Threshold-Selective Raman Purification

• Threshold-selective Raman purification is an optical technique that uses sharp Raman gain thresholds to selectively transfer or amplify target states in polarization and velocity domains.
• Key methods include all-optical purification in PM-AR-HCFs and velocity-selective π-pulse transitions in Λ-systems, resulting in high polarization extinction ratios and narrow velocity distributions.
• Experimental implementations demonstrate efficient Stokes generation with PER up to 35 dB and quantum conversion efficiencies near 67%, ensuring robustness even under fiber bending conditions.

Threshold-selective Raman purification is an optical technique enabling selective population transfer or signal amplification based on sharp threshold criteria imposed by Raman processes. Two principal realizations are supported: (1) all-optical purification of polarization states via threshold-selective stimulated Raman scattering in birefringent gas-filled polarization-maintaining (PM) anti-resonant hollow-core fibers (AR-HCF), and (2) velocity-selective purification of particle ensembles by π-pulse driven stimulated Raman transitions in Λ-systems. Each approach leverages intrinsic or engineered disparities in Raman gain thresholds to "purify" an initially mixed state into a highly selective target distribution, whether polarization, frequency, or velocity. This article presents fundamental mechanisms, governing equations, and experimental implementations as established in the literature (Qi et al., 31 Jan 2026, 2002.04511).

1. Fundamental Principles of Threshold-Selective Raman Purification

Threshold-selective Raman purification operates by exploiting sharp nonlinearities in the probability of Raman transitions as system parameters cross an axis-dependent or velocity-dependent threshold. In PM-AR-HCFs, differential Raman thresholds originate from structural birefringence, resulting in preferential amplification along a principal axis when the global pump power exceeds only one threshold. In Λ-systems for particle beams, selective coherent π-pulse transfer is achieved for particles fulfilling a two-photon resonance—precisely those in a narrow velocity window. In both cases, selectivity is governed by gain or transfer efficiency that rises steeply above threshold for the desired subensemble while remaining negligible for orthogonal components or off-resonant velocities.

2. Birefringence-Engineered Polarization Purification in PM-AR-HCF

The PM-AR-HCF platform consists of a single-ring anti-resonant hollow-core fiber with six silica capillaries asymmetrically arranged (bi-thickness semi-tube geometry), leading to strong structural birefringence. Orthogonal capillary-wall thicknesses yield distinct anti-resonances for fast and slow polarization axes, with measured phase birefringence of Δn ≃ 1.48×10^–5 (1064 nm, pump) and ≃ 8.21×10^–5 (1415 nm, Stokes). Resultant polarization-beat lengths (B_p ≃ 7.2 cm, B_s ≃ 1.7 cm) are much shorter than typical fiber lengths, suppressing linear polarization cross-talk. The net effect is near-decoherence between axes, allowing axis-specific stimulated Raman scattering dynamics (Qi et al., 31 Jan 2026).

3. Coupled Maxwell-Bloch Equations and Raman Threshold Dynamics

Stimulated Raman scattering in the PM-AR-HCF is described under slowly-varying envelope and single-mode assumptions by independent 1D coupled Maxwell-Bloch equations for each principal axis (i ∈ {x, y}, i.e., "fast", "slow"):

$$\begin{aligned} \partial_z E_{P,i} + (1/v_{gP,i})\partial_t E_{P,i} &= - i (\omega_P/c) \chi_{R,i} Q_i E_{S,i} - \alpha_{P,i} E_{P,i} \ \partial_z E_{S,i} + (1/v_{gS,i})\partial_t E_{S,i} &= - i (\omega_S/c) \chi_{R,i} Q^*_i E_{P,i} - \alpha_{S,i} E_{S,i} \ \partial_t Q_i + (1/T_2) Q_i &= - i (g_i/\hbar \omega_S) E_{P,i} E^*_{S,i} \end{aligned}$$

Here, $E_{P,i}$ and $E_{S,i}$ are the pump and Stokes envelopes, $Q_i$ is the Raman coherence (dephasing time $T_2 \sim 8$ ps), $g_i$ the axis-dependent Raman gain, and $\alpha_{P,i}$, $\alpha_{S,i}$ the losses. Under long-pulse, steady-state, and undepleted-pump conditions, gain on each axis is exponential above threshold pump power:

$$P_{\mathrm{th},i} = \frac{A_{\mathrm{eff},i}\,\Delta\omega}{g_{i}\,L_{\mathrm{eff}}}$$

where $A_{\mathrm{eff},i}$ is mode area, $E_{P,i}$0 the Raman linewidth, and $E_{P,i}$1 the effective interaction length.

Due to birefringence-induced disparities ($E_{P,i}$2, $E_{P,i}$3), two thresholds split: $E_{P,i}$4. For global pump power $E_{P,i}$5 such that $E_{P,i}$6, Stokes amplification proceeds exclusively along the fast axis, while the orthogonal component remains unamplified.

4. Polarization Extinction and Purification Mechanism

The polarization extinction ratio (PER) of the output Stokes field is sharply enhanced within the threshold window. For an initial pump with polarization angle $E_{P,i}$7 to the fast axis, and undepleted regime, output Stokes powers are:

$E_{P,i}$8

with $E_{P,i}$9, $E_{S,i}$0.

The PER is:

$E_{S,i}$1

For $E_{S,i}$2, $E_{S,i}$3 increases rapidly. At sufficiently high $E_{S,i}$4, both axes amplify and $E_{S,i}$5 saturates at the limit dictated by residual cross-axis coupling (measured ≃40 dB, realized ≃35 dB in PM-AR-HCF). This behavior underlies intrinsic polarization purification: even with an incident pump PER ≃2 dB, the vibrational Stokes can reach PER ≈35 dB, without external polarizers (Qi et al., 31 Jan 2026).

5. Experimental Implementation and Performance

An exemplary implementation is realized in 2.5 m N₂-filled PM-AR-HCF with 15 μm core diameter, birefringence as above, and pure N₂ at 22 bar. Using 6 ns, 1064 nm pump pulses (up to 300 μJ, average ≈140 mW), the first vibrational Stokes at 1415 nm is generated with observed PER up to 35 dB and quantum conversion efficiency up to 67%. A key property is bend tolerance: both PER and Stokes power remain unchanged for coil radii down to 5 cm, a regime where non-PM HCFs exhibit severe degradation at <25 cm radius. This stabilizes polarization over deployment-relevant fiber deformations (Qi et al., 31 Jan 2026).

Parameter	Value (PM-AR-HCF Implementation)	Note
Core diameter	≈15 μm
Pump wavelength	1064 nm	6 ns pulses
Gas / pressure	Pure N₂ at 22 bar	With positive gradient
Max Stokes PER	35 dB
Quantum efficiency (Stokes)	up to 67%
Polarization-beat lengths $E_{S,i}$6	7.2 cm (pump), 1.7 cm (Stokes)
Bend insensitivity	Down to 5 cm coil radius

6. Velocity-Selective Raman Purification in Λ-Systems

Threshold-selective Raman purification is also applicable to selecting a narrow velocity class in particle beams, as in caesium or Ca$E_{S,i}$7 ion spectroscopy (2002.04511). This method uses two far-detuned, counter-propagating lasers driving a Λ-system, inducing a stimulated two-photon transition that is Doppler-selective. Only particles at velocity $E_{S,i}$8 (fulfilling $E_{S,i}$9 in the two-photon detuning) undergo a complete π-pulse population transfer, forming a purified ensemble with width (FWHM):

$Q_i$0

Here, $Q_i$1, $Q_i$2 are one-photon Rabi frequencies, $Q_i$3 is the detuning, and $Q_i$4 the laser wavenumbers. For $Q_i$5Ca$Q_i$6 ions, realistic parameters ($Q_i$7 GHz, $Q_i$8 MHz) yield velocity windows $Q_i$9–$T_2 \sim 8$0 m/s, corresponding to relative sensitivities at the few ppm level in beam energy. Master-equation analysis including laser phase noise and spontaneous emission refines the selectivity and efficiency limits (2002.04511).

7. Applications and Technological Impact

Threshold-selective Raman purification enables robust, intrinsic polarization purification in all-fiber gas-photonic sources (for metrology, quantum communication, sensing) and Doppler-selective purification for high-precision atomic and ionic beam manipulation. The method’s immunity to fiber bending and twisting, reliance on engineered birefringence or resonance criteria, and ab-initio design flexibility render it suitable for deployment in field environments and scalable architectures. In gas-filled PM-AR-HCFs, polarization can be considered an actively engineerable degree of freedom rather than an afterthought. In Λ-system spectroscopies, sub-ppm velocity selection is feasible with straightforward experimental control (Qi et al., 31 Jan 2026, 2002.04511).