and entanglement in atomic ensembles.
  • The slow-light effect also has applications which utilise continuous-wave light:
  • Interferometers used to measure electric and magnetic fields with high spectral sensitivity.
  • Off-resonant laser locking with a dynamically tunable lock point over >10 GHz frequency range.
  • Highly frequency dependent optical isolation and filtering.
    Schematic of the experimental apparatus (right). The output of an external cavity diode laser (ECDL) is split by a polarization beamsplitter (PBS), for the c.w. (red) and pulsed (blue) experiments. Optical pulses are generated using a Pockels Cell (PoC). A 50:50 beam splitter is then used to produce a time delayed second pulse. The c.w. light is attenuated with a neutral density (ND) filter and a small fraction of the beam is used to perform sub-Doppler spectroscopy in a reference cell. Waveplates (λ/2 and λ/4) control the polarization of both beams before they pass through the experiment cell. The two orthogonal linear components of the pulse are collected on separate fast photodiodes (FPD), and the two components of the c.w. beam are collected on a differencing photodiode (DPD).
    Having a theoretical model which predicts the absorption and refractive index of a medium is useful, for example, in predicting the magnitude of pulse propagation effects. We have performed a comprehensive study of the Doppler-broadened absorption of a weak monochromatic probe beam in a thermal rubidium vapour cell on the D lines.

    Figure (right). Shows transmission plots for the comparison between experiment and theory, at temperatures of 16.5 °C (top), 25.0 °C (middle) and 36.6 °C (bottom). Red and black lines show measured and expected transmission, respectively. Below the main figure is a plot of the difference in transmission between theory and experiment for the 16.5 °C measurement.
    Beginning from the exact lineshape calculated for a two-level atom, a series of approximations to the electric susceptibility are made. These simplified functions facilitate direct comparison between absorption and dispersion and show that dispersion dominates the atom-light interaction far from resonance.

    Figure (left). Comparison between experiment and theory for the transmission of a weak probe beam through a vapour cell. (a) The thick solid red curve shows experimental data, whilst the dashed black curve shows the transmission calculated using the Voigt function. The Gaussian and Lorentzian approximations to the Voigt function are shown as solid black and dashed blue curves, respectively. (b) The difference in transmission between theoretical and measured data. The experimental data were obtained with red-detuned light, but plotted against Δ’ = -Δ. The origin of the detuning axis is from the 87Rb Fg = 2 → Fe = 1 transition.
    The Faraday effect is a magneto-optical phenomenon, where the rotation of the plane of polarization is proportional to the applied magnetic field in the direction of the beam of light. In slow-light the Faraday effect results in large dispersion and high transmission over tens of gigahertz. This large frequency range opens up the possibility of probing dynamics on a nanosecond timescale. In addition, we show large rotations of up to 15π rad for continuous-wave light.
    Optical pulse propagation in a slow-light medium. Pulse form at various temperatures for a pulse centred at zero detuning, Figure (right). The first observations of delayed optical pulses in Durham (September 2008). The pulse from a laser at wavelength 795 nm was sent through a Rb vapour cell. The output was then compared to a non-interacting reference pulse at various temperatures.
    Probe differencing signal produced by scanning the probe versus red detuning, Δ, from the D1 87Rb F = 2 → F’ = 1 transition. The dashed black curve shows the experimentally measured signal. The measured data is compared to theory (red).
    Broadband Faraday rotation in a slow-light medium. An input 1.5 ns pulse initially linearly polarized in the x-direction (red) is delayed by 0.6 ns with respect to a non-interacting reference pulse (black), in the absence of an applied magnetic field. The pulse is red-detuned from the weighted D1 transition centre by 10 GHz. For a temperature of 135 °C and a field of 80 G (green) or 230 G (blue) the pulse is rotated into the y-direction while retaining its linear polarization and intensity.
    Optical control of the Faraday effect could be used for all-optical single qubit rotations for photons and consequently opens new perspectives for all-optical quantum information processing.
    The measured rotation angles, for no optical control (squares), a red-detuned control field (circles), and a blue-detuned control field (triangles).
    A pulsed field will be used to drive population into the excited state in a time less than the excited state lifetime. The nanosecond switching time, combined with the Gigahertz bandwidth off-resonant. Faraday effect could permit rapid high-fidelity switching at low light levels.

    Content © Lee Weller, Durham University 2008