of the revived probe pulse is reduced compared to that of the pulse
in Fig. 2b. Figure 2d shows the measured transmission for a series of
pulses as a function of their storage time in the atom cloud. The data
are consistent with an exponential decay with a 1/e decay time of
0.9 ms, comparable to the calculated mean free time of 0.5 ms
between elastic collisions in the atom cloud with a density of
11 mm-3
. Further studies of the decoherence mechanisms are
planned but are beyond the scope of this Letter.
We have veri®ed experimentally that the probe pulse is regenerated
through stimulated rather than spontaneous emission. To do
this, we prepared all atoms in state |2i and subsequently turned on
the coupling laser alone. The coupling laser was completely
absorbed for tens of microseconds without generating any signal
in the probe PMT.
In Fig. 3a±c, we show three PMT signal traces recorded under
similar conditions except that we vary the intensity, Ic2, of the
coupling laser when it is turned back on. When Ic2 is larger than the
original coupling intensity, Ic1, the amplitude of the revived probe
pulse increases and its temporal width decreases (Fig. 3a). For
Ic2 , Ic1, the opposite occurs (Fig. 3c). These results support our
physical picture of the process. The stored atomic coherence dictates
the ratio of the Rabi frequencies of the coupling and revived probe
®elds, as well as the spatial width of the regenerated pulse. In Fig. 3d
we show that with a large Ic2, the peak intensity of the revived probe
pulse exceeds that of the original input pulse by 40%.
of the revived probe pulse is reduced compared to that of the pulse
in Fig. 2b. Figure 2d shows the measured transmission for a series of
pulses as a function of their storage time in the atom cloud. The data
are consistent with an exponential decay with a 1/e decay time of
0.9 ms, comparable to the calculated mean free time of 0.5 ms
between elastic collisions in the atom cloud with a density of
11 mm-3
. Further studies of the decoherence mechanisms are
planned but are beyond the scope of this Letter.
We have veri®ed experimentally that the probe pulse is regenerated
through stimulated rather than spontaneous emission. To do
this, we prepared all atoms in state |2i and subsequently turned on
the coupling laser alone. The coupling laser was completely
absorbed for tens of microseconds without generating any signal
in the probe PMT.
In Fig. 3a±c, we show three PMT signal traces recorded under
similar conditions except that we vary the intensity, Ic2, of the
coupling laser when it is turned back on. When Ic2 is larger than the
original coupling intensity, Ic1, the amplitude of the revived probe
pulse increases and its temporal width decreases (Fig. 3a). For
Ic2 , Ic1, the opposite occurs (Fig. 3c). These results support our
physical picture of the process. The stored atomic coherence dictates
the ratio of the Rabi frequencies of the coupling and revived probe
®elds, as well as the spatial width of the regenerated pulse. In Fig. 3d
we show that with a large Ic2, the peak intensity of the revived probe
pulse exceeds that of the original input pulse by 40%.
การแปล กรุณารอสักครู่..
