Efficient All-Optical Switching Using Slow Light within a Hollow Fiber

We demonstrate a fiber-optical switch that is activated at tiny energies corresponding to a few hundred optical photons per pulse. This is achieved by simultaneously confining both photons and a small laser-cooled ensemble of atoms inside the microscopic hollow core of a single-mode photonic-crystal fiber and using quantum optical techniques for generating slow light propagation and large nonlinear interaction between light beams.

Figure 1

(a) Schematic of the experimental setup. (b) Schematic detail of the fiber loading. Inside the fiber, the atoms are transversely trapped by an optical trap formed by a single beam guided by the fiber. After exiting the top end of the fiber the divergent beam creates a funnel potential guiding the atoms into the fiber core from a distance up to $\sim 100\mu \mathrm{m}$. The atomic cloud is moved into this region by a magnetic guide consisting of four current-carrying wires that fan out from the fiber tip. (c) Scanning electron microscope image of the fiber cross section. The diameter of the central hollow region is $7\mu \mathrm{m}$. (d) Transmission through the fiber loaded with $\sim 3000$ atoms as a function of probe detuning from the $(5S_{1/2},F=1)\to (5P_{1/2},F^{\prime }=2)$ transition. Red points show the absorption profile with the dipole trap on continuously. Comparison to the calculated profile based on the dipole trap parameters (solid red line) confirms that the atoms are loaded inside the fiber. Black points show the observed absorption profile when probe and dipole trap are modulated as shown in (e). Here, a homogeneous broadening is caused solely by the large optical depth of the system.

Figure 2

(a) Four-state atomic system for nonlinear saturation based on incoherent population transfer. (b) Transmission of the probe as a function of the number of incident pump photons with exponential fit. 50% reduction of probe transmission is achieved with $\sim 300$ pump photons.

Figure 3

(a) Three-level $\Lambda$-scheme used for EIT in the mesoscopic atomic medium. (b) Both probe and control field are broken into a set of $\sim 100$ synchronized pulses sent through the fiber during the off-times of the dipole trap. The probe pulses have Gaussian envelopes with rms width $t_p\sim 150\mathrm{ns}$. (c) Probe transmission through the fiber without (black data points) and with (red data points) control field (each data point corresponds to 100 individual pulses). Solid lines are fits of Eq. (2). (d) Individual probe pulse shape and delay. The EIT pulse (red) is delayed by $\sim 100\mathrm{ns}$ compared to the reference pulse (black), for control pulses containing $\sim 3500$ photons. Data points represent the average number of photons detected in a 30 ns time bin. (e) Observed transmission of the probe pulses on resonance as a function of average number of photons in the $1\mu \mathrm{s}$ control field pulse and the prediction (gray line) based on Eq. (2).

Figure 4

(a) Four-level system for nonlinear optical switching. (b) Timing sequences for probe, control, and switch fields. (c) Probe transmission through the fiber without (red) and with (blue) the switch field present. Solid lines are fits of Eq. (2). (d) Observed transmission versus average number of switch photons per pulse. The solid gray line is the prediction based on Eq. (2). The transmission is normalized to the EIT transmission in the absence of the switch photons. (e) Truth table of the switch, showing the detected photons in the output port of the switch system as a function of presence (denoted by 1) or absence (denoted by 0) of the probe and switch pulses. Data are presented for probe pulses containing on average $\sim 2$ photons and $\sim 1/e$ attenuation of transmission in the presence of the switch photons.