Research Of Optical Information Storage And Control Of Coherent Superpositions By Stimulated Raman Adiabatic Passage Technique

1. Optical signal storage and switching between two wavelengthsLight is the ideal carrier of quantum information. Storage and recovery of light signal in an atomic ensemble are exciting topics in the field of quantum optical information. Our technique is different from the conventional"light-storage"technique, which is based on the phenomenon of ultraslow light group velocity made by electromagnetically induced transparency (EIT). In our research, the first writing pulse (λ1 = 794.9842 nm) and the second write pulse (λ2 = 794.9698 nm) are turned on to write their optical information into the medium. The two pulses with the same back edge drive the atoms of ensemble into a maximum coherent superposition between the lower levels. That is, the strong fields write their optical information into the coherence superposition states of atomic levels by F-STIRAP technique. After a time interval, we turn on the reading pulse at 794.9842 nm, the recovered pulse at 794.9698 nm is released. Similarly, we turn on the reading pulse at 794.9698 nm, the revived pulse at 794.9698 nm is released.An energy levels diagram of the experiment is shown in Fig.1(a). Hyperfine levels |5² S_1/2,F= 1,M_F=-1,0>,showed as state |1>, and levels |5² S_1/2,F= 2,M_F=+1,+2>,showed as state |3>, serve as the lower states. The upper state is |5² P_1/2,F'= 1,M_F'=0,+1>,showed as level |2>. The first writing pulse with Rabi frequencyΩ₁ (λ₁ =794.9842nm) couples the ground state |3> and the excited state |2> with a frequency detuning of ? s. It is right circularly polarized. The power of the first writing pulse is 6mw, with duration 170ns, which can drive all populations into state |1> before the second writing Rabi frequency of the first writing and the second writing pulses, respectively. ? 3and ? 4are the Rabi frequency of the reading pulse or the restored pulse, respectively.(b)Schematic of the experimental setup. ECDL1 P1P2P3ECDL2I1I2L 1L2L 3L4AOM1AOM22λ4PλB S1Rb4λL 6L5Magnetic shieldPD1PD2 PBS2(b)?p(52 P1/2,F'=1,MF'=0,+1)(52 S1/2,F=2,MF=+1,+2)(52 S1/2,F=1,MF=?1,0)?4?2? 1?3?s|1>|2>|3>(a) pulse is applied. The second writing pulse with Rabi frequencyΩ₂(λ₂ =794.9698nm) couples the lower ground state |1> and the excited state |2> with a frequency detuning ofΔ_p. It is left circularly polarized. The power of the second writing pulse is 6.7mw, with duration 30ns. The first writing and the second writing pulses with the same back edge to store their optical information into the maximum coherent superpositions between lower hyperfine levels. The reading pulse turns to the |3>→|2> transition with Rabi frequencyΩ₃(λ₁ =794.9842nm), and read out the restored optical pulse with Rabi frequencyΩ₄(λ₂ =794.9698nm). The time delay between the end of the writing pulses and the beginning of the reading pulse is 80ns, as shown in Fig.2. The duration of the reading pulse is 30ns. Similarly, the reading laser can turn to the |1>→|2> transition with Rabi frequencyΩ₄, and obtain the released pulse with Rabi frequencyΩ₃.The experimental arrangement is shown in Fig.1(b). Both ECDL1 and ECDL2 are the DL-100 external-cavity diode lasers with linearly polarized output beams. The acousto-optic modulator AOM1, with the frequency shift 200MHZ, driven by a pulse generator, is used to switch on and off of the ECDL1 laser to generate the first writing pulse or the first writing and reading pulses. The polarization of the first writing pulse beam is rotated by 900 after passing throughλ/2 wave plate. The AOM2 is used to turn on and off ECDL2 laser to generate the second writing pulse or the second writing and reading pulse. The focus length of L1, L2, L3, and L4 is 5 cm. The two beams are combined with a polarizing beam splitter. They propagate collinearly through aλ/4 wave plate, which results in opposite circular polarization of the two beams. Then the two beams are focused by a lens (focus length 30 cm) into the atomic Rb vapor cell which is 3.5 cm long, 2.5 cm in diameter. The temperature of the Rb vapor cell was set to 85 ⁹80C, corresponding to atomic density of 10¹²cm^-3. After the cell, the beams with opposite circular polarization can be separated by another polarizing beam splitter after they pass through anotherλ/4 wave plate and become linearly polarized with orthogonal polarization, respectively. The first writing pulse, the second writing pulse, the reading pulse and the restored pulse are detected by PD1 and PD2, respectively.The time sequences of the light storage and release processes are shown in Fig.2. Fig.2(a) gives the experimental demonstration of the first writing pulse with Rabi frequencyΩ₁ (λ₁ =794.9842nm), the second writing pulse with Rabi frequencyΩ₂(λ₂ =794.9698nm) and the reading pulse with Rabi frequencyΩ₃(λ₁ =794.9842nm) before the Rb cell. The duration of the first writing pulseΩ₁ , the second writing pulseΩ₂ are 170 ns and 30 ns, respectively. The first writing pulse and the second writing pulse with the same edge prepare the maximum coherence between hyperfine levels. After a time interval of 80 ns, the reading pulse at 794.9842 nm is launched to release the stored optical pulse at 794.9698 nm, as shown in Fig.2(b). The duration of the reading pulse is 30 ns. Similarly, we can also turn on the reading field Ω₄(λ₂ =794.9698nm), as shown in Fig.2(c). Then the recovered pulseΩ₃(λ₁ =794.9842nm) is released, as shown in Fig.2(d). We also studied the storage time of revived probe pulses at 794.9698 nm and 794.9842 nm. Fig.3 shows the intensity of recovered pulse versus the storage time. An exponential degradation of the intensity due to the relaxation of the coherence between hyperfine levels ( |5² S_1/2,F= 1,M_F=-1,0> and levels |5² S_1/2,F= 2,M_F=+1,+2>) is clearly seen. The dots are the experimental data points from the released pulse at 794.9698 nm by the reading pulse at 794.9842 nm. And the triangle are the experimental points from the released pulse at 794.9842 nm by the reading pulse at 794.9698 nm. It is worth to note that the amplitude of the revived pulse at 794.9842nm has a higher amplitude than that of the revived pulse at 794.9698 nm. We believe that this phenomenon is caused by the population distribution of the hyperfine levels in the system. In the experiment, the writing pulses are stored in the atomic medium for more than 360 ns before they are read out by the reading pulse.In summary, we experimentally demonstrated that the optical signal can be stored into and controllably released from the atomic medium at 794.9698 nm by the reading pulse at 794.9842 nm or released at 794.9842 nm by the reading pulse at 794.9698 nm in a three-levelΛ-type atomic system. Because the maximum coherence are prepared between the lower levels, we can obtain the strong restored signal. We also studied the storage time of revived pulses at 794.9698 nm and 794.9842 nm. Such controlled release of stored optical pulses may extend the capabilities of the optical information storage technique, and can have applications in multichannels all-optical switching, optical information, optical networking and image storage systems.2. Optical switching and frequency conversion by Stimulated Raman Adiabatic PassageThe essence of the optical switching and frequency conversion is that we can control the switching and frequency conversion of Probe pulse by changing the duration of pump pulse. The Probe pulse is switching off when it passes through the medium, as shown in Fig.4(a). In this case, we only turn on the Stokes pulse, which wavelength isλ₂, with no Pump pulse. After a time interval, the probe pulse enters into the Rb vapor cell and is completely absorbed. So we say that the probe pulse is switched off. A short pump pulse is turned on with its back edge being the same with the Stokes pulse to create a coherence between lower levels, as shown in Fig.4(b). Then the probe pulse withλ₁ is changed to the pulse withλ₂ after it passing through the Rb vapor cell by a Raman process. Therefore an optical frequency conversion of probe pulse is performed. If the duration of the pump pulse is increased, as shown in Fig.4(c), the pump pulse is varied to make its front part overlaps with the back part of the Stokes pulse. In this case, the probe pulse has negligible absorption after it passing through the Rb vapor cell. So we say that the probe pulse is switched on.The experiment is performed in 87 Rb vapor which was maintained in a cell with the temperature of 110 ¹20(?), corresponding to atomic density of 10¹³cm^-3. The Rb cell is 3.5 cm long, 2.5 cm in diameter. The energy levels configuration and the schematic diagram of the experimental setup is shown in Fig.5. The Stokes pulse with Rabi frequencyΩ_s(λ 2=794.9842nm) drives the |5² S_1/2,F= 2>→|52P1/2,F'=1> transition resonantly. It is the right circularly polarized with power 6.8 mw. The pump pulse with Rabi frequencyΩ_p(λ₁ =794.9698nm) and probe pulse with Rabi frequencyΩ_probe(λ₁ =794.9698nm) drives the |5² S_1/2,F= 1>→|5²P_1/2,F'=1> transition resonantly. They are the left circularly polarized with the same power 4.9 mw. The prepared coherence and the distribution of coherence population between the lower levels are lied on the Stokes and pump pulses. The laser at 794.9842 nm is generated by the external-cavity diode laser ECDL1. The acousto-optic modulator AOM1 is used to switch on and off this beam to generate the Stokes pulse. The laser at 794.9698 nm is generated by ECDL2. The AOM2, driven by a pulse generator, is used to switch on and off the laser to generate the pump and probe pulses. The polarization of the pump and probe pulses is rotated by 900 after passing throughλ/2 wave plate. The two beams with orthogonal polarization are overlapped by the polarizing beam splitter PBS1, then converted to circular polarization by aλ/4 wave plate. Then the two beams focus into the Rb atomic vapor cell. After the cell, the two beams can be separated by PBS2 after they pass through anotherλ/4wave plate. The Stokes, pump, and probe pulses are detected by the fast photodiode PD1 and PD2, respectively.To predict the experimental phenomenon, we have performed numerical simulations for the optical switching and frequency conversion. The time evolution of the laser pulses is shown in Fig.6. The upper-side of the Fig.6(a), (b), (c) shows the time evolution of laser pulses before the medium. And the lower-side of the Fig.6(a), (b), (c) displays the laser pulses after passing through the medium. The dash curves correspond to the laser pulses at wavelengthλ 2 while the solid curves mark the laser pulses at wavelengthλ₁ . The case with no pump pulse is shown in Fig.6(a). The Stokes pulse with 150 ns is first turned on and most of population are prepared in the state |1>. After a time interval, the probe pulse with a temporal length of 20 ns enters into the Rb vapor cell and is completely absorbed. So we say that the probe pulse is switched off. The pump pulse with 20 ns, is turned on with its back edge being the same with the Stokes pulse, which is described as F-STIRAP technique, to create a coherence between levels |3> and |1>, as shown in Fig.6(b). Because the power of the Stokes is stronger than the pump, a weak coherence is prepared between the two lower levels. So the probe pulse withλ₁ is changed to the pulse withλ 2 after it passing through the medium. Therefore an optical frequency conversion of probe pulse is performed. If the duration of the pump pulse is increased to be 110 ns, as shown in Fig.6(c), the pump pulse is varied to make its front part overlaps with the back part of the Stokes pulse, as required by the STIRAP process. In this case, most of the population is transferred to level |3>, and the probe pulse can go through the medium with a small absorption, and the wavelength of probe pulse is not changed. So we say that the probe pulse is switched on. The experimental results of the optical switching and frequency conversion processes are shown in Fig.7. The direct comparison between the Fig.7 and Fig.6 shows that the experiments are in good agreement with numerical simulations. The upper-side frame of the Fig.7(a), (b), (c) gives the experiment demonstration of all laser pulses before the Rb cell. The lower-side frame of Fig.7(a), (b), (c) displays the laser pulses after passing through the Rb vapor cell. The beam atλ 2 is described by the thick solid curves and the one atλ₁ is described by the thin solid curves. The experimental result of the probe being switched off is shown in Fig.7(a). The duration of Stokes pulse atλ 2, the pump pulse atλ₁ , and the probe pulse atλ₁ are 150 ns, 0 ns and 20 ns, respectively. The probe fields experiences a strong absorption after passing though the cell. One can see that the Stokes pulse, which transfers most of the population to the state |1>, is absorbed at the front edge of the pulse. When the pump pulse with 20 ns duration is turned on, as shown in Fig.7(b), the optical frequency conversion is achieved. The probe pulse atλ₁ becomes the pulse atλ 2 after passed through the Rb vapor cell. The frequency conversion efficiency is about 0.4 by the technique. The probe pulse is switched on as long as the duration of pump pulse is varied to 110 ns, as shown in Fig.7(c). In Fig.7(b) and Fig.7(c), the peak at the back of the Stokes pulse arises from stimulated Raman resonance scattering of the two-photon process.In summary, this part reported the experimental results of optical switching and frequency conversion based on STIRAP technique in high-temperature Rb vapor cell. We experimentally demonstrated that by changing the duration of the pump pulse, we could achieve the optical switching and the optical frequency conversion. The propagation of the laser pulses in the medium was numerically simulated. We noticed that the numerical simulations had good agreement with the experimental results. Our work is expected to be a good candidate for optical processing in the future optical networks, optical computing, optical telecommunications, and so on.3. The storage and switching of multi-optical signals among three channels based on F-STIRAP technique in tripod-type atomic systemThe most experimental and theoretical works on "light storage" focus on one light signal storage and retrieval. In this letter, we report an experimental result of the storage and retrieval of multi-optical signals by F-STIRAP technique in a tripod-type four-level Rb atomic system as shown in Fig.8. In this work, instead of EIT, the F-STIRAP technique is used for optical storage. And, multi-optical signals storage and retrieval are realized in the atomic system. The writing pulsesΩ₀(λ₁ = 794.9842 nm; left circularly polarized lightσ^-),Ω₁ (λ 2= 794.9698 nm; linearly polarized lightÏ€), andΩ₂(λ₁ = 794.9842 nm; right circularly polarized lightσ⁺) with the same back edge drive the atoms into the optimization coherent superposition among the atomic ground levels and store their optical information in the maximum coherent superposition of the collective atomic medium. After a time interval, we turn on the reading pulseΩ₄(λ 2= 794.9698 nm; linearly polarized lightÏ€) , the two pulsesΩ₃ andΩ₅ are simultaneously released. Instead ofΩ₄, ifΩ₃(λ₁ = 794.9842 nm; left circularly polarized lightσ^-) is turned on as the reading pulse, the two pulsesΩ₄ andΩ₅ are simultaneously read out. The two stored pulsesΩ₃ andΩ₄ can be read out by turningΩ₅ (λ₁ = 794.9842 nm; right circularly polarized lightσ⁺) as the reading pulse.The energy levels diagram of the tripod-type Rb atomic system and the schematic diagram of the experimental setup are shown in Fig.8. The Rb cell is 3.5 cm long, 2.5 cm in diameter. The temperature of the Rb vapor cell was set to 88 ⁹80C, corresponding to the atomic density of 10^{11 1}0¹²cm^-3. The atomic levels |5₂ S_1/2,F= 2,M_F=-2,-1,0>, |5₂ S_1/2,F= 1,M_F=-1,0,+1>, and |5₂ S_1/2,F= 2,M_F=0,+1,+2>, are used as the three ground states in the tripod-type four-level system, which are labeled as |0>, |1>, and |2>, respectively. The atomic level 5₂ P_1/2,F'= 1,M_F'=-1,0,+1>, which is showed as level |3>, is used as the upper state in the tripod-type four-level system. The two beams of writing pulse with Rabi frequencyΩ₀ and the writing pulse with Rabi frequencyΩ₂ are obtained by one extended cavity diode laser (ECDL1), which generated the wavelength 794.9842 nm. The writing pulseΩ₀ and the reading pulseΩ₃ are left circularly polarized and tuned to |5₂ S_1/2,F= 2,M_F =-2,-1,0>→|5²P_1/2,F'=1,M_F'=?1,0,+1>the transition of 87 Rb. The writing pulseΩ₂ and the reading pulseΩ₅ are right circularly polarized and tuned to the |5₂ S_1/2,F= 2,M_F =0,+1,+2>→|5²P_1/2,F'=1,M_F'=-1,0,+1>transition of 87 Rb. The beam of writing pulseΩ₁ and the reading pulseΩ₄ is obtained by another extended cavity diode laser ECDL2, which generated the wavelength 794.9698 nm, to couple the atomic levels |5₂ S_1/2,F= 1,M_F=-1,0,+1>and the |5₂ P_1/2,F'= 1,M_F'=-1,0,+1>. The beam ofΩ₁ andΩ₄ is linearly polarized. The beam radiated from the ECDL1 is split into two beams by a beamsplitter. One beam passes through acousto-optic-modulator AOM1, which is used to switch on and off the laser to generate one pulseΩ₀ or two pulsesΩ₀ andΩ₃. Another beam passes through AOM2 to generate one pulseΩ₂ or two pulsesΩ₂ andΩ₅. The two beams become orthogonal polarized after one of them passed through aλ/2 wave plate. Then they are converted to the left circularly polarized and the right circularly polarized beams after passing through aλ/4 wave plate, respectively. The beam, which is generated by ECDL2, passes through AOM3 to generate one pulseΩ or two pulsesΩ andΩ₄. The focus length of L1, L2, L3, L4, L5 and L6 is 10 cm. Then the three beams are collimated and focused by the lens L7 (focus length 30 cm) into the atomic Rb atomic vapor cell to store their optical information in the coherence among atomic levels. Input peak powers for the three fields are about 8 mw, 6 mw and 15 mw, respectively. Coming out of the cell, the two circularly polarized beams pass through anotherλ/4 wave plate. Then they are detected by the fast photodiode PD1 and PD2, respectively. Another beam with linear polarization is detected by PD3.The experimental results of the multi-Optical signals storage and release processes are shown in Fig.9. Fig.9(a), (c) and (e) give the experimental pulse shapes recorded before the Rb cell. Fig.9(b), (d) and (f) describe the reading pulse and restored pulses after their propagation through the cell, corresponding the reading fieldsΩ₅,Ω₃, andΩ₄,respectively. TheΩ₂,Ω₀, andΩ are the writing pulses, respectively. The writing pulseΩ₂ with duration 400 ns is first turned on. After 200 ns, the writing pulseΩ₀ with a temporal length of 200 ns enters into the cell. Then the writing pulse Ω with duration 40 ns is turned on. The three writing fields are simultaneously and abruptly switched off and a maximal atomic coherence among the three ground levels is established. The optical information ofΩ₀,Ω andΩ₂ are stored in the coherence. Fig.9(a) and (b) show that, after 32 ns time delay following the writing pulses, the reading pulseΩ₅ with duration 40 ns is launched to release optical pulsesΩ₃ andΩ₄, which are the stored optical pulsesΩ₀ andΩ , respectively. In Fig.9(c) and (d), theΩ₃ with duration 40 ns being turned on as the reading pulse to simultaneously read out the optical informationΩ₄ andΩ₅, which are the stored optical informationΩ andΩ₂, respectively. In Fig.9(b),(d), it is worth to note that the peak power of recovered pulseΩ₃ andΩ₅ is stronger thanΩ₄. The first reason is that the beam ofΩ₀,Ω₃ and the beam ofΩ₂,Ω₅ are generated by one laser ECDL1. The created coherence using single laser system is better than using two independent ECDLs. The second reason is that, in this experiment, the populations interacted with the two beams ofΩ₀ andΩ₂ are more than the populations interacted with the three beams ofΩ₀,Ω andΩ₂. As shown in Fig.9(e) and (f), this time we turn on the pulseΩ₄ instead ofΩ₃ orΩ₅, so the pulsesΩ₃ andΩ₅ are read out. We emphasize that the restored pulses have the same frequency, polarization and propagation direction with the writing pulses.Fig.10 shows the peak intensities of revived pulses as a function of storage time, corresponding to Fig.9. At different delay times, we turn on the reading pulseΩ₅, then we plot the intensity of restored pulsesΩ₃ andΩ₄ versus the time delay, as shown in Fig.10(a). The hollow dots are the experimental data points of released pulseΩ₃. And the hollow triangles are the experimental data points of revived pulseΩ₄. One can see that both power of the revived pulsesΩ₃ andΩ₄ decrease following the storage time due to the atomic decoherence. The two curves show the signals attenuate as approximate functions of exponential distributions. Fig.10(b) describes the peak powers of pulsesΩ₅ andΩ₄ versus storage time, which are read out by turning on the reading pulseΩ₃ at different delay times. The squares are the experimental data points of released pulseΩ₅. The hollow triangles are the experimental data points of restored pulseΩ₄ versus storage time. The two curves are the exponential decay fit of the experimental data, respectively. WhenΩ₄ as the reading pulse, the recovered intensity of pulsesΩ₃ andΩ₅ versus the storage time are shown in Fig.10(c). The squares are the experimental data points of released pulseΩ₅. The hollow dots describe the experimental date points ofΩ₃ versus storage time. The two curves are the exponential decay fit of the experimental data. And the two curves are nearly overlapped and have approximately the same decline slope. Our Rb vapor cell doesn't contain buffer gas. So there is a large decoherence, which results the short storage time in our experiment.In summary, this part experimentally demonstrated that multi-optical signals can be stored and recovered by F-STIRAP technique in a tripod-type four-level Rb atomic system. The optical pulses stored can be controllably released into three different channels. In this experiment, the three writing fields are strong fields to write their optical information in the coherence among the ground atomic levels. After a time interval, we can turn on one of the three channels as the reading field to simultaneously read out another two stored pulses. Multi-optical signals stored being simultaneously retrieved is an extension of the optical storage techniques in atomic assembles. The restored pulses have the same frequency, polarization and propagation direction with the writing pulses. We also observed the intensity of the recovered optical signals versus storage time. Based on the mechanism of multi-optical signals simultaneously storage and retrieval and controllable release among three different channels of store light, one can expend the capabilities of optical information "storage" technique, effective multichannel optical switch, and image storage system.