Optical response in six-level 85Rb and 87Rb atomic media Authors: Le Quang Trung, Hoang Thi Bich Ngoc, Nguyen Thu An, Ho Nguyen An, Dang Khanh Huyen, Nguyen Huy Bang, Hoang Dinh Hai, Le Van Doai Published: 2026 ABSTRACT AbstractIn this work, we analytically investigate the optical response including absorption, dispersion, and group index of a six-level atomic system under electromagnetically induced transparency (EIT). A key advantage of this configuration is its ability to simultaneously generate four transparency windows at distinct frequency regions using only two controlling laser fields. The analytical model is applied to both 87Rb and 85Rb atoms in cold and hot conditions, enabling exploration under various experimental environments. The results provide a comprehensive picture of the fundamental optical properties of 87Rb and 85Rb atomic systems with the six-level configuration, offering useful insights for experimental predictions and potential applications.Subject terms: Optics and photonics, Physics Introduction Understanding and controlling the optical response of atomic media-how atoms absorb, refract, and delay light-has been a longstanding goal in both classical and quantum optics. These fundamental properties not only reveal the underlying energy structure and coherence dynamics of atomic systems, but also underpin a wide range of photonic applications, from slow-light buffers to quantum memories and nonlinear optical devices. Among the most striking phenomena related to optical response is electromagnetically induced transparency (EIT)1, where quantum interference enables a resonant atomic medium to become nearly transparent to a weak probe beam. Along with transparency, EIT introduces sharp dispersion features, leading to highly tunable group indices and slow or fast light propagation2. These dramatic modifications of the optical response, suppressed absorption, enhanced dispersion, and large group delay have made EIT a key effect in the design of next-generation light-based technologies. Since its discovery in 19911, EIT has become a cornerstone of research in coherent light-matter interactions, finding applications in slow light3, quantum memory4, all-optical switching5, high-resolution hyperfine spectroscopy6, and low-power nonlinear optics7 and so on. Early studies focused primarily on three-level atomic systems8–10, which offer a single transparency window and are relatively simple to analyze both theoretically and experimentally. However, advances in laser technology and atomic control have enabled exploration of more complex multi-level schemes, allowing for the generation of multiple EIT windows and richer optical responses. Multi-window EIT configurations are particularly attractive for multi-frequency slow-light applications, optical signal multiplexing, and multi-channel quantum control. These can be realized through various architectures, including N-type, ladder-type, M-type, K-type schemes, often requiring several control fields to simultaneously address different transition pathways11–14. Recent studies have demonstrated that even with a limited number of control fields, multi-window EIT can emerge in atomic systems possessing closely spaced hyperfine or Zeeman sublevels15–18. Such configurations are not only experimentally feasible but also offer improved spectral flexibility and enhanced control over optical dispersion. In parallel, the role of temperature has received increasing attention. While cold atom systems offer high spectral resolution and reduced Doppler broadening, hot vapor cells are easier to operate and integrate into compact devices. Understanding how EIT manifests under both cold and thermal conditions is crucial for real-world applications, particularly in atomic systems such as rubidium, where both isotopes (87Rb and 85Rb) exhibit rich hyperfine structures19–21. Despite considerable experimental progress, most multi-level EIT studies rely on numerical simulations. Analytical approaches are scarce but essential, as they provide direct physical insight and scalability to broader parameter spaces. Recent analytical developments have paved the way for compact modeling of complex atomic systems, enabling accurate prediction of absorption, dispersion, and group index behavior under varying conditions16,19,22–24. This work presents an analytical study of the optical response in a six-level atomic system driven by two coherent control fields. We focus on key quantities of interest in EIT media, namely, the absorption spectrum, dispersion profile, and group refractive index, and explore their behavior in both cold and hot ensembles of 87Rb and 85Rb atoms. A key characteristic of our model is the simultaneous generation of four distinct EIT windows, achieved without the need for additional control lasers. This multi-window structure offers enhanced flexibility for manipulating light propagation at multiple frequencies, a feature relevant for quantum information processing and broadband optical buffering. By systematically comparing the EIT spectra under varying isotopes and thermal conditions, we aim to identify optimal regimes for achieving large, tunable group indices and high-resolution transparency features. Our findings advance the theoretical understanding of EIT in complex atomic systems and support the development of practical schemes for light control in both laboratory settings and integrated photonic devices. Theoretical model The six-level atomic configuration considered in this study is coherently driven by three laser fields, as illustrated in Fig. 1. A weak probe laser field with frequency ωp​ and Rabi frequency Ωp​ couples the transition |1〉↔|2〉, while a signal field of frequency ωs​ and Rabi frequency Ωs drives the |6〉↔|2〉 transition. Simultaneously, a strong coupling laser with frequency ωc and Rabi frequency Ωc interacts with the |2〉↔|3〉, |2〉↔|4〉, and |2〉↔|5〉 transitions. The energy separations between the excited states |3〉, |4〉 and |5〉 are denoted as δ1 and δ2, corresponding to |3〉-|4〉 and |5〉-|3〉, respectively. Six-level atomic system interacts with probe, coupling and signal fields. The density matrix equations describing the response of the atomic system in the laser fields are derived from the Liouville equation: where ρ is the density-matrix of order 6 × 6, Λρ describes the relaxation term associated with the population decay from the upper to the lower level and is introduced into the density matrix equations in a phenomenological way, and H is the total Hamiltonian of the system which is given by: In the framework of the semiclassical theory and using the rotating wave and electric dipole approximations under rotating frame, the density-matrix equations of motion of this scheme can be written as: The density matrix elements describing population are interrelated through normalized condition ; where,, and are the frequency detuning of the probe, coupling and signal lasers, respectively; Ωp = d21Ep/ћ, Ωc = d32Ec/ћ and Ωs = d26Es/ћ are respectively Rabi frequency induced by the probe, coupling and signal fields; dmn is a dipole moment of the |m〉-|n〉 transition; a32 = d32/d32, a42 = d42/d32 and a52 = d52/d32 are the relative transition strengths; Γmn is the population decay rate from level |m〉 to level |n〉, and γmn represents the coherence decay rate, . Now, we solve the density matrix equations in the steady state () and under the weak-probe approximation, then from Eqs. (10)–(13), we easily obtain: By substituting Eqs. (24a-d) into Eq. (9), the analytical solution for the coherence term ρ21​, which determines the probe optical response, is given by: where we assumed that all atoms were initially prepared in the ground state ∣1⟩ with ρ11 = 1, while all other populations were set to zero; A, B, C and D are controllable parameters given by The susceptibility of the atomic medium for the probe field is given by where, N is density of particles, ε0 is the permittivity in vacuum and Ep is electric field of probe light and For atomic vapors at room temperature, the Doppler effect must be taken into account in the expression of the optical susceptibility. To suppress the first-order Doppler broadening, we assume that the coupling and probe laser beams co-propagate through the atomic medium, while the signal laser field propagates in the opposite direction. Thus, if the atoms move with velocity v along the direction of the probe laser propagation, the frequencies of the probe and coupling laser fields are Doppler-shifted by , , respectively; in contrast, the signal laser field, which counter-propagates relative to the atomic motion, experiences a Doppler shift of . This results in modifications of the frequency detunings of the laser fields. Specifically, the detunings of the probe and coupling lasers are shifted by and , respectively, while the detuning of the signal laser is shifted by . The components of the atomic velocities along the beam axis follow the Maxwell distribution: where, is the root-mean-square speed of the atoms, N0 is the total atomic density in the sample. Thus, the electric susceptibility when the Doppler effect is considered as: Performing the integration with v over, we get: where, and is the complex conjugate of z, with erf being the error function. The absorption and dispersion coefficients of the atomic medium for the probe laser beam are given by: Applying to 87Rb and 85Rb atoms In general, an analytical model can be applied to any atomic or molecular systems that have an energy-level diagram similar to that presented in Fig. 1. In this paper, we apply the analytical model for the Doppler broadened medium for the 5S1/2 – 5P3/2 – 5D5/2 transition of 85Rb and 87Rb atoms. The natural linewidths of the 5P3/2 and 5D5/2 states are 6.1 MHz and 0.97 MHz, respectively. For 85Rb atom, the probe laser drives the transition 5S1/2, F = 2→5P3/2, F′ = 3, while the coupling laser couples three sets of transitions 5P3/2, F′ = 3→5D5/2, F″ = 4, F″ = 3 and F″ = 2, and the signal laser drives the transition 5S1/2, F = 3→5P3/2, F′ = 3. For 87Rb atom, the probe laser drives the transition 5S1/2, F = 1→5P3/2, F′ = 2, and the coupling laser drives three transitions 5P3/2, F′ = 2→5D5/2, F″ = 3, F″ = 2 and F″ = 1, and the signal laser drives the transition 5S1/2, F = 2→5P3/2, F′ = 2. The frequency separations between the hyperfine levels of 87Rb atom are δ1 = 23.0 MHz and δ2 = 15.9 MHz while for the 85Rb atom, the corresponding values are as δ1 = 9.0 MHz and δ2 = 7.6 MHz. The relative transition strengths are a32 : a42 : a52 = 1 : 1.46 : 0.615. Within the scope of this study, we investigate the optical responses of the atomic medium, including absorption, dispersion, and group refractive index. For each case applied to 87Rb and 85Rb atoms, we consider both cold atoms (cooled and trapped in MOT) and hot atoms at room temperature (T = 300 K). The coupling laser frequency is chosen to be resonant with the atomic transition, while the signal laser frequency is detuned from atomic resonance in order to generate four distinct EIT windows. In principle, this six-level configuration allows for a maximum of four EIT windows, corresponding to the four terms A, B, C, and D in Eq. (25). The positions of these EIT windows are determined by the two-photon resonance conditions: Δp + Δc = 0, Δp + Δc + δ1 = 0, Δp + Δc − δ2 = 0 and Δp − Δs = 0. Accordingly, if we set Δc = 0, the positions of the four EIT windows become: Δp = 0, Δp = −δ1, Δp = δ2 and Δp = Δs. It should be noted that if the signal laser is also tuned to atomic resonance (Δs = 0), two EIT windows will overlap at Δp = 0, resulting in only three observable EIT windows. First, we apply the theoretical model to cold 87Rb atoms with hyperfine level separations of δ1 = 23 MHz and δ2 = 15.9 MHz. The absorption spectra (upper curves) and dispersion spectra (lower curves) of the cold 87Rb atoms are shown in Fig. 2. Here, the Rabi frequencies of the controlling laser fields are chosen as Ωc = Ωs = 10 MHz, Δc = 0, and Δs = −10 MHz (a) and Δs = 10 MHz (b). In the case of Fig. 2(a), with Δs = −10 MHz, four EIT windows are clearly observed at the probe detunings Δp = 0 (induced by the coupling laser on the |2〉−|3〉 transition), Δp = −δ1 = −23 MHz (induced by the coupling laser on the |2〉−|4〉 transition), Δp = δ2 = 15.9 MHz (induced by the coupling laser on the |2〉−|5〉 transition), and Δp = Δs = −10 MHz (induced by the signal laser on the |6〉−|2〉 transition). Similarly, in Fig. 2b, where Δs = 10 MHz, the EIT windows appear at Δp = 0, Δp = −δ1 = −23 MHz, Δp = δ2 = 15.9 MHz, and Δp = Δs = 10 MHz. Thus, the coupling laser can induce three EIT windows due to its ability to simultaneously couple three hyperfine transitions, while the signal laser contributes an additional EIT window in the probe absorption spectrum. Absorption with four EIT windows (upper curve) and dispersion (lower curve) of cold 87Rb atoms, with the controlling laser parameters set as Ωc = Ωs = 10 MHz, Δc = 0, and Δs = −10 MHz (a) and Δs = 10 MHz (b). Our subsequent investigation is applied to hot 87Rb atoms at room temperature (T = 300 K), corresponding to a Doppler broadening of = 520 MHz. In this case, the EIT effect is more difficult to observe than in the cold atom scenario9, thus requiring stronger controlling laser fields. The simulations presented in Fig. 3 are carried out similarly to those in Fig. 2, except that the Rabi frequencies of the controlling lasers are increased to Ωc = Ωs = 30 MHz. From Fig. 3, four EIT windows can still be clearly identified in the absorption spectra at the probe detunings Δp = 0, Δp = −23 MHz, Δp = 15.9 MHz and Δp = −10 MHz (a) or Δp = 10 MHz (b). Compared to the cold atom case, the absorption profiles here are significantly broadened due to Doppler effect, which makes the EIT windows less resolved. The corresponding dispersion curves, shown as the blue lines in the lower parts of Fig. 3, also exhibit characteristic features associated with the EIT windows. Absorption with four EIT windows (upper curve) and dispersion (lower curve) of hot 87Rb atoms at T = 300 K, with the controlling laser parameters set as Ωc = Ωs = 30 MHz, Δc = 0, and Δs = −10 MHz (a) and Δs = 10 MHz (b). Next, we apply the theoretical model to the 85Rb atomic system, in which the frequency separations between hyperfine levels are δ1 = 9 MHz and δ2 = 7.6 MHz. The absorption and dispersion spectra of cold and hot 85Rb atoms are presented in Figs. 4 and 5, respectively. For the case of cold 85Rb atoms, the parameters of the controlling laser fields are chosen as Ωc = Ωs = 10 MHz, Δc = 0, and Δs = −15 MHz (a) and Δs = 15 MHz (b). Figure 4 shows four EIT windows located at Δp = 0 (induced by the coupling laser on the |2〉−|3〉 transition), Δp = −δ1 = −9 MHz (induced by the coupling laser on the |2〉−|4〉 transition), Δp = δ2 = 7.6 MHz (induced by the coupling laser on the |2〉−|5〉 transition), and Δp = Δs = −15 MHz (a) or Δp = Δs = 15 MHz (b), both induced by the signal laser on the |2〉−|4〉 transition. Due to the smaller frequency separation between the hyperfine levels of the 5D5/2​ state in 85Rb atom compared to 87Rb atom, the corresponding EIT windows in 85Rb atom are more closely spaced, resulting in lower spectral resolution, especially in the presence of Doppler broadening, as illustrated in Fig. 5. In Fig. 5, for hot 85Rb atoms at room temperature (T = 300 K), the controlling laser parameters are set as Ωc = Ωs = 30 MHz, Δc = 0, and Δs = −15 MHz (a) and Δs = 15 MHz (b). Although four EIT windows are still formed at the same positions as in the cold atom case, their observation becomes more difficult due to the fact that the hyperfine frequency separations are much smaller than the Doppler width. In some experiments, these EIT windows may overlap significantly, such that only a single EIT peak can be observed9. From this analysis, we can conclude that for room-temperature applications involving Rb atomic media, the EIT spectra of 87Rb atoms offer better resolution than those of 85Rb atoms. Absorption with four EIT windows (upper curve) and dispersion (lower curve) of cold 85Rb atoms, with the controlling laser parameters set as Ωc = Ωs = 10 MHz, Δc = 0, and Δs = −15 MHz (a) and Δs = 15 MHz (b). Absorption with four EIT windows (upper curve) and dispersion (lower curve) of hot 85Rb atoms at T = 300 K, with the controlling laser parameters set as Ωc = Ωs = 30 MHz, Δc = 0, and Δs = −15 MHz (a) and Δs = 15 MHz (b). One of the hallmark features of EIT spectra is the emergence of extremely steep dispersion curves with large amplitudes, significantly greater than those found in conventional materials. This steep dispersion leads to a dramatically enhanced group index, enabling the slowing down or even complete stopping of light within the EIT transparency window. In the following, we investigate the group index of 87Rb and 85Rb atomic media under both cold and hot conditions, as discussed previously. As is well known, the group index experienced by a probe beam is determined by the relation which is directly proportional to the slope of the dispersion curve. Using the same laser parameters as in Figs. 2, 3, 4 and 5, we calculate and plot the corresponding group index curves. Specifically, the group index curves shown in Fig. 6a,b correspond to the conditions of Fig. 2a,b, respectively, for cold 87Rb atoms. In Fig. 6a, where Δs = −10 MHz, four regions of positive group index (slow-light regions) are clearly observed at the EIT window positions: Δp = 0, Δp = −23 MHz, Δp = 15.9 MHz and Δp = −10 MHz. Between these regions, negative group index values (fast-light regions) appear, with amplitudes significantly larger than those of the positive group index. Similarly, in Fig. 6b, for Δs = 10 MHz, the slow-light regions are located at Δp = 0, Δp = −23 MHz, Δp = 15.9 MHz and Δp = 10 MHz, interleaved again with fast-light regions exhibiting negative group indices. Group index with four pairs of positive-negative peaks of cold 87Rb atoms, with the controlling laser parameters set as Ωc = Ωs = 10 MHz, Δc = 0, and Δs = −10 MHz (a) and Δs = 10 MHz (b). The presence of alternating slow-light (positive group index) and fast-light (negative group index) regions reflects the rapid variation of the dispersion slope near the transparency windows. In particular, the negative group index corresponds to anomalous dispersion, which is known to be associated with pulse advancement and fast-light propagation. Although fast light does not imply superluminal information transfer, such behavior can be exploited for applications such as temporal cloaking, optical delay compensation, and enhanced phase sensitivity in interferometry. In contrast, the positive group index regions correspond to normal dispersion and light slowing, which are essential for optical buffering, quantum memory, and nonlinear enhancement. The ability to tailor the group index through laser detuning and atomic configurations thus provides a versatile tool for controlling light propagation in atomic media. The results here demonstrate that cold 87Rb atoms under EIT conditions can support both slow and fast light over multiple spectral regions, offering rich possibilities for tunable light–matter interaction. Figure 7 presents the group index curves for hot 87Rb atoms at room temperature (T = 300 K), corresponding to the absorption and dispersion profiles shown in Fig. 3. In this case, the Rabi frequencies of the controlling laser fields are set to Ωc = Ωs = 30 MHz, while the detuning parameters are kept the same as in Fig. 6. As in the cold atom case, both positive (slow-light) and negative (fast-light) group index regions emerge at similar spectral positions. However, the peak values of the positive group index are significantly enhanced compared to the negative ones. Moreover, the overall amplitude of the group index in the hot atom case is reduced by several orders of magnitude relative to that of the cold atoms, due to Doppler broadening. In addition, the spectral features are broadened significantly; resulting in reduced resolution and less sharply defined EIT-induced group delay structures. Group index with four pairs of positive-negative peaks of hot 87Rb atoms at T = 300 K, with the controlling laser parameters set as Ωc = Ωs = 30 MHz, Δc = 0, and Δs = −10 MHz (a) and Δs = 10 MHz (b). These observations highlight the substantial influence of Doppler broadening on the group index behavior in hot atomic vapors. While the EIT features and associated slow- and fast-light regions persist, the reduction in peak group index and spectral resolution limits their effectiveness for applications requiring strong dispersion or long optical delays. Nevertheless, the enhanced stability, simplicity, and practicality of room-temperature atomic systems make them attractive for compact and robust implementations of EIT-based technologies. In particular, hot 87Rb vapor still supports tunable slow-light propagation, with sufficiently high group indices for applications such as optical delay lines, magneto-optical sensing, and phase-sensitive measurements. The results suggest that careful optimization of the laser parameters, especially the Rabi frequencies and detunings can partially compensate for Doppler-induced broadening. Thus, room-temperature Rb vapor cells remain viable platforms for EIT-based functional devices, especially where cryogenic or laser-cooling setups are impractical. Similarly, Figs. 8 and 9 present the group index curves for cold and hot 85Rb atoms, respectively, corresponding to the absorption and dispersion conditions shown in Figs. 4 and 5. For 85Rb atoms, positive group index regions appear at spectral positions associated with EIT windows: Δp = 0, Δp = −9 MHz, Δp = 7.6 MHz and Δp = −15 MHz (a) or Δp = 15 MHz (b), interleaved with regions of negative group index. In the cold atom case (Fig. 8), prominent peaks of negative group index are observed, whereas in the hot atom case (Fig. 9), the positive group index peaks become significantly enhanced. Due to the relatively small hyperfine level separations in 85Rb, the corresponding group index features are more closely spaced, making individual EIT contributions more difficult to resolve. Group index with four pairs of positive–negative peaks of cold 85Rb atoms, with the controlling laser parameters set as Ωc = Ωs = 10 MHz, Δc = 0, and Δs = −15 MHz and Δs = 15 MHz. Group index with four pairs of positive-negative peaks of hot 85Rb atoms at T = 300 K, with the controlling laser parameters set as Ωc = Ωs = 30 MHz, Δc = 0, and Δs = −15 MHz (a) and Δs = 15 MHz (b). The close spacing of group index peaks in 85Rb atom is a direct consequence of its smaller hyperfine separations compared to 87Rb atom, which leads to spectral crowding and reduced resolution in both the absorption and group index spectra. This spectral overlap becomes particularly problematic in the presence of Doppler broadening, where individual EIT-induced features merge, thus limiting the ability to isolate narrow slow-light windows. In contrast to 87Rb atom, where the group index structure is well separated and symmetric, 85Rb exhibits more complex and asymmetric group index profiles. Notably, in the cold atom regime, the fast-light (negative group index) regions dominate in amplitude, while in the hot atom case, slow-light (positive group index) peaks become more pronounced. This inversion of dominance between cold and hot conditions is indicative of different sensitivity of the EIT features to temperature and Doppler effects in the two isotopes. Overall, while 85Rb atoms can still exhibit EIT and associated dispersion control, the more favorable spectral separation in 87Rb makes it a superior choice for applications requiring high group index contrast, sharper resolution, and better tunability, especially in room-temperature environments. Conclusion In this study, we have analytically investigated the optical response of a six-level atomic system under electromagnetically induced transparency (EIT), focusing on absorption, dispersion, and group index. This configuration enables the generation of four distinct transparency windows using only two control fields, offering flexible multi-channel light manipulation within a single atomic ensemble. Our model has been systematically applied to both 87Rb and 85Rb isotopes under cold and hot conditions. The results highlight several key findings. In cold 87Rb atoms, the EIT windows are spectrally well-resolved, producing sharply varying dispersion profiles and giant group indices on the order of 104 − 106, suitable for slow-light and optical buffering applications. The ability to control the positions and strength of these transparency windows by adjusting laser detunings provides a high degree of spectral tunability and selectivity. In contrast, hot atomic vapors at room temperature suffer from Doppler broadening, which reduces the amplitude and sharpness of the EIT and group index features. Nevertheless, observable transparency windows and meaningful dispersion effects are retained when the Rabi frequencies of the control fields are increased appropriately. This demonstrates the practicality of implementing EIT-based slow- and fast-light phenomena even in thermal vapor cells, which are more accessible than cold-atom setups. For 85Rb, the narrower hyperfine level separations lead to denser spectral features and more significant overlap between adjacent EIT windows, particularly in the hot atom case. This limits the achievable spectral resolution but still allows for light-matter interaction modulation in regimes where 87Rb may be less favorable, such as in compact or isotopically mixed systems. The significance of this research lies in its comprehensive characterization of the fundamental optical properties of a complex atomic system under realistic conditions. The ability to predict and control EIT behavior across multiple channels and temperature regimes provides a strong foundation for the development of advanced quantum and classical optical technologies. Looking forward, this six-level EIT scheme can be extended to include nonlinear effects, interactions with structured light, or integration with nanophotonic structures such as hollow-core fibers and metasurfaces. Experimental validation of the predicted multi-window EIT spectra and group index dynamics, especially under room-temperature conditions, will be the next step toward practical implementations in optical memory, magnetometry, delay lines, and all-optical switching. Acknowledgements This research was funded by Vingroup Innovation Foundation (VINIF) under project code VINIF.2022.DA00076. Author contributions Le Quang Trung, Hoang Thi Bich Ngoc, Nguyen Thu An, Ho Nguyen An, Dang Khanh Huyen, Nguyen Huy Bang, Hoang Dinh Hai and Le Van Doai conceived of the presented idea, developed the theory and performed the analytical calculations. All authors co-wrote the paper, discussed the results and contributed to the final manuscript. Funding This research was funded by Vingroup Innovation Foundation (VINIF) under project code VINIF.2022.DA00076. Data availability All the data generated/analyzed during the current study available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement All the data generated/analyzed during the current study available from the corresponding author on reasonable request.