Observation of Rb two-photon absorption directly excited by an erbium-fiber-laser-based optical frequency comb via spectral control

We present the observation of the Rb two-photon spectrum directly excited by an Er:fiber laser comb in a hot atomic vapor system. The influences of the atomic velocity distribution on direct and sequential two-photon transitions are discussed respectively and a combination of corresponding spectral control techniques is developed to eliminate the Doppler broadening background. A direct frequency comb spectroscopy is obtained with high-resolution two-photon transition lines. These techniques would benefit spectroscopy measurement in hot atomic vapor systems, and pave the way for establishing a high-stability optical frequency comb in 1.5 $\mu$m optical communication bands based on the Rb two-photon transitions.

Figure 1

(a) Spectral components of the comb involved in the TPT excitation (left) and the regular experiment scheme (right) for TPT excitation using ultrashort pulses. (b) The optical control chain for second harmonic generation of the Er:fiber optical comb. EDFA, erbium-doped fiber amplifier; SHG, second harmonic generation.

Figure 2

Two cases of pulses, (a) one excluding the comb lines around 780 nm and 776 nm and (b) the other incorporating these parts, and the corresponding contributions of the velocity groups of atoms to the TPTs. Part 1 and part 2 denote the contribution to the S-TPTs and D-TPTs, respectively.

Figure 3

Variation of the Doppler broadening background with the $f_{\mathrm{rep}}$ scanned around 144.555 MHz. The contribution of D-TPTs performs a constant background, while the contribution of the S-TPTs varies greatly with the $f_{\mathrm{rep}}$. The red lines show the measured background intensity. The blue bars demonstrate the calculated strengths and positions of the resonant S-TPT lines, and the dashed blue line shows the corresponding simulated contribution of these bars to the S-TPT background intensity.

Figure 4

TPT excitation scheme using spectrum split pulses. Each of the pulses is split into two subpulses with $>$778 nm and $<$778 nm spectral components, respectively, and a time delay $\tau$ is introduced between the first and second subpulses.

Figure 5

(a) Simulated and (b) experimental Doppler broadening background intensity versus the time delay $\tau$ between the two subpulses at different $f_{\mathrm{rep}}$ ($f_1=144.555$ MHz). A, B, and C correspond to the three $f_{\mathrm{rep}}$ cases as marked in Fig. 3.

Figure 6

The observed spectrum of ${}^{87}$Rb TPTs (red line) with $f_{\mathrm{rep}}$ scanned near 144.565 MHz. The blue and green bars show the calculated strengths and center positions of the TPT transition lines. The label “$i$-$j$” denotes the 5$S_{1/2}$ ($F=i$) to 5$D_{5/2}$ ($F=j$) TPT transition.