---
title: 'Noncyclic Pancharatnam Phase: Geometric Insights'
url: https://www.emergentmind.com/topics/noncyclic-pancharatnam-phase
type: topic
---

# Noncyclic Pancharatnam Phase: Geometric Insights

Searching arXiv for relevant papers on noncyclic Pancharatnam phase and closely related formulations.
Noncyclic Pancharatnam phase is the geometric phase associated with evolution between distinct, nonorthogonal rays rather than with a strictly cyclic return of the state. In polarization language it is defined by the phase of the overlap between initial and final states, $\Phi_P=\arg\langle i|f\rangle$, while in geometric language an open path in ray space is closed by the short geodesic segment on the Poincaré or Bloch sphere, so that the phase is expressed through half of the corresponding solid angle or enclosed area [1510.03799][2211.08519]. This open-path construction underlies a wide range of phenomena, including polarization interferometry, matter-wave condensates, intensity interferometry, generalized measurements, Bloch-band transport, and entanglement diagnostics [1709.00533][1909.00818][1509.04310].

## 1. Geometric meaning of the open-path phase

The operational starting point is Pancharatnam’s in-phase criterion: two states are “in phase” when $\arg(\psi_1^\dagger\psi_2)=0$, so that their superposition interferes maximally. For a noncyclic transformation, the phase between nonorthogonal initial and final states is then
$$
\Phi_P=\arg\langle i|f\rangle,
$$
with $|f\rangle=U|i\rangle$ in the unitary polarization setting [1510.03799][1312.0980].

Its standard geometric interpretation uses the Poincaré or Bloch sphere. In one formulation, when the polarization state of light goes along a closed contour on the Poincaré sphere, the acquired phase is half the enclosed solid angle,
$$
\gamma_{\rm PB}=\frac{\Omega}{2},
$$
and for a noncyclic evolution the same phase is obtained by closing the path geometrically with the short geodesic segment [1709.00533]. In another formulation, the accumulated phase is equal to minus one half of the area enclosed by the actual trajectory and the shortest geodesic joining the endpoints on the Bloch sphere [2409.19692]. The coexistence of $\Omega/2$ and $-\omega/2$ formulas in this literature suggests that orientation and sign convention are intrinsic to the chosen construction rather than to the existence of the phase itself.

A more explicitly spherical treatment writes the noncyclic phase in terms of angular excess. For polarized beams split into states $A$, $B$, and $C$, one has
$$
2\delta=\angle BAC+\angle ACB+\angle ABC-\pi,
$$
so the phase is half the area of the corresponding spherical triangle. In the limiting case where the triangle degenerates, the phase becomes half the area of a lune,
$$
\Delta=\pi-\angle C_0AC^{\prime}=\angle C_0AC,
$$
again emphasizing that the open-path phase is governed by spherical geometry rather than by dynamical recurrence [1411.5603].

## 2. Formal definitions and mathematical frameworks

A general kinematic formulation separates geometric and dynamical contributions. For a ray-space curve $\mathrm C$ lifted to a Hilbert-space path $\mathcal C$, the geometric phase is
$$
\phi_{\mathrm g[\mathrm C]}=\arg\big(\psi(s_1)^\dagger\psi(s_2)\big)-\operatorname{Im}\int_{s_1}^{s_2} ds\,\frac{\psi^\dagger\dot\psi}{\psi^\dagger\psi},
$$
or equivalently $\phi_{\mathrm g[\mathrm C]}=\phi_{\mathrm t[\mathcal C]}-\phi_{\mathrm d[\mathcal C]}$. In the horizontal lift, where $\operatorname{Im}(\psi^\dagger\dot\psi)=0$, the geometric phase reduces directly to the overlap phase, while for an open path it remains equal to minus half the area bounded by the actual trajectory and the geodesic joining its endpoints [1312.0980].

For discrete noncyclic sequences the natural invariant is a Bargmann product. Given
$$
\mathcal C=\{\ket{\psi_0},\ket{\psi_1},\ldots,\ket{\psi_M}\},
$$
the Pancharatnam phase is
$$
\eta_{\mathcal C}=\Im\log\!\Big(\braket{\psi_0}{\psi_1}\braket{\psi_1}{\psi_2}\cdots\braket{\psi_M}{\psi_0}\Big).
$$
This is the discrete analog of Berry’s phase, and the construction via null-phase curves and continuous piecewise null-phase curves makes the discrete Pancharatnam phase equal to the Berry phase of a suitable interpolating continuous path [2312.04500].

The same logic extends to generalized measurements. For a measurement sequence with Kraus operators $M^{(j)}_{r_j}$, the generalized phase is
$$
\chi_{\{r_j\}}=\arg\,\langle \psi_0|M^{(N)}_{r_N}\cdots M^{(1)}_{r_1}|\psi_0\rangle,
$$
and for Hermitian Kraus operators it admits the geometric interpretation
$$
\chi_{\{r_j\}}=\frac{\Omega}{2}.
$$
Here the final state need not coincide with the initial state; the phase is defined through the overlap after the nonunitary sequence, closed conceptually by a “fake” projection onto the initial state [2211.08519].

In periodic lattices, the same open-path correction appears as a boundary term. The Pancharatnam-Zak phase is
$$
\gamma_g(n)=\mathrm{Arg}\,\langle u_n(0)|u_n(2\pi/a)\rangle+i\int_0^{2\pi/a}dq\,\langle u_n(q)|\partial_q|u_n(q)\rangle,
$$
which supplements the usual Zak integral by the endpoint overlap. In the formulation given, this restores gauge invariance and independence of the unit-cell origin, because the boundary Pancharatnam term cancels the gauge- and origin-dependent part of the line integral [1909.00818].

## 3. Interferometric access and phase readout

The noncyclic Pancharatnam phase is directly observable in first-order interference. For two interfering states $\lvert i\rangle$ and $\lvert f\rangle$,
$$
I=\left|e^{i\phi}\lvert i\rangle+\lvert f\rangle\right|^2
=2+2\left|\langle i|f\rangle\right|\cos\!\left(\phi-\arg\langle i|f\rangle\right),
$$
so the fringe displacement yields $\arg\langle i|f\rangle$. In the polarization experiments based on arbitrary $SU(2)$ transformations, the same framework yields the explicit visibility relation $v=\cos\beta$, and robust readout is achieved by feeding the interferometer with two copropagating beams that are orthogonally polarized with respect to each other, so that common mechanical and thermal disturbances largely cancel in the relative fringe shift [1510.03799].

A second interferometric route appears in condensates of indirect excitons. There the measured interference pattern is
$$
I_{\rm interf}({\bf r})=
\left|\psi({\bf r}-\delta{\bf r}/2)+e^{iq_ty}\psi({\bf r}+\delta{\bf r}/2)\right|^2,
$$
with
$$
q_t=\frac{2\pi\alpha}{\lambda}.
$$
From this one extracts the interference amplitude $A_{\rm interf}(x,y)$ and phase $\phi_{\rm interf}(x,y)$. The reported signature is a sharp phase shift in the fringes at essentially the same radius where spontaneous coherence onsets and the linear polarization pattern changes into a helical texture, leading to the emphasized relation
$$
r_{\rm phase}\approx r_{\rm linear}.
$$
The phase shift is interpreted as a Pancharatnam-Berry phase acquired through coherent spin precession in the condensate [1709.00533].

Generalized-measurement realizations also use interferometric readout. In the optical implementation based on null weak measurements, the accumulated phase appears in the probabilities
$$
P_{0/1}=\frac12\left[1\pm {\rm Re}\,e^{-i\delta}\langle \psi_0|M_-^{(N)}\cdots M_-^{(1)}|\psi_0\rangle\right],
$$
so an interference shift directly reveals the measurement-induced geometric phase [2211.08519].

## 4. Physical realizations

The noncyclic Pancharatnam phase is realized in several distinct physical settings. In all of them, the phase enters observables through overlap factors, polarization or spin transport, or geodesically closed state sequences.

| Platform | Phase construction | Observable |
|---|---|---|
| Polarization optics | $\Phi_P=\arg\langle i|U|i\rangle$ for arbitrary $SU(2)$ transformations | Fringe shift and visibility [1510.03799] |
| Indirect-exciton condensate | Poincaré-sphere polarization evolution from coherent spin precession | Shift of IX interference fringes and momentum jump [1709.00533] |
| Two-photon HBT optics | Projection loop $|R\rangle\to|3\rangle\to|L\rangle\to|4\rangle\to|R\rangle$ | Coincidence correlation only [1002.1547] |
| QSHE edge states with SPEs | Spinor-overlap products forming Bloch-sphere loops | Current and cross-correlated noise [1710.05266] |
| Surface sound waves | Loop $A\to B\to C^\pm\to B'\to A$ on an acoustic Poincaré sphere | Direction-dependent acoustic PB phase [2408.03513] |
| Dual Stern–Gerlach interferometers | $\Phi(t)=\mathrm{Arg}[\langle \Psi(0)|\Psi(t)\rangle]$ | Phase jump or continuous phase, depending on gravity model [2409.19692] |

In the indirect-exciton system, the notable feature is that the phase shift appears exactly where IX coherence onsets, is tightly correlated with a change in polarization texture, and can be translated into an effective momentum shift. Using fringe displacements, the momentum map is extracted from
$$
k_x(x,y)=-\frac{2\pi}{D}\frac{\delta y_N(x,y)}{\delta x},\qquad
k_y(x,y)=-\frac{2\pi}{D}\frac{\delta y_N(x,y)}{\delta y},
$$
and the observed momentum jump is of the same order as the estimate
$$
k_{\rm PB}\sim \frac{\Omega}{2l}\sim 5~\mu{\rm m}^{-1}
$$
obtained from one polarization cycle on the Poincaré sphere [1709.00533].

In two-photon intensity interferometry, the phase is explicitly nonlocal: local detector counts do not contain it, while the normalized coincidence correlation does,
$$
{\cal C}=\frac{3}{2}+\frac{1}{2}\cos\!\left[\vec d_D\cdot(\vec k_2-\vec k_1)+\frac{\Omega}{2}\right].
$$
The phase is controlled by the relative detector polarization angle and is described as an optical analog of the multiparticle Aharonov-Bohm effect [1002.1547].

In the QSHE proposal, the phase is a spin-geometric phase generated locally in spin space rather than by real-space orbital circulation. For three spinors,
$$
Z=\langle n_1|n_2\rangle\langle n_2|n_3\rangle\langle n_3|n_1\rangle
= z\,e^{i\Omega/2},
$$
and the resulting phase oscillations appear in current and especially in cross-correlated noise. In the two-SPE geometry, the relevant quantity becomes a quadrilateral loop on the Bloch sphere, giving a genuine multi-electron Pancharatnam phase that survives orbital dephasing [1710.05266].

In acoustics, the phase arises for surface sound waves carrying transverse spin. The relevant geometric phase is
$$
\Phi_{\mathrm{PB}}^{\pm}=\frac{\Omega^{\pm}}{2},
$$
with the two propagation directions corresponding to different points $C^\pm$ on the acoustic Poincaré sphere. Because of spin-momentum locking, $\Phi_{\mathrm{PB}}^{-}$ can cover the full $2\pi$ range while $\Phi_{\mathrm{PB}}^{+}$ cannot, and this asymmetry is used for nearly arbitrary wavefront manipulation of surface sound waves [2408.03513].

## 5. Topological, many-body, and information-theoretic extensions

Beyond single-particle interferometry, the noncyclic Pancharatnam phase functions as a trajectory invariant in measurement-driven dynamics. In the generalized-measurement experiment on a single qubit, the family of phases obeys
$$
\Delta\chi=\chi(\pi)-\chi(0)=2\pi m,
$$
with
$$
\Delta\chi_{\eta\to0}=0,\qquad \Delta\chi_{\eta\to\infty}=2\pi.
$$
The transition between these values occurs at a critical measurement strength where the interference contrast vanishes and the phase becomes ill-defined, providing an experimentally realized topological transition of a generalized Pancharatnam-Berry phase [2211.08519].

In Floquet quantum error-correcting codes, the Pancharatnam phase becomes a noncyclic invariant of a many-body measurement trajectory. For the $\mathbb Z_2$ Floquet toric code, the trajectory phase is tied to the logical action
$$
\mathcal M[\phi]=e^{i\phi}(H_a\otimes H_b)\circ \mathrm{SWAP},
$$
whose nontrivial sector carries a universal $\pi$ shift. The phase can be extracted through a computationally assisted interferometry protocol using the measurement record for error correction and gauge matching, and it matches the Berry phase of an associated continuous gapped unitary evolution [2312.04500].

The same open-path phase also serves as a diagnostic of nonclassical correlations. For bipartite local evolution, the Pancharatnam phase deficit is
$$
\Delta=\Phi_T^{AB}-[\Phi_T^A+\Phi_T^B].
$$
It vanishes for product states because the global overlap factorizes, whereas a nonzero value is a sufficient condition for entanglement. In the examples discussed, the deficit detects macroscopic superpositions of coherent states and can even be related directly to concurrence for distant boundary spins under appropriate local phases [1509.04310].

A further extension uses the noncyclic Pancharatnam phase as an optimization criterion in postselected metrology. There the key condition is
$$
\Theta_{\parallel}=\Theta_{\perp}=\operatorname{Im\,ln}\langle \hat{O}_{\lambda}^{P}\rangle,
$$
which aligns the postselection phase with the intrinsic Pancharatnam phase of the meter channel, suppresses parallel evolution, and maximizes usable orthogonal quantum Fisher information. The reported comparison includes an approximately $83.33\%$ reduction of observable QFI for a nonoptimized comparison channel and more than a tenfold improvement in information retention per trial for the optimized Pancharatnam-phase design [2508.13934].

## 6. Distinctions, limitations, and recurrent misunderstandings

The noncyclic Pancharatnam phase should be distinguished from the more familiar cyclic Berry phase, even though the two are often discussed together under the label “Pancharatnam-Berry phase.” A cyclic polarization evolution generates a geometric phase that enters as a relative phase in non-separable polarization–OAM states and modulates Bell-CHSH correlations; the broader lesson is that a geometric phase changes observable correlations unless the measurement basis is adjusted to compensate, but that particular construction is explicitly cyclic rather than noncyclic [1605.05478].

It should also be distinguished from the complete geometric phase of a nonunitary transformation. In the quantum-kinematic treatment of polarizing processes, a light wave passing through a polarizer may acquire a nonzero complete geometric phase even when the initial and final polarization states are in phase according to the Pancharatnam criterion and therefore show no interferometric Pancharatnam phase. Total-reflection-based polarizers are the key example: the ray-space path is a loxodrome rather than a geodesic, so the complete geometric phase is nonzero even though ordinary interferometric superposition does not reveal it [1312.0980].

A separate caution concerns correlation functions that contain Bargmann-like or solid-angle-like factors. In the classical analysis of Hanbury-Brown-Twiss correlations with polarizers, the phase
$$
\arg\Delta_4=-\frac{\Omega}{2}
$$
appears in a fourth-order polarization trace, but the argument advanced there is that this is not, strictly speaking, a genuine Pancharatnam phase because no physical polarization state is being transported through a meaningful ray-space path; the phase is an algebraic property of the correlation function, and its solid-angle form depends sensitively on source statistics [1611.08071].

Finally, conventional metasurface formulas such as $\Phi_{\mathrm{PB}}=\pm 2\theta$ describe a special symmetry-restricted setting. In the non-axisymmetric metasurface analysis, the equal-magnitude opposite-sign rule for orthogonal circular polarizations is traced to axisymmetry of the meta-atom, and non-axisymmetric structures permit different phase offsets in the two circular channels. This does not redefine the noncyclic Pancharatnam phase, but it does show that standard textbook PB behavior is only one geometric-phase limit among several broader constructions [2301.01118].

Source: https://www.emergentmind.com/topics/noncyclic-pancharatnam-phase