Phaseonium Gas: Coherence-Controlled Quantum Systems

• Phaseonium gas is a quantum medium defined by controlled phase coherence between atomic states, enabling phase-sensitive optical and thermodynamic responses.
• It leverages three-level Λ systems and multilevel configurations to manipulate light transmission, resonance features, and spin–orbit interactions through coherent superpositions.
• Experimental realizations show that tuning the coherence phase can switch effective heating, cooling, and work extraction in quantum thermodynamic models.

Phaseonium gas denotes an atomic ensemble or beam in which the controlled phase of internal coherences is itself an operative degree of freedom. In the usage associated with Scully and subsequent quantum-optical and quantum-thermodynamic work, phaseonium is not defined solely by populations in a multilevel atom, but by a phase-coherent superposition—typically between two low-lying or degenerate states—whose relative phase alters propagation, absorption, gain, effective temperature, and work extraction. Across the literature, the term encompasses closely related realizations: three-level $\Lambda$ media with phase-sensitive susceptibility, coherently prepared media for structured-light propagation, and streams of phase-coherent ancillas acting as non-thermal reservoirs (Jha et al., 2012, Amato et al., 2023, Permana et al., 21 Nov 2025, Amato et al., 2 Sep 2025, Türkpençe et al., 2015).

1. Core definition and representative realizations

The defining property of a phaseonium medium is that its response depends explicitly on atomic coherence phases, not only on level populations. In a standard thermal gas, the density matrix is diagonal in the energy basis. In phaseonium, off-diagonal density-matrix elements between low-lying states are prepared or sustained and become experimentally accessible control parameters. In the optical setting, this makes the susceptibility phase-sensitive. In the thermodynamic setting, it makes the reservoir effectively non-thermal and assigns it a phase-dependent apparent temperature (Jha et al., 2012, Amato et al., 2023).

Paper	Configuration	Phase-controlled effect
(Jha et al., 2012)	$\Lambda$-type medium in a photonic crystal with a microwave/maser field	Transmission, reflection, and hyperfine resonances
(Permana et al., 21 Nov 2025)	Coherently prepared $\Lambda$ phaseonium probed by optical vector vortices	$2|l|$-fold azimuthal transparency and SAM exchange
(Amato et al., 2023)	Cascade of two cavities interacting with phaseonium ancillas	Heating, cooling, and phase-dependent steady temperature
(Amato et al., 2 Sep 2025)	Phaseonium-driven optomechanical Otto engine	Coherence-induced effective hot and cold reservoirs
(Türkpençe et al., 2015)	Multilevel phaseonium fuel with $N$ coherent lower levels	Quadratic scaling of work and efficiency

A common misconception is to identify phaseonium with any $\Lambda$-system exhibiting EIT. The literature distinguishes the broader EIT mechanism from the more specific phaseonium condition: the coherence phase itself must be part of the controllable physics. That control can be established dynamically, for example by closing a three-field loop with a maser, or kinematically, by preparing the atomic ensemble in a coherent ground-state superposition before the probe arrives (Jha et al., 2012, Permana et al., 21 Nov 2025).

2. Canonical $\Lambda$-system formulations

A canonical phaseonium model uses a three-level $\Lambda$ configuration with excited state $|a\rangle$ or $|e\rangle$ and two ground or metastable states. In the phase-sensitive photonic-crystal realization, the transitions $\Lambda$0, $\Lambda$1, and $\Lambda$2 are driven respectively by control, probe, and microwave/maser fields, with Rabi frequencies

$\Lambda$3

and

$\Lambda$4

Because the maser closes the loop, the relative phases of the three fields cannot all be gauged away, and the linear response becomes explicitly phase-sensitive (Jha et al., 2012).

Under EIT conditions and in the simplified limit $\Lambda$5, $\Lambda$6, the probe coherence is written as

$\Lambda$7

so the susceptibility

$\Lambda$8

contains a maser-induced term proportional to $\Lambda$9. This is the operational signature of phaseonium in that model: both $\Lambda$0 and $\Lambda$1 depend on the coherence phase $\Lambda$2 (Jha et al., 2012).

In the weak-probe structured-light realization, the medium is instead pre-prepared in

$\Lambda$3

before the optical vector vortex arrives. To first order in the fields, the Maxwell–Bloch equations are

$\Lambda$4

with

$\Lambda$5

Here the off-diagonal terms proportional to $\Lambda$6 directly encode coherence-induced coupling between right- and left-circular components. The phaseonium preparation therefore acts as an intrinsic polarization mixer even in the absence of an additional control field or magnetic bias (Permana et al., 21 Nov 2025).

These two formulations differ in implementation but share the same conceptual structure: phaseonium is the regime in which the complex atomic coherence enters the constitutive response at leading order.

3. Phase-sensitive spectroscopy in photonic crystals

A particularly concrete realization embeds phaseonium layers into a one-dimensional photonic superlattice composed of alternating dielectric and doped layers. In the doped layers, the dielectric function is

$\Lambda$7

and the probe wavenumber is

$\Lambda$8

Because $\Lambda$9 depends on the maser amplitude and phase, the local propagation constant and the entire transfer matrix inherit phase sensitivity (Jha et al., 2012).

The photonic structure is analyzed using Fresnel coefficients for single interfaces, slab formulas for double interfaces, and transfer matrices

$2|l|$0

for the full superlattice. Reflection and transmission then follow from the global matrix. The resulting spectra do not merely reproduce homogeneous-medium EIT; multiple scattering and photonic band-structure effects reshape the quantum resonances into narrow transmission and reflection features (Jha et al., 2012).

The maser controls both the strength and the phase of the phase-sensitive contribution. For $2|l|$1, increasing $2|l|$2 shifts the twin EIT peaks to lower frequency until they merge, after which they split again and their separation grows with $2|l|$3. For $2|l|$4, sharp and tall resonance peaks in both reflection and transmission appear around $2|l|$5. The paper identifies these as constructive interference effects between the standard $2|l|$6-EIT contribution and the maser-induced term in $2|l|$7 (Jha et al., 2012).

The “hyperfine resonances” discussed in that work arise because the microwave field couples the two lower states, interpreted physically as hyperfine levels in realistic atoms such as $2|l|$8Rb. The control field and maser together dress the ground-state manifold, and the probe interrogates transitions to those dressed states. A plausible implication is that phaseonium in a structured photonic environment should be understood as a dressed-state spectroscopy platform as much as an EIT medium: the phase-controlled coherence is amplified by multiple scattering into extremely sharp spectral signatures (Jha et al., 2012).

4. Structured light, anisotropy, and optical spin–orbit coupling

In coherently prepared phaseonium probed by optical vector vortices, the essential effect is a spatial mapping from optical OAM onto matter coherence. The incident field consists of two Laguerre–Gaussian components with opposite SAM and opposite OAM charges $2|l|$9,

$N$0

with

$N$1

For the symmetric resonant case $N$2, $N$3, and balanced input, the analytical solution is

$N$4

$N$5

The optical coherences then inherit the azimuthal dependence $N$6, so the medium acquires a vortex-imprinted coherence pattern (Permana et al., 21 Nov 2025).

The absorption vanishes where $N$7, namely at

$N$8

which gives $N$9 transparency directions within $\Lambda$0. The paper therefore predicts $\Lambda$1-fold azimuthal transparency structures and associated petal-like output intensities: two lobes for $\Lambda$2, four for $\Lambda$3, six for $\Lambda$4, and so forth (Permana et al., 21 Nov 2025).

This is not only a spatial filtering effect. Because the susceptibilities for right- and left-circular components become azimuthally dependent and unequal, the medium is effectively birefringent and dichroic, with a polarization-dependent refractive response tied to the OAM structure. The off-diagonal coherence terms convert RCP into LCP and vice versa, and since these components carry opposite OAM, spin flips are linked to changes in the helical phase sector. The paper identifies the resulting dynamics as optical spin–orbit coupling manifested through SAM exchange, polarization rotation, and evolution of polarization textures described by the Stokes parameters

$\Lambda$5

(Permana et al., 21 Nov 2025).

The dependence on the initial phaseonium preparation is central. Balanced phaseonium $\Lambda$6 tends to symmetrize the two circular components, whereas unequal preparation $\Lambda$7 introduces an intrinsic bias favoring one polarization. On resonance, an unbalanced phaseonium can drive a full polarization flip from LCP to RCP; off resonance, it adds oscillatory SAM exchange. This shows that, in the structured-light context, phaseonium functions as a coherence-programmed anisotropic medium rather than merely a transparent one (Permana et al., 21 Nov 2025).

5. Phaseonium as a non-thermal reservoir

In quantum thermodynamics, phaseonium gas is modeled as a stream of identically prepared three-level atoms whose ground-state coherence changes the effective emission and absorption rates of a cavity mode. In one collision-model formulation, each ancilla has an excited-state population $\Lambda$8, total ground-state population $\Lambda$9, and density matrix

$\Lambda$0

The cavity thermalizes exactly to a Gibbs state with effective temperature

$\Lambda$1

which depends only on the excited-state population and the coherence phase, not on the interaction time $\Lambda$2 (Amato et al., 2023).

That framework also analyzes a cascade of two cavities. The first cavity interacts with fresh atoms and therefore follows a Markovian CPTP dynamics, while the second cavity interacts with atoms already modified by the first cavity and therefore follows a non-Markovian reduced dynamics. Despite this asymmetry, both cavities eventually thermalize to the same $\Lambda$3. By selecting $\Lambda$4, the phaseonium beam can heat or cool the cavities to a designed target temperature, and the rate of approach depends on the collision Rabi phase $\Lambda$5 (Amato et al., 2023).

A related optomechanical engine model uses phaseonium to implement both effectively hot and effectively cold reservoirs. There the apparent temperature is written as

$\Lambda$6

while the incoherent reference temperature is

$\Lambda$7

For $\Lambda$8, one has $\Lambda$9; for phases with $\Lambda$0, the bath is hotter than the incoherent one; for $\Lambda$1, it is colder. The same physical gas can therefore implement both hot and cold non-thermal reservoirs by changing only the coherence phase (Amato et al., 2 Sep 2025).

Within that Otto-engine setting, phaseonium drives isochoric heating and cooling strokes, while adiabatic expansion and compression are performed by changing cavity length and frequency. The paper compares the resulting efficiency to the Curzon–Ahlborn benchmark

$\Lambda$2

defined with respect to corresponding incoherent baths, and reports that for favorable phases the efficiency reaches about

$\Lambda$3

whereas for less favorable phases it can drop to about $\Lambda$4 (Amato et al., 2 Sep 2025).

The two thermodynamic models use different explicit parametrizations of the phase dependence, one through $\Lambda$5 and the other through $\Lambda$6. This suggests that the detailed formula is model dependent, whereas the central physical statement is robust: phaseonium coherence changes the apparent temperature of the reservoir without being reducible to a change in populations alone (Amato et al., 2023, Amato et al., 2 Sep 2025).

6. Multilevel phaseonium, scaling laws, and constraints

The multilevel generalization replaces the original three-level phaseonium by an $\Lambda$7-level atom with one excited state $\Lambda$8 and $\Lambda$9 lower states $|a\rangle$0. The lower manifold contains

$|a\rangle$1

pairwise coherences, each with amplitude and phase. In the symmetric degenerate case, the paper assumes equal lower-state populations and phase-locked coherences $|a\rangle$2, $|a\rangle$3, so the coherent contribution to the cavity dynamics scales with the number of coherence links (Türkpençe et al., 2015).

For the photonic Carnot engine, the cavity mean photon number obeys a coarse-grained equation of the form

$|a\rangle$4

where coherence enters only through

$|a\rangle$5

In the degenerate high-temperature limit and with dephasing factor $|a\rangle$6, the paper gives

$|a\rangle$7

so the coherence contribution is proportional to $|a\rangle$8, while the cavity-loss term is proportional to $|a\rangle$9 (Türkpençe et al., 2015).

The central result is that efficiency and work scale quadratically with the number of coherent lower levels. This is presented as a distinctive quantum-coherence effect, since the number of pair coherences itself scales as $|e\rangle$0. The paper also states that multilevel phaseonium can overcome decoherence in realistic resonator settings where the original three-level phaseonium could not (Türkpençe et al., 2015).

At the same time, the literature emphasizes that phaseonium is not a free thermodynamic resource. Preparing the lower-manifold coherence has an energetic cost. For the multilevel fuel, the preparation energy to reach steady state is written as

$|e\rangle$1

with an estimate

$|e\rangle$2

The work per cycle remains much smaller than this coherence-preparation cost, so the second law is not violated once the full resource accounting is included (Türkpençe et al., 2015).

Experimental feasibility is discussed across the literature in terms of cavity QED with Rydberg atoms, circuit QED, optical Fabry–Pérot resonators, and optomechanical cavities, together with coherent-state preparation methods such as STIRAP, fractional STIRAP, f-SCRAP, Morris–Shore transformations, and quantum Householder reflections (Amato et al., 2023, Amato et al., 2 Sep 2025, Türkpençe et al., 2015). The main limitations repeatedly identified are atomic dephasing, cavity loss, control of the coherence phase, and in multi-cavity or correlated setups the nontrivial definition of work and generalized efficiency bounds (Amato et al., 2 Sep 2025, Türkpençe et al., 2015).

Phaseonium gas therefore occupies a distinctive position at the interface of quantum optics and quantum thermodynamics. In spectroscopy and photonics it is a phase-sensitive coherent medium whose susceptibility can be shaped by prepared or driven ground-state coherence. In structured-light propagation it is a medium that converts atomic coherence into anisotropy, azimuthal transparency, and spin–orbit coupling. In thermodynamic models it is a coherence-engineered reservoir with a tunable apparent temperature. Across these settings, the unifying principle is the same: the phase of atomic coherence is elevated from an incidental quantity to a functional control parameter (Jha et al., 2012, Permana et al., 21 Nov 2025, Amato et al., 2023, Amato et al., 2 Sep 2025, Türkpençe et al., 2015).