## Abstract

We propose an efficient scheme for the robust and controlled generation of beating signals in a sample of stationary atoms driven into the tripod configuration. This scheme relies on an asymmetric procedure of light storage and retrieval where the two classical coupling fields have equal detunings in the storage stage but opposite detunings in the retrieval stage. A quantum probe field, incident upon such an atomic sample, is first transformed into two spin coherence wave-packets and then retrieved with two optical components characterized by different time-dependent phases. Therefore the retrieved quantum probe field exhibits a series of maxima and minima (beating signals) in intensity due to the alternative constructive and destructive interference. This interesting phenomenon involves in fact the coherent manipulation of two dark-state polaritons and may be explored to achieve the fast quantum limited measurement.

© 2011 OSA

## 1. Introduction

Quantum information storage and quantum state engineering are two key elements for the practical implementation of quantum information processing. Photons are believed to be the most fast and robust carriers of quantum information while atomic ensembles may provide the best units for reversible storage and nonlinear processing of weak light signals [1]. To efficiently control the propagation and interaction of weak light signals, there has been a growing interest in the research of laser induced atomic coherence, which is the basis of many interesting phenomena such as electromagnetically induced transparency (EIT) [2] and stimulated Raman adiabatic passage (STIRAP) [3]. Using the EIT technique, one can either suppress resonant absorption to achieve low-light-level nonlinear optics [4–7] or manipulate light group velocity to attain quantum information storage [8–11]. Typically, the three-level Lambda system with a non-degenerate dark state is adopted to demonstrate the quantum memory of photonic states by transforming a slowly propagating light field into a stationary wave-packet of spin coherence and vice versa [12]. This dynamic process and the associated slow light propagation are usually interpreted in terms of dark-state polariton (DSP) [13] defined as a joint atom+field excitation. To store a photonic qubit with two basis states, however, we have to consider four-level systems driven into, e.g. the tripod configuration exhibiting multiple degenerate dark states and allowing for two DSP modes [14, 15]. In the tripod system, a pair of light fields can be simultaneously slowed down, respectively stored into two different wave-packets of spin coherence, and then retrieved at the same time or with a controlled time delay [16–18]. The tripod system has also been explored to achieve the enhanced cross-phase modulation between two weak light fields with the ultimate goal to build conditional quantum phase gates for manipulating photonic polarization states [19–22]. Note, in particular, that beating signals due to the interference between two weak probe fields (two DSP modes) have been observed recently in the tripod system as a manifestation of the preserved phase information during light storage [23]. Beating signals can also be observed in the Lambda system when a weak probe field (a DSP mode) interferes with a strong coupling field during light retrieval in the presence of a magnetic field [24, 25]. Last but not least, Heinze et al. demonstrate that the total intensity of a retrieved spatial image may oscillate against the storage time, which is also identified as a beating of DSP modes [26]. Such interferometric beating is claimed to be important in that it has potential applications in the fast quantum limited measurement.

In this paper, we present an alternative scheme for the robust and controlled generation of beating signals via the dynamic EIT technique in a tripod system of stationary atoms. This tripod atomic system is driven by a weak (quantum) probe field and two strong (classical) coupling fields and therefore is different from that in ref. [23] interacting with two weak probe fields and a strong coupling field. Our numerical calculations show that beating signals can be reliably generated and flexibly controlled in an *asymmetric* procedure of light storage and retrieval. To be more specific, we first store the probe field into two different wave-packets of spin coherence by switching off the two coupling fields with *equal* detunings, and then retrieve it after a short storage time by switching on the two coupling fields with *unequal* detunings instead. In this case, the only retrieved probe field consists of two components characterized by different time-dependent phases so that beating signals (a series of maxima and minima in intensity) arise due to the alternate constructive and destructive interference. The generation of beating signals can be well understood in the polariton picture where a pair of DSP modes are simultaneously excited with their structures totally determined by amplitudes, phases, and detunings of the two coupling fields. Experimental examination of such beating signals may be used to acquire the frequency difference, the relative phase, and their stabilities of two classical coupling fields or to achieve the fast quantum limited measurement of magnetic field amplitudes and atomic transition frequencies between ground state sublevels.

## 2. Model and equations

We consider here a medium of length *L* consisting of an ensemble of stationary atoms coherently driven into the four-level tripod configuration (see Fig. 1). The first dipole-allowed transition |*a*〉 ↔ |*e*〉 is coupled by one classical field with frequency *ω _{c}*

_{1}and amplitude

**E**

_{c}_{1}and the second dipole-allowed transition |

*b*〉 ↔ |

*e*〉 is coupled by another classical field with frequency

*ω*

_{c}_{2}and amplitude

**E**

_{c}_{2}. The third dipole-allowed transition |

*c*〉 ↔ |

*e*〉, however, is probed by a weak quantum field described by

*ε*is the polarization vector,

_{p}*ω*the carrier frequency,

_{p}*V*the quantization volume, and

*E*(

_{p}*z,t*) the slowly-varying dimensionless operator.

The tripod-type stationary atoms, on the other hand, are described by the collective atomic operators

*N*denotes the atomic number in a small volume

_{z}*V*centered at

_{z}*z*while ${\widehat{\sigma}}_{\mu \nu}^{j}\hspace{0.17em}\left(z,\hspace{0.17em}t\right)$ is defined as the atomic flip operator |

*μ*〉 |

_{j}*ν*〉 with {

_{j}*μ*,

*ν*} ∈ {

*a*,

*b*,

*c*,

*e*}. For simplicity and convenience, we further introduce the slowly-varying operators

*σ*(

_{ce}*z, t*),

*σ*(

_{cb}*z, t*), and

*σ*(

_{ca}*z, t*) for three relevant coherence terms

The weak probe assumption allows us to perturbatively solve the original Heisenberg-Langevin equations to the first order in the probe field so that we don’t need to consider the dynamic evolutions of operators *$\widehat{\sigma}$ _{ab}*(

*z,t*),

*$\widehat{\sigma}$*(

_{ae}*z,t*),

*$\widehat{\sigma}$*(

_{be}*z,t*),

*$\widehat{\sigma}$*(

_{aa}*z,t*),

*$\widehat{\sigma}$*(

_{bb}*z,t*),

*$\widehat{\sigma}$*(

_{cc}*z,t*), and

*$\widehat{\sigma}$*(

_{ee}*z,t*). The adiabatic control assumption, on the other hand, is the solid foundation for us to safely neglect the

*δ*-correlated Langevin noise operators

*F*,

_{ca}*F*, and

_{cb}*F*in Eqs. (4). In Eqs. (4),

_{ce}*γ*,

_{ca}*γ*, and

_{cb}*γ*represent the coherence decay rates of operators

_{ce}*σ*,

_{ca}*σ*, and

_{cb}*σ*while Δ

_{ce}*=*

_{p}*ω*–

_{p}*ω*, Δ

_{ec}

_{c}_{1}=

*ω*

_{c}_{1}–

*ω*, and Δ

_{ea}

_{c}_{2}=

*ω*

_{c}_{2}–

*ω*are frequency detunings of the three quantum or classical fields. In addition, Ω

_{eb}

_{c}_{1}=

**E**

_{c}_{1}·

**d**

*/2*

_{ea}*h̄*and Ω

_{c}_{2}=

**E**

_{c}_{2}·

**d**

_{eb}*/*2

*h̄*are complex Rabi frequencies of the first and second (classical) coupling fields while ${g}_{p}\hspace{0.17em}=\hspace{0.17em}\sqrt{{\omega}_{p}/2\overline{h}{\varepsilon}_{0}V}{\varepsilon}_{p}\hspace{0.17em}\cdot \hspace{0.17em}{\mathbf{\text{d}}}_{\mathit{\text{ec}}}$ is the real coupling constant of the (quantum) probe field with

**d**

*being the dipole matrix element on transition |*

_{μν}*μ*〉 ↔ |

*ν*〉. To be more specific, we may further set

*G*

_{c}_{1}(

*G*

_{c}_{2}) and Φ

_{c}_{1}(Φ

_{c}_{2}) as the modulus and the argument of the complex Rabi frequency Ω

_{c}_{1}(Ω

_{c}_{2}), respectively.

Solving Eqs. (4) in the steady state, we can easily examine the absorption and dispersion spectra, proportional respectively to Im[*σ _{ce}*(Δ

*)] and Re[*

_{p}*σ*(Δ

_{ce}*)], of a continuous-wave probe. To study the dynamic evolution of the dimensionless operator*

_{p}*E*(

_{p}*z,t*), however, we also need the wave propagation equation

*N*refers to the total number of active atoms in the quantization volume

*V*of the weak probe field while

*c*is defined as the light speed in vacuum. Eq. (5) coupled with Eqs. (4) will be numerically solved in the next section to examine an interesting light propagation dynamics concerning the generation and control of quantum limited beating signals imposed on a probe field.

## 3. Results and discussions

In what follows, we show by numerical calculations how to generate and control robust beating signals in an *asymmetric* procedure of light storage and retrieval via the dynamic EIT technique. Without loss of generality, we will set *γ* = *γ _{ce}* as the frequency unit,
${t}_{0}\hspace{0.17em}=\hspace{0.17em}{\gamma}_{\mathit{\text{ce}}}^{-1}$ as the time unit, and
${l}_{a}\hspace{0.17em}=\hspace{0.17em}c{\gamma}_{\mathit{\text{ce}}}/{g}_{p}^{2}N$ (the absorption length in the absence of EIT) as the position unit. As we can see from Fig. 2, a quantum field

*E*(

_{p}*z,t*) goes slowly into the atomic sample at the velocity

*υ*= 0.25

_{g}*l*/

_{a}*t*

_{0}and is totally transformed into stationary spin coherences

*σ*(

_{ca}*z,t*) and

*σ*(

_{cb}*z,t*) at the sample center when the two coupling fields with Δ

_{c}_{1}= Δ

_{c}_{2}= 0.0

*γ*are simultaneously turned off at

*t*= 48.0

*t*

_{0}. After a short storage time of Δ

*t*= 24.0

*t*

_{0}, we switch on the two coupling fields to retrieve the quantum field from

*σ*(

_{ca}*z,t*) and

*σ*(

_{cb}*z,t*) at the same time but with opposite detunings, i.e. Δ

_{c}_{1}= −Δ

_{c}_{2}= 0.02

*γ*in (a, b), Δ

_{c}_{1}= −Δ

_{c}_{2}= 0.05

*γ*in (c, d), and Δ

_{c}_{1}= −Δ

_{c}_{2}= 0.08

*γ*in (e, f). As expected, the retrieved quantum field appears and vanishes in a periodic pattern when it propagates inside the atomic sample once again at the velocity

*υ*= 0.25

_{g}*l*/

_{a}*t*

_{0}. That is, the retrieved quantum field exhibits a series of maxima and minima (beating signals) in its average intensity, which oscillate at the frequency of Δ

*ω*= Δ

_{beat}

_{c}_{1}− Δ

_{c}_{2}when |Δ

_{c}_{1}| = |Δ

_{c}_{2}| are large enough [see Fig. 2(e, f)].

The interesting beating signals in Fig. 2 can be qualitatively understood in terms of DSP as a coherent mixture of quantum field and spin coherence. Adopting the standard procedure [13], we may adiabatically transform Eqs. (4) and Eqs. (5) into

*z,t*) = Ψ

*(*

_{a}*z,t*) + Ψ

*(*

_{b}*z,t*) with

*z,t*) originates from the pure quantum field excitation

*E*(

_{p}*z,t*) or

*z,t*) originates from the pure atomic coherence excitation $\sqrt{N/2}\left[{\sigma}_{\mathit{\text{ca}}}\hspace{0.17em}\left(z,t\right)\hspace{0.17em}+\hspace{0.17em}{\sigma}_{\mathit{\text{cb}}}\hspace{0.17em}\left(z,t\right)\right]$. In deriving Eqs. (7) and Eqs. (8), we have set $\text{tan}\theta \hspace{0.17em}=\hspace{0.17em}g\sqrt{N/\left({G}_{c1}^{2}\hspace{0.17em}+\hspace{0.17em}{G}_{c2}^{2}\right)}$ and

*G*

_{c}_{1}=

*G*

_{c}_{2}for simplicity so that both DSP modes Ψ

*(*

_{a}*z,t*) and Ψ

*(*

_{b}*z,t*) can attain the same propagating velocity

*υ*=

_{g}*c*cos

^{2}

*θ*at the retrieval stage.

Now we begin to explain the main results shown in Fig. 2 with Eqs. (7) and Eqs. (8) in the limit of Φ_{c}_{1} = Φ_{c}_{2} = 0.0. At *t* = 0.0*t*
_{0}, a fast probe field *E _{p}*(

*z,t*) enters the sample of stationary atoms and evolves into two slowly-moving DSP modes described by Eqs. (7) through its two-photon resonant interaction with both coupling fields Ω

_{c}_{1}and Ω

_{c}_{2}. In this case, the field component in Ψ

*(*

_{a}*z,t*) and that in Ψ

*(*

_{b}*z,t*) have the same vanishing phase so that no beating signals are observed in the probe intensity before

*t*= 48.0

*t*

_{0}. At

*t*= 48.0

*t*

_{0}, when both coupling fields Ω

_{c}_{1}and Ω

_{c}_{2}are switched off, the two slowly-moving DSP modes described by Eqs. (7) turn into a pair of stationary atomic excitations $\sqrt{N/2}{\sigma}_{\mathit{\text{ca}}}\hspace{0.17em}\left(z,t\right)$and $\sqrt{N/2}{\sigma}_{\mathit{\text{cb}}}\hspace{0.17em}\left(z,t\right)$ limited by Δ

_{c}_{1}= Δ

_{c}_{2}= 0.0

*γ*. This is why the probe intensity is exactly zero during the period of

*t*= 48.0

*t*

_{0}∼ 72.0

*t*

_{0}. At

*t*= 72.0

*t*

_{0}, when the two coupling fields are switched on with Δ

_{c}_{1}≠ Δ

_{c}_{2}instead, the pair of stationary atomic excitations become the origin of two slowly-moving DSP modes described by Eqs. (8). In this case, the field component in Ψ

*(*

_{a}*z,t*) and that in Ψ

*(*

_{b}*z,t*) attain, respectively, time-dependent phases Δ

_{c}_{1}

*t*and Δ

_{c}_{2}

*t*so that they interfere with each other to produce a series of beating signals in the probe intensity after

*t*= 72.0

*t*

_{0}. At the sample exit, the two DSP modes described by Eqs. (8) finally turn into a pair of fast field components

*e*

^{iΔc1t}

*E*(

_{p}*z,t*)/2 and

*e*

^{iΔc2t}

*E*(

_{p}*z,t*)/2 with beating signals perfectly reserved. It is clear that the field components of both DSP modes experience little absorptive loss even if we set Δ

_{c}_{1}≠ Δ

_{c}_{2}after

*t*= 72.0

*t*

_{0}because they are well contained in two separate EIT windows located, respectively, at Δ

*= Δ*

_{p}

_{c}_{1}and Δ

*= Δ*

_{p}

_{c}_{2}[27].

To see how beating signals are formed, we plot in Fig. 3 the pair of field components at the sample exit by increasing the coupling detuning difference |Δ_{c}_{2} – Δ_{c}_{1}| in very small steps. As we can see there is only a single maximum when |Δ_{c}_{2} – Δ_{c}_{1}| < 0.003*γ* while a second maximum tends to arise when |Δ_{c}_{2} – Δ_{c}_{1}| > 0.003*γ*. In addition, a part of light energy is redistributed into a later time region and the destructive interference occurs between the former and later time regions when a wider EIT window is gradually split into two narrower ones by an absorption line [27]. From Eqs. (7) and Eqs. (8) we also can see that the two DSP modes are sensitive to arguments Φ_{c}_{1} and Φ_{c}_{2} of the two complex Rabi frequencies Ω_{c}_{1} and Ω_{c}_{2}. So one may simply modulate the relative phase ΔΦ = Φ_{c}_{2} – Φ_{c}_{1} to control the beating signals attained in an *asymmetric* procedure of light storage and retrieval, which is illustrated in Fig. 4. It is found that beating signals with ΔΦ = 0.0 (*π/*2) and beating signals with ΔΦ = *π* (3*π*/2) are exactly staggered by a half period, i.e. a maximum in the black-solid curves corresponds to a minimum in the red-dashed curves. In this case, it is *e ^{i}*

^{(Δc1t−Φc1)}

*E*(

_{p}*z,t*)/2 and

*e*(

^{i}^{Δc2t − Φc2)}

*E*(

_{p}*z,t*)/2 that describe the pair of field components at the sample exit.

We would like to note that the stationary atoms mentioned above can be either stray impurities doped in a solid materials or cold atoms confined in a magneto-optical trap (MOT). For the former, the Pr^{3+}: Y_{2}SiO_{4} crystal working at the cryogenic temperature [28, 29] should be a good candidate for attaining beating signals via the dynamic EIT technique. But the inhomogeneous broadening of absorption lines due to random crystalline fields should be suitably included into the coherence decay rates *γ _{ca}*,

*γ*, and

_{cb}*γ*. For the latter, we may choose the D1 line of cold

_{ce}^{87}Rb atoms to construct the tripod system with |

*a*〉, |

*b*〉, |

*c*〉, and |

*e*〉 referring to |5

^{2}

*S*

_{1/2},

*F*= 2,

*m*= +1〉, |5

_{F}^{2}

*S*

_{1/2},

*F*= 2,

*m*= −1〉, |5

_{F}^{2}

*S*

_{1/2},

*F*= 1,

*m*= +1〉, and |5

^{2}

*P*

_{1/2},

*F*′ = 1,

*m*= 0〉, respectively [30].

## 4. Conclusions

In summary, we have demonstrated by numerical calculations and analyzed in the polariton picture an efficient scheme for the dynamic generation and flexible control of beating signals via an *asymmetric* light storage and retrieval technique. In experiment, the frequency difference *ω _{c}*

_{2}−

*ω*

_{c}_{1}, the relative phase Φ

_{c}_{2}− Φ

_{c}_{1}, and their stabilities can be easily inferred from the beating signals imposed on a weak probe field. If the detuning difference |Δ

_{c}_{1}− Δ

_{c}_{2}| is induced by shifting hyperfine state sublevels with a magnetic field

**B**, we may have

*h̄*|Δ

_{c}_{1}− Δ

_{c}_{2}| = 2

*gμ*·

_{B}**B**with

*g*being the Lander factor and

*μ*the Bohr magneton. In this case such interferometric beating signals can be used to measure magnetic field amplitudes. If the two coupling fields have the identical frequency and differ only by

_{B}*σ*

^{+}and

*σ*

^{−}polarizations, however, we may have |Δ

_{c}_{1}− Δ

_{c}_{2}| = |

*ω*| so that atomic transition frequencies can be determined from such interferometric beating signals. In particular, when a nonclassical squeezed light is used, one can achieve the quantum limited measurement with a sub-shot noise precision because quantum properties are well conserved during light storage and retrieval [10, 11]. Moreover, this novel method for measuring atomic transition frequencies is much faster than the standard spectroscopy method because a light storage experiment can be implemented in a fraction of the time required for the acquisition of a complete EIT spectrum.

_{ab}Finally it is worth emphasizing what are special in our four-level tripod scheme involving an asymmetric procedure of light storage and retrieval. First, we use only one probe field to generate two DSPs, whose optical components of different frequencies interfere to generate beating signals. In refs. [23–26], however, beating signals are resulted from the interference between two probe fields or between a probe field and a coupling field. Second, it is easy for us to let the two DSPs have the same amplitude and experience the same loss and diffusion by simply manipulating the two coupling fields so that beating signals imposed on the only probe field can be kept perfect (i.e. always have vanishing minima in intensity).

## Acknowledgments

This work is supported by NBRP of China (No. 2011CB921603), NSFC of China (No. 10874057 and No. 10904047), BSRF of Jilin University (No. 200905019), and the Graduate Innovation Fund of Jilin University (No. 20101051).

## References and links

**1. **M. D. Lukin, “Colloquium: trapping and manipulating photon states in atomic ensembles,” Rev. Mod. Phys. **75**, 457–472 (2003). [CrossRef]

**2. **M. Fleischhauer, A. Imamoglu, and J. P. Marangos, “Electromagnetically induced transparency: Optics in coherent media,” Rev. Mod. Phys. **77**, 633–673 (2005). [CrossRef]

**3. **K. Bergmann, H. Theuer, and B. W. Shore, “Coherent population transfer among quantum states of atoms and molecules,” Rev. Mod. Phys. **70**, 1003–1025 (1998). [CrossRef]

**4. **D. A. Braje, V. Balic, G. Y. Yin, and S. E. Harris, “Low-light-level nonlinear optics with slow light,” Phys. Rev. A **68**, 041801(R) (2003). [CrossRef]

**5. **H. Kang and Y. Zhu, “Observation of large Kerr nonlinearity at low light intensities,” Phys. Rev. Lett. **91**, 093601 (2003). [CrossRef] [PubMed]

**6. **C.-Y. Wang, Y.-F. Chen, S.-C. Lin, W.-H. Lin, P.-C. Kuan, and I. A. Yu, “Low-light-level all-optical switching,” Opt. Lett. **31**, 2350–2352 (2006). [CrossRef] [PubMed]

**7. **C. Hang and G.-X. Huang, “Giant Kerr nonlinearity and weak-light superluminal optical solitons in a four-state atomic system with gain doublet,” Opt. Express **18**, 2952–2966 (2010). [CrossRef] [PubMed]

**8. **C. H. van der Wal, M. D. Eisaman, A. Andre, R. L. Walsworth, D. F. Phillips, A. S. Zibrov, and M. D. Lukin, “Atomic memory for correlated photon states,” Science **301**, 196–200 (2003). [CrossRef] [PubMed]

**9. **H.-H. Wang, X.-G. Wei, L. Wang, Y.-J. Li, D.-M. Du, J.-H. Wu, Z.-H. Kang, Y. Jiang, and J.-Y. Gao, “Optical information transfer between two light channels in a Pr^{3+}:Y_{2}SiO_{5}crystal,” Opt. Express **15**, 16044–16050 (2007). [CrossRef] [PubMed]

**10. **J. Appel, E. Figueroa, D. Korystov, M. Lobino, and A. I. Lvovsky, “Quantum memory for squeezed light,” Phys. Rev. Lett. **100**, 093602 (2008). [CrossRef] [PubMed]

**11. **K. S. Choi, H. Deng, J. Laurat, and H. J. Kimble, “Mapping photonic entanglement into and out of a quantum memory,” Nature (London) **452**, 67–71 (2008). [CrossRef]

**12. **C. Liu, Z. Dutton, C. H. Behroozi, and L. V. Hau, “Observation of coherent optical information storage in an atomic medium using halted light pulses,” Nature (London) **409**, 490–493 (2001). [CrossRef]

**13. **M. Fleischhauer and M. D. Lukin, “Dark-State Polaritons in Electromagnetically Induced Transparency,” Phys. Rev. Lett. **84**, 5094–5097 (2000). [CrossRef] [PubMed]

**14. **R. Unanyan, M. Fleischhauer, B. W. Shore, and K. Bergmann, “Robust creation and phase-sensitive probing of superposition states via stimulated Raman adiabatic passage (STIRAP) with degenerate dark states,” Opt. Commun. **155**, 144–154 (1998). [CrossRef]

**15. **J.-H. Wu, C.-L. Cui, N. Ba, Q.-R. Ma, and J.-Y. Gao, “Dynamical evolution and analytical solutions for multiple degenerate dark states in the tripod-type atomic system,” Phys. Rev. A **75**, 043819 (2007). [CrossRef]

**16. **A. Raczynski, M. Rzepecka, J. Zaremba, and S. Zielinska–Kaniasty, “Polariton picture of light propagation and storing in a tripod system,” Opt. Commun. **260**, 73–80 (2006). [CrossRef]

**17. **A. Raczynski, J. Zaremba, and S. Zielinska–Kaniasty, “Beam splitting and Hong-Ou-Mandel interference for stored light,” Phys. Rev. A **75**, 013810 (2007). [CrossRef]

**18. **C.-L. Cui, J.-K. Jia, J.-W. Gao, Y. Xue, G. Wang, and J.-H. Wu, “Ultraslow and superluminal light propagation in a four-level atomic system,” Phys. Rev. A **76**, 033815 (2007). [CrossRef]

**19. **D. Petrosyan and Y. P. Malakyan, “Magneto-optical rotation and cross-phase modulation via coherently driven four-level atomsin a tripod configuration,” Phys. Rev. A **70**, 023822 (2004). [CrossRef]

**20. **S. Rebic, D. Vitali, C. Ottaviani, P. Tombesi, M. Artoni, F. Cataliotti, and R. Corbalan, “Polarization phase gate with a tripod atomic system,” Phys. Rev. A **70**, 032317 (2004). [CrossRef]

**21. **Y.-X. Han, J.-T. Xiao, Y.-H. Liu, C.-H. Zhang, H. Wang, M. Xiao, and K.-C. Peng, “Interacting dark states with enhanced nonlinearity in an ideal four-level tripod atomic system,” Phys. Rev. A **77**, 023824 (2008). [CrossRef]

**22. **S.-J. Li, X.-D. Yang, X.-M. Cao, C.-H. Zhang, C.-D. Xie, and H. Wang, “Enhanced Cross-Phase Modulation Based on a Double Electromagnetically Induced Transparency in a Four-Level Tripod Atomic System,” Phys. Rev. Lett. **101**, 073602 (2008). [CrossRef] [PubMed]

**23. **L. Karpa, F. Vewinger, and M. Weitz, “Resonance beating of light stored using atomic spinor polaritons,” Phys. Rev. Lett. **101**, 170406 (2008). [CrossRef] [PubMed]

**24. **A. Mair, J. Hager, D. F. Phillips, R. L. Walsworth, and M. D. Lukin, “Phase coherence and control of stored photonic information,” Phys. Rev. A **65**, 031802(R) (2002). [CrossRef]

**25. **L. Karpa, G. Nikoghosyan, F. Vewinger, M. Fleischhauer, and M. Weitz, “Frequency matching in light-storage spectroscopy of atomic Raman transitions,” Phys. Rev. Lett. **103**, 093601 (2009). [CrossRef] [PubMed]

**26. **G. Heinze, A. Rudolf, F. Beil, and T. Halfmann, “Storage of images in atomic coherences in a rare-earth-ion-doped solid,” Phys. Rev. A **81**, 011401(R) (2010).

**27. **E. Paspalakis and P. L. Knight, “Electromagnetically induced transparency and controlled group velocity in a multilevel system,” Phys. Rev. A **66**, 015802 (2002). [CrossRef]

**28. **E. Kuznetsova, O. Kocharovskaya, P. Hemmer, and M. O. Scully, “Atomic interference phenomena in solids with a long-lived spin coherence,” Phys. Rev. A **66**, 063802 (2002). [CrossRef]

**29. **B. S. Ham, “Reversible quantum optical data storage based on resonant Raman optical field excited spin coherence,” Opt. Express **16**, 14304–14313 (2008). [CrossRef] [PubMed]

**30. **H. Wang, S.-J. Li, Z.-X. Xu, X.-B. Zhao, L.-J. Zhang, J.-H. Li, Y.-L. Wu, C.-D. Xie, K.-C. Peng, and M. Xiao, “Quantum interference of stored dual-channel spin-wave excitations in a single tripod system,” Phys. Rev. A **83**, 043815 (2011). [CrossRef]