Wielding Singularities: How Moving Exceptional Points Control Wave Behavior

Author: Denis Avetisyan

New research reveals that manipulating exceptional points within band structures allows for precise control over the switching of wave functions based on momentum and parameter tuning.

The evolution of extended parameter space trajectories traced by exceptional points (EPs) defines closed loops in momentum space, geometrically organizing regions where crystal momenta either undergo an eigenmode switch <span class="katex-eq" data-katex-display="false">\psi_{1} \rightarrow \psi_{2}</span> after a single cycle or return to their initial state <span class="katex-eq" data-katex-display="false">\psi_{0} \rightarrow \psi_{0}</span>, thus demonstrating how EP motion dictates momentum-space switching domains. — The evolution of extended parameter space trajectories traced by exceptional points (EPs) defines closed loops in momentum space, geometrically organizing regions where crystal momenta either undergo an eigenmode switch $\psi_{1} \rightarrow \psi_{2}$ after a single cycle or return to their initial state $\psi_{0} \rightarrow \psi_{0}$ , thus demonstrating how EP motion dictates momentum-space switching domains.

Mobile exceptional points generate momentum-space switching domains in non-Hermitian band systems, dictating eigenmode behavior via cyclic modulation.

While non-Hermitian systems offer a route to manipulate band topology through exceptional points (EPs), the consequences of dynamically modulating these degeneracies remain largely unexplored. In the work ‘Mobile Exceptional Points Generate Momentum-Space Switching Domains’, we demonstrate that driving EPs in a band structure generates momentum-space switching domains-regions within the Brillouin zone where eigenmodes exchange character with periodic parameter modulation. These switching domains are dictated by the projection of EP trajectories in an extended parameter space, leading to global band switching as modulation strength increases. Could this EP-driven control over eigenmode behavior unlock new functionalities in photonic and other non-Hermitian materials?

Beyond Hermitian Constraints: Exploring the Realm of Non-Hermitian Physics

For generations, the foundations of physics have been built upon the principle of Hermitian Hamiltonians – mathematical operators ensuring that physical quantities like energy remain real and measurable. However, this formalism inherently struggles to accurately depict open systems – those that exchange energy and matter with their surroundings. Real-world scenarios, such as lasers with optical gain, decaying particles experiencing loss, or even biological systems constantly interacting with their environment, defy this closed-system constraint. The limitation arises because Hermitian Hamiltonians demand energy conservation, precluding the natural gain or loss inherent in these dynamic processes. Consequently, a more versatile framework is required – one that can accommodate complex potentials and describe the intriguing phenomena arising when energy is neither strictly conserved nor lost, paving the way for the exploration of non-Hermitian physics.

Conventional physics often employs Hermitian Hamiltonians to describe the energy of a system, but this approach struggles with scenarios involving gain and loss – crucial in open systems where energy isn’t conserved. Non-Hermitian physics provides an alternative framework, utilizing Hamiltonians that are not self-adjoint, allowing for the modeling of these complex interactions. This shift unlocks a range of unique spectral properties, most notably the appearance of complex energy eigenvalues, which don’t simply correspond to energy but also describe rates of growth or decay. Consequently, non-Hermitian systems can exhibit phenomena unseen in their Hermitian counterparts, such as unidirectional propagation of waves or enhanced sensitivity to perturbations, making them relevant to areas like lasers, metamaterials, and even quantum biology. These systems aren’t merely mathematical curiosities; they represent a powerful tool for understanding and manipulating open quantum systems.

Exceptional Points represent a radical departure from conventional physics, arising within non-Hermitian systems as singularities where the usual requirement of distinct eigenvalues breaks down. At these points, eigenvalues coalesce and the corresponding eigenvectors become indistinguishable, leading to a loss of observability and a breakdown of the standard perturbation theory. This isn’t simply a mathematical curiosity; Exceptional Points dictate drastically altered system behavior, manifesting as enhanced sensitivity to perturbations and unidirectional wave propagation. Unlike traditional bifurcations where systems smoothly transition, Exceptional Points represent a qualitative change in the nature of the system, opening doors to novel device functionalities and offering a new lens through which to view phenomena ranging from laser physics to topological insulators. The behavior around these singularities challenges fundamental assumptions about stability and symmetry, prompting a re-evaluation of established physical principles and driving innovation in areas like sensing and signal processing.

Trajectories of exceptional points (EPs) in extended parameter space (<span class="katex-eq" data-katex-display="false">k_x</span>, <span class="katex-eq" data-katex-display="false">k_y</span>, φ) reveal two branches (blue and purple) that merge and re-emerge at pair-annihilation/creation points, coinciding with switching-domain boundaries of the band-permutation invariant <span class="katex-eq" data-katex-display="false">W(\mathbf{k})</span> as the cyclic control parameter φ varies. — Trajectories of exceptional points (EPs) in extended parameter space ( $k_x$ , $k_y$ , φ) reveal two branches (blue and purple) that merge and re-emerge at pair-annihilation/creation points, coinciding with switching-domain boundaries of the band-permutation invariant $W(\mathbf{k})$ as the cyclic control parameter φ varies.

Sculpting Complex Spectra: The Power of Cyclic Modulation

Cyclic parameter modulation involves the periodic alteration of a quantifiable system characteristic, such as gain, loss, or refractive index. In non-Hermitian systems, which lack the symmetry constraints of Hermitian systems, this modulation effectively creates a time-dependent potential. This dynamic alteration influences the system’s energy landscape, allowing traversal of regions otherwise inaccessible under static conditions. The frequency and amplitude of the modulation directly impact the rate and extent of navigation within this landscape, and can be tailored to control transitions between different system states and manipulate the flow of energy. The modulation does not change the underlying static Hamiltonian, but rather provides a pathway to explore the complex energy surface defined by that Hamiltonian.

Cyclic parameter modulation enables transitions around Exceptional Points (EPs) in non-Hermitian systems, resulting in the formation of Momentum Space Switching Domains (MSSDs). These MSSDs are characterized by spatially defined regions within the system’s momentum space where the characteristics of the eigenmodes – specifically, their energy and spatial distribution – undergo abrupt changes. The size and location of these MSSDs are directly dependent on the modulation frequency and amplitude relative to the separation between EPs, and the system’s underlying Hamiltonian. Consequently, manipulating the modulation parameters allows for precise control over the boundaries and properties of these switching regions, influencing the system’s overall response and potentially enabling novel functionalities.

The formation and characteristics of Momentum Space Switching Domains are directly influenced by the underlying topological properties of the non-Hermitian system. Specifically, the system’s band structure topology – including features like Weyl points or nodal lines – dictates the size, shape, and stability of these domains. Crucially, manipulating the parameters governing cyclic modulation allows for control over energy flow within the system; by precisely tailoring the modulation, it becomes possible to direct energy preferentially through specific pathways defined by the switching domains, effectively acting as an optical or acoustic switch. This control is predicated on the domains’ ability to selectively support or suppress certain momentum-space modes, altering the direction and magnitude of energy transport and enabling functionalities such as unidirectional transmission or energy localization.

Mobile exceptional points (EPs) induce momentum-space switching domains-regions in the Brillouin zone characterized by band-permutation invariance <span class="katex-eq" data-katex-display="false">W(\mathbf{k})</span>-that expand and merge with increasing modulation amplitude μ, eventually covering nearly the entire Brillouin zone and aligning with the projected EP trajectories. — Mobile exceptional points (EPs) induce momentum-space switching domains-regions in the Brillouin zone characterized by band-permutation invariance $W(\mathbf{k})$ -that expand and merge with increasing modulation amplitude μ, eventually covering nearly the entire Brillouin zone and aligning with the projected EP trajectories.

Revealing the Topology: Band Structure and Switching Domains

Cyclic modulation of a lattice system introduces time-periodic driving, which can significantly alter the electronic band structure and, consequently, the band topology. This modulation creates momentum space switching domains – regions in reciprocal space where the effective Hamiltonian undergoes qualitative changes. These switching domains are characterized by alterations in the band connectivity and can induce topological transitions, potentially leading to the emergence of novel topological phases not present in the static, unmodulated system. The degree of these topological modifications is directly related to the characteristics of the cyclic modulation and the resulting behavior of the system’s energy bands in momentum space, potentially giving rise to protected edge states or other hallmark features of topological materials.

The Two-Band Non-Hermitian Lattice Model serves as a mathematically tractable system for analyzing the effects of cyclic modulation on band topology. This model, typically described by a $2 \times 2$ Hamiltonian with complex potentials, allows for the explicit calculation of energy bands and their associated eigenstates as a function of crystal momentum. By varying the modulation parameters within this framework, researchers can observe how the band structure evolves, including the emergence or annihilation of topological features such as Dirac points or exceptional points. Furthermore, the simplicity of the model facilitates the computation of relevant topological invariants and the characterization of eigenmode switching, providing a direct connection between modulation parameters and topological phase transitions.

The Band-Permutation Invariant is a quantitative metric used to assess topological changes induced by cyclic modulation within a band structure. This invariant functions as a binary indicator of eigenmode switching; a value of 0 signifies no switching has occurred between eigenstates during modulation, indicating a topologically trivial alteration. Conversely, a value of 1 denotes the presence of eigenmode switching, which is a characteristic feature of non-trivial topological transitions and the emergence of novel topological phases. The calculation of this invariant relies on tracking the permutation of eigenstates across the Brillouin zone as the modulation parameters are varied, providing a clear and concise measure of topological alteration.

Eigenvalue trajectories within a switching domain exhibit braiding in the <span class="katex-eq" data-katex-display="false">$\\mathrm{Re}E,\\mathrm{Im}E,\\phi\$</span> space, indicating an exchange of eigenmodes after one modulation cycle (<span class="katex-eq" data-katex-display="false">\\phi:0\\rightarrow 2\\pi</span>), while trajectories outside the domain simply return to their original branches. — Eigenvalue trajectories within a switching domain exhibit braiding in the $\\mathrm{Re}E,\\mathrm{Im}E,\\phi\$ space, indicating an exchange of eigenmodes after one modulation cycle ( $\\phi:0\\rightarrow 2\\pi$ ), while trajectories outside the domain simply return to their original branches.

Realizing Non-Hermitian Topology: A Path Towards Novel Devices

Photonic crystals, structures engineered to control the flow of light, are proving to be a fertile ground for realizing non-Hermitian physics – a realm of quantum mechanics traditionally difficult to observe directly. Incorporating lossy dielectric materials into these crystals introduces dissipation, mimicking the effects of particle decay or environmental interactions that define non-Hermitian systems. This deliberate introduction of loss isn’t a hindrance, but rather a crucial design element, allowing researchers to sculpt the behavior of light within the crystal. The resulting structures exhibit unique properties, such as unidirectional light propagation and enhanced sensitivity to external perturbations, opening doors for novel optical devices and a deeper understanding of fundamental physics. This approach circumvents many of the challenges associated with creating non-Hermitian systems in traditional materials, providing a robust and accessible platform for both experimental investigation and potential technological applications.

The creation of non-Hermitian topological systems relies heavily on the precise manipulation of photonic structures. Researchers are leveraging the unique properties of light within these engineered materials-specifically, incorporating lossy dielectric materials into photonic crystals-to mimic complex quantum phenomena. This careful design allows for the observation of predicted topological behaviors, such as the formation of exceptional points and the associated switching dynamics. By controlling the geometry and composition of the photonic crystal, it becomes possible to tailor the flow of light in unconventional ways, creating platforms where fundamental concepts from quantum mechanics can be experimentally verified and potentially harnessed for novel optical devices. The ability to dynamically adjust these structures opens avenues for controlling light propagation and realizing functionalities not achievable in traditional photonic systems.

The hallmark of topological transitions within these non-Hermitian systems lies in a phenomenon called eigenvalue braiding, where the eigenvalues of the system intertwine as external parameters are varied. This intricate dance of eigenvalues isn’t merely a mathematical curiosity; it provides a direct, experimentally observable signature of the underlying topological change, detectable through spectroscopic measurements. Investigations reveal that specific modulation amplitudes induce distinct behaviors: an amplitude of $\mu = 0.4$ generates two exceptional points (EPs), critical points where the system’s properties dramatically change, while a significantly larger amplitude of $\mu = 2.7$ results in a fully gapped spectrum, effectively eliminating the EPs and signifying a fundamentally different topological state. This precise control over the spectrum, achieved through parameter modulation, allows researchers to navigate and characterize these complex non-Hermitian phases of matter.

Complex-energy braiding reveals the topological properties of the band structure, demonstrating the emergence of exceptional points (black) and branch cuts (red/green) connected to the vanishing of the real and imaginary parts of the band gap at representative momentum points for <span class="katex-eq" data-katex-display="false">\mu = 0.4</span> and <span class="katex-eq" data-katex-display="false">\mu = 2.7</span>, calculated for <span class="katex-eq" data-katex-display="false">t_x = 1</span>, <span class="katex-eq" data-katex-display="false">t_y = 0.5</span>, and <span class="katex-eq" data-katex-display="false">\epsilon_0 = 1.2</span> with <span class="katex-eq" data-katex-display="false">\phi = \pi/2</span>. — Complex-energy braiding reveals the topological properties of the band structure, demonstrating the emergence of exceptional points (black) and branch cuts (red/green) connected to the vanishing of the real and imaginary parts of the band gap at representative momentum points for $\mu = 0.4$ and $\mu = 2.7$ , calculated for $t_x = 1$ , $t_y = 0.5$ , and $\epsilon_0 = 1.2$ with $\phi = \pi/2$ .

Beyond Current Designs: Charting a Course for Future Innovation

The convergence of non-Hermitian physics, cyclic modulation, and band topology presents a remarkably fertile ground for advancements in condensed matter physics and photonics. Non-Hermitian systems, characterized by effective Hamiltonians that lack symmetry, exhibit unique spectral properties and sensitivity to external perturbations. When coupled with cyclic modulation – the periodic alteration of system parameters – these properties become dynamically tunable, leading to novel phenomena such as non-reciprocal light propagation and topologically protected states that are robust against disorder. Crucially, the underlying band topology – the mathematical description of electronic energy bands – dictates the emergence of these protected states and influences the system’s overall response. Further investigation into this interplay promises the realization of devices with functionalities unattainable in traditional Hermitian systems, potentially revolutionizing fields like signal processing, sensing, and lasing through the precise control of light and matter at the nanoscale.

A complete understanding of the complex energy spectrum in non-Hermitian systems, as revealed through the Riemann Surface representation, promises a paradigm shift in their control and application. Traditionally, energy is understood as a real value; however, non-Hermitian systems exhibit complex energies, where the imaginary component dictates decay or gain. The Riemann Surface provides a visual and mathematical framework to map these complex energies, revealing hidden connections and symmetries not apparent in conventional energy plots. This allows researchers to move beyond simply observing non-Hermitian phenomena and instead proactively engineer systems with desired properties – controlling wave localization, enhancing sensitivity to external stimuli, and designing novel topological states of light or matter. By fully exploiting the information encoded within the Riemann Surface, the potential for creating advanced devices – including highly sensitive sensors and lasers with unique beam characteristics – is significantly expanded, moving these concepts from theoretical curiosities to practical technological innovations.

The principles explored in this study offer a pathway toward a new generation of photonic devices boasting capabilities previously considered unattainable. By harnessing the unique properties of non-Hermitian systems and their complex energy landscapes, researchers anticipate innovations spanning multiple fields. Enhanced sensing applications are envisioned, where subtle changes in the environment can be detected with unprecedented sensitivity due to the system’s amplified responses. Furthermore, the foundations laid by this work could lead to the realization of topological lasers – devices exhibiting robust and directional light emission protected from backscattering and defects. These advances promise not only improved performance in existing photonic technologies but also the creation of entirely new functionalities, potentially revolutionizing fields like optical communications, biomedical imaging, and advanced materials science.

The presence of exceptional points (EPs), indicated by black points, is contingent on the parameter μ, with lower values (μ=0.4) supporting EPs and connected spectral sheets via branch cuts (red/green points), while higher values (μ=2.7) result in a fully gapped spectrum devoid of EPs.

The study illuminates how manipulating non-Hermitian band systems through mobile exceptional points creates predictable, switchable behavior within momentum space. This control over eigenmode transitions, dictated by cyclic parameter modulation, echoes a fundamental tenet of understanding any complex system: recognizing patterns to predict outcomes. As Jean-Jacques Rousseau observed, “The best way to know a thing is to make it.” This research embodies that principle; by actively making exceptional points mobile, the researchers reveal a new mechanism for controlling wave behavior and, consequently, understanding the broader landscape of band topology. The creation of these momentum-space switching domains demonstrates how targeted interventions can induce observable, predictable changes within a system’s inherent structure.

Beyond the Switching Domain

The observation of momentum-space switching domains generated by mobile exceptional points presents a fascinating, if somewhat predictable, analogue to phase transitions in condensed matter physics. Just as a system seeks minimal energy, these non-Hermitian band structures appear to gravitate toward configurations that maximize mode switching sensitivity. However, the precise control needed to reliably position and modulate these exceptional points raises questions about robustness. Real-world implementations-particularly those venturing beyond idealized, perfectly cyclic modulations-will inevitably encounter imperfections. The resulting ‘smearing’ of the exceptional points could lead to incomplete switching or, more subtly, to a statistical distribution of switching probabilities-a move away from deterministic control toward probabilistic computation.

Furthermore, the current work primarily focuses on systems with a single band. Extending this framework to multi-band systems, where interactions between bands could dramatically alter the landscape of exceptional points, remains a significant challenge. One might envision ‘exceptional point networks’-complex topologies of switching domains that could support more sophisticated information processing. The potential for harnessing the complex energy spectrum-a fingerprint of the non-Hermitian nature-for sensing or even energy harvesting remains largely unexplored.

Ultimately, the exploration of non-Hermitian systems is, at its heart, a search for new forms of symmetry and order. The discovery of mobile exceptional points is not merely a quirk of mathematical formalism, but a hint that the rules governing wave phenomena may be more flexible-and potentially more powerful-than previously imagined. The limitations are not in the math, but in the ability to translate these abstract concepts into tangible, controllable devices.

Original article: https://arxiv.org/pdf/2604.27417.pdf

Contact the author: https://www.linkedin.com/in/avetisyan/