Phase-Shifted Order Parameter

• Phase-shifted order parameter is defined by its phase degree of freedom (e.g., spatial shift, phase lag, sign reversal), offering a richer description than scalar amplitude alone.
• It plays a key role in systems ranging from topological chains and superconductors to AC-driven and finite-size arrays by diagnosing phase transitions and ordering patterns.
• Engineered designs and algorithmic methods utilize phase-shifted order frameworks to optimize device performance and reveal quantum critical behavior in complex systems.

A phase-shifted order parameter is an order descriptor whose physically relevant content is carried not only by amplitude but also by a phase degree of freedom, a spatial displacement of its support, or a measurable phase lag in its response. In contemporary literature, the expression appears in several technically distinct senses: as a bond-shifted topological marker in the Su–Schrieffer–Heeger model, as a local $\pi$ shift of the superconducting gap near magnetic impurity wires, as a phase lag between drive and response in AC magnetocaloric measurements, as a phase-shifted angular harmonic in Weyl-semimetal planar transport, and as an order-parameter-like design descriptor in phase-shifted distributed-feedback gratings (Yu et al., 2016, Björnson et al., 2016, Aliev et al., 2023, Bera et al., 25 Jun 2026, Sun et al., 2024). Taken together, these usages suggest that the concept is best understood as a family of constructions in which phase information organizes the ordering pattern, its diagnostics, or its control.

1. Conceptual scope and principal definitions

Several mathematically precise objects play the role of phase-shifted order parameters in different subfields. In finite-size two-dimensional superfluids, the phase order parameter is

$$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$

with $\Phi(\mathbf r)$ the coarse-grained phase field and $\bar{\Phi}$ its spatial average; this is the XY magnetization amplitude in the helium–XY mapping (Bramwell et al., 2015). In topological planar transport of Weyl semimetals, the relevant object is the complex combination

$\mathcal O=\alpha_1+i\alpha_2,$

whose phase gives

$$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$

thereby shifting the canonical $\sin 2\phi$ and $\cos 2\phi$ angular harmonics into $\sin 2(\phi+\phi_r)$ and $\cos 2(\phi+\phi_r)$ (Bera et al., 25 Jun 2026). In a multi-wavelength DFB laser, the distributed set of cavity phase defects is described as

$$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$0

with $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$1 the number of global $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$2 phase shifts, $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$3 their magnitudes, and $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$4 their positions; the paper explicitly treats this set as an order-parameter-like descriptor of the mode structure (Sun et al., 2024).

These definitions share a common structural feature: the relevant order is encoded by phase alignment, phase displacement, or phase-sensitive interference rather than by a single scalar amplitude alone. A plausible implication is that the phrase is most useful when the dominant invariant is a phase, sign, or shifted support that would be invisible to a purely magnitude-based Landau description.

2. Spatially shifted and topological order

In one-dimensional topological systems, a phase-shifted order parameter often denotes a real-space displacement of the operator that diagnoses the phase. For the spinless SSH model, the reduced-density-matrix construction yields a trivial-phase operator

$$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$5

which is intra-cell, and a topological-phase operator

$$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$6

which is inter-cell and therefore shifted by one lattice spacing relative to the unit-cell decomposition (Yu et al., 2016). The same work identifies the topological region by a Berry phase $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$7 and the trivial region by $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$8, so the spatial shift of $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$9 is the local manifestation of a global phase change in the ground state under twisted boundary conditions. The topological non-trivial phase is reported to be robust to repulsive inter-site interactions and can also appear in a topologically trivial parameter region when appropriate interactions are added (Yu et al., 2016).

A closely related but distinct construction appears for one-dimensional symmetry-protected phases. There the order parameter is built from symmetry strings and swaps, and its asymptotic value is the sign of a projective commutator such as

$\Phi(\mathbf r)$0

which distinguishes the trivial and Haldane classes for $\Phi(\mathbf r)$1 symmetry (Haegeman et al., 2012). The normalized order parameter therefore takes quantized values $\Phi(\mathbf r)$2 or $\Phi(\mathbf r)$3, so the physically meaningful change is again a phase shift by $\Phi(\mathbf r)$4, now in virtual-symmetry space rather than in real space (Haegeman et al., 2012).

These examples also delimit a common misconception. The relevant operators are not standard Landau order parameters with ordinary long-range connected correlations. In the SSH case, the connected correlations of the diagnostic operators can vanish even though the operators still distinguish topological sectors; in the SPT case, the order parameter is explicitly string-like and symmetry-based rather than local and symmetry-breaking (Yu et al., 2016, Haegeman et al., 2012).

3. Complex phase, sign reversal, and interferometric order in paired states

In superconducting systems, a phase-shifted order parameter is often literal. For a conventional on-site $\Phi(\mathbf r)$5-wave superconductor,

$\Phi(\mathbf r)$6

and a $\Phi(\mathbf r)$7 phase shift means $\Phi(\mathbf r)$8 (Björnson et al., 2016). In a finite ferromagnetic impurity wire embedded in a two-dimensional $\Phi(\mathbf r)$9-wave superconductor, the self-consistent gap

$\bar{\Phi}$0

is reduced in discrete steps as Yu–Shiba–Rusinov-derived states cross the Fermi level and contribute out of phase with the condensate (Björnson et al., 2016). The paper traces the $\bar{\Phi}$1-shift to a resonance condition in the Bogoliubov–de Gennes spectrum: when non-resonating states localized at impurity sites are pulled into the condensate, the local order parameter on the wire is progressively suppressed and eventually changes sign relative to the bulk (Björnson et al., 2016). With finite Rashba spin–orbit coupling, the stepwise structure becomes smoother because level crossings become avoided crossings, but the $\bar{\Phi}$2-shift is preserved and occurs in a large portion of the topologically non-trivial phase (Björnson et al., 2016).

Ultracold-atom interferometry generalizes the same idea to momentum-space phase structure. For a Feshbach molecule,

$\bar{\Phi}$3

and for BCS states the relevant phases are those of $\bar{\Phi}$4 (Kitagawa et al., 2010). Bragg-pulse beam splitters produce coincidence or density-noise signals proportional to $\bar{\Phi}$5, allowing direct measurement of relative phases between different momentum components of $\bar{\Phi}$6-wave, $\bar{\Phi}$7-wave, and particle-hole order parameters (Kitagawa et al., 2010). The same formalism applies to non-trivial particle-hole order such as a $\bar{\Phi}$8-density wave, where products like $\bar{\Phi}$9 encode the phase difference between two momentum sectors (Kitagawa et al., 2010).

In a more minimal quantum-information setting, a relative phase can play an order-parameter-like role even when the entanglement magnitude is unchanged. For phase-shifted Bell states,

$\mathcal O=\alpha_1+i\alpha_2,$0

the tunable phase $\mathcal O=\alpha_1+i\alpha_2,$1 controls the full correlation landscape and can move the CHSH parameter from near Tsirelson saturation to the classical bound at special quarter-turn values (Davis et al., 16 Jan 2025). This is not a thermodynamic order parameter, but it is a closely analogous instance in which the physically decisive variable is the phase of a coherent superposition rather than its norm.

4. Dynamical and transport phase shifts as order diagnostics

In nonequilibrium and transport settings, the phrase frequently denotes a response phase rather than a static order field. In AC magnetocaloric measurements, a weak sinusoidal magnetic field

$\mathcal O=\alpha_1+i\alpha_2,$2

drives an adiabatic temperature response

$\mathcal O=\alpha_1+i\alpha_2,$3

where $\mathcal O=\alpha_1+i\alpha_2,$4 is the phase lag between drive and response (Aliev et al., 2023). The order-parameter field is written as

$\mathcal O=\alpha_1+i\alpha_2,$5

with relaxational dynamics

$\mathcal O=\alpha_1+i\alpha_2,$6

At a second-order phase transition, the paper sets $\mathcal O=\alpha_1+i\alpha_2,$7 at the critical point and concludes that the phase shift does not depend on magnetic-field magnitude, whereas at a first-order transition finite-lifetime fluctuations produce

$\mathcal O=\alpha_1+i\alpha_2,$8

with strong field dependence and a field-induced shift of the temperature of maximum phase lag (Aliev et al., 2023). The resulting criterion is explicitly diagnostic: absence of field dependence indicates a second-order transition, while field dependence and temperature shift of the maximum phase shift indicate a first-order transition (Aliev et al., 2023).

An analogous but collective phenomenon occurs in multiplex Kuramoto networks with interlayer phase-shifted coupling

$\mathcal O=\alpha_1+i\alpha_2,$9

The standard Kuramoto order parameters,

$$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$0

retain their usual definition, but the self-consistency relations that determine them are modified by the phase lag $$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$1 (Kumar et al., 2021). The interlayer term decomposes into an attractive part $$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$2 and a bias term $$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$3, so increasing $$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$4 suppresses smooth synchronization and can induce explosive synchronization with hysteresis, especially near $$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$5 and for nonzero mirror-node frequency mismatch (Kumar et al., 2021). The paper emphasizes that the phase shift enters the order parameter indirectly, through the dynamics and the relative phase structure rather than by altering the formal definition of $$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$6 (Kumar et al., 2021).

In Weyl semimetals, phase-shifted order appears in angular response tensors. The planar Hall and longitudinal conductivities acquire the forms

$$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$7

$$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$8

with $$\phi_r=\frac{1}{2}\arctan\!\left(\frac{\alpha_2}{\alpha_1}\right),$$9 and $\sin 2\phi$0 (Bera et al., 25 Jun 2026). The additional coefficient $\sin 2\phi$1 is identified as an intrinsic quadratic-in-$\sin 2\phi$2 contribution absent in the conventional semiclassical description, and the same $\sin 2\phi$3 extracted independently from longitudinal and transverse responses quantitatively fits available data, including $\sin 2\phi$4 and $\sin 2\phi$5 in one Cd$\sin 2\phi$6As$\sin 2\phi$7 dataset, $\sin 2\phi$8 in another, and $\sin 2\phi$9, $\cos 2\phi$0, and $\cos 2\phi$1 in PtBi$\cos 2\phi$2 fits (Bera et al., 25 Jun 2026). Here the phase-shifted order parameter is not symmetry breaking but an angular-order descriptor of the planar transport tensor.

5. Finite-size phase order and thermodynamic constraints

In finite two-dimensional superfluids, phase order is intrinsically finite-size. The helium-film analysis maps the system to the 2D-XY model with

$\cos 2\phi$3

and defines

$\cos 2\phi$4

as a finite-size phase order parameter (Bramwell et al., 2015). The key scaling relation is

$\cos 2\phi$5

with $\cos 2\phi$6 (Bramwell et al., 2015). Near the size-dependent transition temperature, the order parameter follows

$\cos 2\phi$7

and the paper reports a universal collapse of helium-film and ferromagnetic-film data under the appropriate finite-size rescaling (Bramwell et al., 2015). This finite-size phase order does not violate the Mermin–Wagner theorem, because the infinite-system order parameter still vanishes for any $\cos 2\phi$8; the nonzero $\cos 2\phi$9 is an effective order parameter induced by the finite integral scale $\sin 2(\phi+\phi_r)$0 (Bramwell et al., 2015).

Thermodynamically constrained mean-field theory provides a complementary amplitude-focused perspective. In the classical limit, imposing continuity at $\sin 2(\phi+\phi_r)$1 and the third law at $\sin 2(\phi+\phi_r)$2 yields

$\sin 2(\phi+\phi_r)$3

together with

$\sin 2(\phi+\phi_r)$4

so the order parameter can be read equivalently from temperature, entropy, or thermal expansion (Santos et al., 2021). In the quantum-mechanical extension, the order parameter becomes

$\sin 2(\phi+\phi_r)$5

which saturates to $\sin 2(\phi+\phi_r)$6 as $\sin 2(\phi+\phi_r)$7 and introduces the soft-mode scale $\sin 2(\phi+\phi_r)$8 below which the quantum ground state is reached (Santos et al., 2021). Applied to SrTiO$\sin 2(\phi+\phi_r)$9, the model fits thermal-expansion data with $\cos 2(\phi+\phi_r)$0, $\cos 2(\phi+\phi_r)$1, and $\cos 2(\phi+\phi_r)$2 (Santos et al., 2021). In this setting the “phase” is not explicit, but the formalism fixes the amplitude envelope that any complex or phase-shifted order parameter would have to obey if it shares the same thermodynamic boundary conditions.

6. Engineered and algorithmic formulations

Phase-shifted order can also be engineered as a design variable or discovered algorithmically from reduced states. In a third-order, four-phase-shifted sampled Bragg grating DFB laser, the global set of true $\cos 2(\phi+\phi_r)$3 phase shifts is treated as an order-parameter-like descriptor of cavity segmentation and defect-mode structure (Sun et al., 2024). For the four-channel device, four $\cos 2(\phi+\phi_r)$4 phase shifts are placed at $\cos 2(\phi+\phi_r)$5, $\cos 2(\phi+\phi_r)$6, $\cos 2(\phi+\phi_r)$7, and $\cos 2(\phi+\phi_r)$8 along a $\cos 2(\phi+\phi_r)$9 DFB section, partitioning the cavity into five subsections; for the seven-channel device, seven $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$00 shifts partition the same total length into eight sub-cavities (Sun et al., 2024). Combined with a linearly chirped sampling period, this yields measured average channel spacings of $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$01 with standard deviation $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$02 for the four-channel laser and $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$03 with standard deviation $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$04 for the seven-channel device, while using a fabrication flow requiring one MOVPE step and a single III–V etch (Sun et al., 2024). The paper’s central point is that the sequence of phase defects, not merely the local grating period, determines how many defect modes are supported and how uniformly they are spaced.

At the opposite end of the spectrum, reduced-fidelity-susceptibility methods discover order parameters without assuming their form in advance. For a two-parameter Hamiltonian, the method constructs the scalar field

$$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$05

from local reduced density matrices, then defines a vector field

$$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$06

whose sources identify quantum critical lines and whose angle field maps phases (Mariella et al., 2024). An optimization problem over few-site Hermitian operators then yields observables whose expectation values are large on one side of a transition and small on the other (Mariella et al., 2024). In benchmark applications, the method recovers an Ising-like observable close to $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$07 in the ANNNI model, reconstructs known five-site string order operators in the cluster SPT model, and identifies local projectors such as $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$08 and $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$09 for the $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$10 and $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$11 Rydberg-crystal phases (Mariella et al., 2024). Its finite-size scaling in the ANNNI model yields $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$12 and $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$13, close to the Ising value $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$14 (Mariella et al., 2024). A plausible implication is that algorithmic discovery extends the phase-shifted-order-parameter paradigm from analytically motivated constructions to data-driven operator synthesis.

Across these settings, the unifying idea is not a single formal definition but a recurring mechanism: phase, sign, spatial displacement, or response lag becomes the decisive carrier of order. In topological chains this appears as a bond shift or projective sign; in superconductors as a local $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$15 inversion of $$\Psi(L,T)=\big\langle \cos\big(\Phi-\bar{\Phi}\big)\big\rangle,$$16; in driven magnetic and synchronization systems as a phase lag; in planar transport as a shifted angular harmonic; in helium films as finite-size phase alignment; and in photonic or many-body design problems as a structured phase-defect configuration that organizes the allowed collective modes (Yu et al., 2016, Björnson et al., 2016, Aliev et al., 2023, Kumar et al., 2021, Bera et al., 25 Jun 2026, Bramwell et al., 2015, Sun et al., 2024, Mariella et al., 2024).