Figure 1

(a) Transmission electron microscope image of the cross section of the fiber (AIR-6-800) used in these experiments. Rubidium vapor is generated in the central core region, which then interacts with the light fields coupled into the core of the fiber. (b) TPA level scheme in ${}^{85}\mathrm{Rb}$ used for performing all-optical modulation. An atom in the ground $5S_{1/2}$ state can simultaneously absorb a photon each from the 780-nm control and 776-nm signal beams to make a resonant transition to the excited $5D_{5/2}$ state. The signal photon can only be absorbed if the control photon is also present. A small fraction of the excited atoms decay through the $6P_{3/2}$ level emitting blue fluorescence.Reuse & Permissions

Figure 2

The polarizations of the signal and control beams are made identical (circular) using polarization beam splitter (PBS) cubes and quarter ($\lambda /4$) wave plates, and the beams are then coupled counterpropagating into the fiber. A pickoff is used to monitor the transmission of the signal beam using a sensitive photodetector (PD). A strong (3 mW) off-resonant vapor generation beam is also coupled into the fiber to generate the desired atomic density and optical depth. The optical depth is monitored during the experiment using a weak 795-nm beam scanning across the $D_1$ resonance of Rb. An electro-optic modulator (EOM) driven by a function or waveform generator is used to make square pulses from the signal beam for the pulsed measurement shown in Fig. 4. An acousto-optic modulator (AOM) driven by another function generator is used to ramp the power of the control beam as a sawtooth wave for the measurement shown in Fig. 5.Reuse & Permissions

Figure 3

Transmission (red dots) of the signal beam as it is scanned across two-photon resonance showing $>20\%$ absorption with 1.4 nW of total power in the fiber. The OD attained in the fiber is $\sim 200$. The solid black line shows a fit produced by summing five exponential line shapes corresponding to the accessible hyperfine states of $5D_{5/2}$ level, from which the transit time $\tau$ was determined to be 5.3 ns.Reuse & Permissions

Figure 4

Absorption (red dots) of the pulsed signal beam at two-photon resonance versus pulse width. Error bars indicate measurement noise (1 standard deviation). The experimental data agree very well with the theoretically predicted curve (dotted black line) for a transit time $\tau =5\mathrm{ns}$ of the Rb atoms across the fiber core and corroborates that the response time of the system is determined by the transit time.Reuse & Permissions

Figure 5

Transmission (red dots) of the signal beam at two-photon resonance with varying power in the control beam. Up to $\sim 3\mathrm{dB}$ attenuation in the signal beam is observed with less than 5-nW of control power which corresponds to an energy density less than one photon per (${\lambda }^2/2\pi$). The solid black line shows the theoretical prediction from numerical simulation of the nonlinear propagation equations.Reuse & Permissions