Bright and Dark Excitons in CrSBr: Local Ligand-Field Character and Band-Coherent Optical Selection Rules

arXiv:2607.08355 · cond-mat.mtrl-sci, cond-mat.str-el, physics.comp-ph · Submitted 2026-07-09 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Bright and Dark Excitons in CrSBr".

Mira: Here's the gist of the paper. Magnetic van der Waals semiconductors such as CrSBr host an intricate exciton landscape whose physical interpretation has converged only recently <ref:2607.08355#pg2> Here we show that the brightness is a band-coherent property of the excitonic eigenfunctions: bright and dark partners are sublattice-symmetric and sublattice-antisymmetric superpositions of the same ligand-field-like Bloch transitions across the two Cr atoms of the orthorhombic primitive cell <ref:2607.08355#pg4> This disentangling of bare excitonic structure is a prerequisite for interpreting the optical response of CrSBr once magnon, phonon, and photon couplings are layered on <ref:2607.08355#pg2> Exciton Character and Symmetry The paper addresses the mystery of why intense bright excitons coexist with companion states that are several orders of magnitude darker despite drawing from essentially the same single-particle transition manifold <ref:2607.08355#pg2> Bright and dark partners in CrSBr live in the same four-band block built from the two topmost valence and two lowest conduction branches <ref:2607.08355#pg4>.

Kai: Specifically, bright and dark partners are sublattice-symmetric (“diagonal”) or sublattice-antisymmetric (“cross”) superpositions of the same ligand-field-like Bloch transitions across the two Cr atoms of the primitive cell <ref:2607.08355#pg2>.

Mira: The diagonal combination adds the constituent transition dipoles constructively and is bright, whereas the cross combination adds destructively and is dark <ref:2607.08355#pg4>.

Kai: Optical Selection Rules The optical amplitude is defined by a sum over transition channels, where brightness depends on the coherence with which these constituent Bloch transitions add <ref:2607.08355#pg2> The oscillator strength is then calculated based on this amplitude <ref:2607.08355#pg2>.

Paper summary: Kai: So to summarize this paper, they tackle the mystery of why intense bright excitons exist alongside companion states that are several orders of magnitude darker, even though they all seem to come from essentially the same starting single-particle transition manifold.

Mira: The main claim is that the brightness we measure is actually a band-coherent property tied directly to how these excitonic eigenfunctions combine. Bright and dark partners live in the same four-band block built from two topmost valence and two lowest conduction branches, but they are distinguished by whether they are sublattice-symmetric or antisymmetric superpositions across the two chromium atoms.

Lev: They use the QSGWc theory to look at this, and it shows that the diagonal combination of these constituent Bloch transitions adds constructively to create a bright exciton, while the cross combination adds destructively to create a dark one.

Kai: The authors argue that this symmetry is actually more important than other factors like onsite d-d weight or the distribution of k-space participation ratio when you're trying to explain the brightness hierarchy.

Mira: They show that this band coherence effect applies across both Frenkel-like and Wannier-like excitons, suggesting it’s a universal mechanism for optical selection rules in this system.

Lev: What this means for us is that the optical response you see isn't just a simple sum of local atomic transitions; it’s an interference pattern dictated by the symmetry of how those transitions add up across the lattice.

Kai: They quantify this coherence using an interference figure of merit, I alpha S, which compares the coherent and incoherent sums of per-channel contributions to measure this effect directly.

Mira: They found that when these nearly identical k-space manifolds form these different combinations—diagonal versus cross—you get constructive brightness with a factor I about ten, or destructive brightness that can be as low as one to ten to the power of negative seven.

Lev: If we were building hardware based on this, we’d have to focus our efforts on engineering the crystal structure so that these k-space manifolds couple in the way you want for maximum constructive interference.

Kai: So, they are arguing that the source of brightness in CrSBr isn't purely onsite d-d physics or purely Wannier spatial extent, but rather this coherent optical interference within a ligand-field derived magnetic band structure.

Conclusion: Kai: The title itself really sets up the idea that we need to look at both the local ligand-field character and the band coherence when interpreting these excitons in CrSBr.

Mira: It moves past just labeling them bright or dark based on some simple structural features and grounds it in a specific quantum mechanical description of how those transitions combine across space.

Lev: This paper suggests that we need to stop viewing these as purely Frenkel or purely Wannier things and start seeing them as inherently band-coherent entities whose optical strength depends on their symmetry within the crystal lattice.

Kai: The implication for us is that understanding the optical response of CrSBr, especially when you layer in things like magnon and phonon couplings, requires this microscopic picture of band hybridization and symmetry-adapted transition dipole addition.

Mira: So, we’re looking at a system where the source of brightness isn't just one thing; it’s a complex interplay where the way bands hybridize dictates whether we get strong optical coupling or not.

Lev: It tells us that when designing experiments or materials for magnetic semiconductors, you have to account for this internal phase effect during your modeling, because it’s not just a simple energy gap problem.

Swagata Acharya, * Jessica McDivitt, Dimitar Pashov, Mark van Schilfgaarde, Justin C. Johnson, Jeffrey L. Blackburn

National Laboratory of the Rockies · University of Colorado Boulder · King’s College London

cond-mat.mtrl-sci, cond-mat.str-el, physics.comp-ph

Submitted: 2026-07-09

Updated: 2026-11-08

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 91/100

The gist: Magnetic van der Waals semiconductors such as CrSBr host an intricate exciton landscape whose physical interpretation has converged only recently <ref:2607.08355#pg2> Here we show that the brightness…

Key concepts

Bright and Dark Partners
These are excitonic states in CrSBr that originate from the same single-particle electronic transitions but exhibit vastly different optical strengths. They arise because they are either 'diagonal' (sublattice-symmetric) or 'cross' (sublattice-antisymmetric) superpositions of the same underlying transitions, leading to constructive or destructive interference.
Sublattice Symmetry
In the crystal structure of CrSBr, there are two different sites for the Chromium atoms. Bright and dark excitons are characterized by how their wavefunctions behave across these two sites—either adding constructively (symmetric) or destructively (antisymmetric) when combined, which directly dictates their optical brightness.
Optical Selection Rules
Brightness is determined by the coherence with which the constituent electronic transitions add up. A dark exciton isn't just a state lacking local character; its eigenvector is nearly orthogonal to the optical vector, meaning it interferes destructively with the light, resulting in very low oscillator strength.

Terminology

Summary

Magnetic van der Waals semiconductors such as CrSBr host an intricate exciton landscape whose physical interpretation has converged only recently <ref:2607. p. 008355#pg2> Here we show that the brightness is a band-coherent property of the excitonic eigenfunctions: bright and dark partners are sublattice-symmetric and sublattice-antisymmetric superpositions of the same ligand-field-like Bloch transitions across the two Cr atoms of the orthorhombic primitive cell <ref:2607. p. 008355#pg4> This disentangling of bare excitonic structure is a prerequisite for interpreting the optical response of CrSBr once magnon, phonon, and photon couplings are layered on <ref:2607. p. 008355#pg2>

Exciton Character and Symmetry

The paper addresses the mystery of why intense bright excitons coexist with companion states that are several orders of magnitude darker despite drawing from essentially the same single-particle transition manifold <ref:2607.08355#pg2> Bright and dark partners in CrSBr live in the same four-band block built from the two topmost valence and two lowest conduction branches <ref:2607. p. 208355#pg4>. Specifically, bright and dark partners are sublattice-symmetric (“diagonal”) or sublattice-antisymmetric (“cross”) superpositions of the same ligand-field-like Bloch transitions across the two Cr atoms of the primitive cell <ref:2607. p. 008355#pg2>. The diagonal combination adds the constituent transition dipoles constructively and is bright, whereas the cross combination adds destructively and is dark <ref:2607. p. 208355#pg4>.

Optical Selection Rules

The optical amplitude is defined by a sum over transition channels, where brightness depends on the coherence with which these constituent Bloch transitions add <ref:2607.08355#pg2> The oscillator strength is then calculated based on this amplitude <ref:2607. p. 108355#pg2>. A dark exciton is not simply a state with no allowed local component; it is a state whose eigenvector is nearly orthogonal to the optical vector <ref:2607. p. 508355#pg2>. The interference figure of merit for brightness, IαS, quantifies this coherence by comparing the coherent and incoherent sums of per-channel contributions <ref:2607. p. 508355#pg2>.

Exciton Hierarchy and Limits

The analysis compares excitons between the Frenkel and Wannier limits to understand their physical nature <ref:2607.08355#pg3>. The Frenkel limit is associated with the XA bright exciton, which has strong onsite Cr d-d character, while the XB exciton is closer to a Wannier-like intersite, ligand-assisted limit <ref:2607. p. 108355#pg2>. The paper concludes that CrSBr is neither a simple molecular ligand-field system nor a conventional weakly bound Wannier exciton system but rather a magnetic band insulator where ligand-field-like orbital excitations acquire optical strength through band hybridization and symmetry-adapted transition dipole addition <ref:2607. p. 908355#pg2>.

Key Findings on Brightness Contrast

The most direct evidence for the coherent selection rule comes from comparing near-partner states, such as the 1.33 eV and 1.38 eV excitons, which have almost identical k-space distributions <ref:2607. p. 708355#pg2>. The difference in brightness is not a population effect but an internal phase effect within a nearly identical transition manifold <ref:2607. p. 808355#pg2>. The strongest bright/dark contrasts arise when nearly identical k-space manifolds form different diagonal and cross band-branch combinations, producing constructive (I ∼10) or destructive (I ∼10−4 to 10−7) optical interference of comparable underlying transition amplitudes <ref:2607. p. 9 p. 908355#pg2>.

Methodology for Analysis

The electronic structure was computed within the self-consistent many-body perturbation theory framework, specifically the QSGWc cycle <ref:2607.08355#pg2>. Excitons were extracted from the Bethe-Salpeter equation (BSE) solutions, and the per-channel transition-density vector was reconstructed to verify oscillator strengths <ref:2607. p. 508355#pg2>. The interference figure of merit was derived by comparing the coherent sum to the incoherent sum of per-channel contributions <ref:2607. p. 508355#pg2>. The k-space participation ratio, kPR(S), was also reported as an auxiliary diagnostic <ref:2607. p. 608355#pg2>.

Conclusion on Physical Interpretation

The brightness hierarchy in CrSBr is not controlled by onsite d-d weight alone <ref:2607.08355#pg2>. Bright and dark excitons are symmetry-adapted coherent superpositions of ligand-field-like and charge-transfer-like band transitions <ref:2607. p. 008355#pg2>. The strongest bright/dark contrasts arise when nearly identical k-space manifolds form different diagonal and cross band-branch combinations, producing constructive (I ∼10) or destructive (I ∼10−4 to 10−7) optical interference of comparable underlying transition amplitudes <ref:2607.08355#pg2>. The resulting picture is that the source of brightness in CrSBr is neither purely Frenkel-like onsite physics nor purely Wannier-Mott spatial extent but a coherent optical-interference effect within a ligandfield-derived magnetic band structure <ref:2607. p. 908355#pg2>.

Data Availability and Contributions

All data required to reproduce the results of this paper – the converged QSGWc single-particle eigenvalues and eigenvectors, the optical matrix elements, and the BSE eigenvalues, oscillator strengths and eigenvectors, together with the analysis scripts that produce Figs. 1, 2, 3 and Table I – are publicly available on Zenodo <ref:2607. p. 008355#pg2>. S.A. conceived the work, developed the methodology, performed and analysed the calculations, and wrote the manuscript <ref:2607.08355#pg2>. J.M., D.P., M.v.S., J.C.J. and J.L.B contributed to the interpretation of the results and to editing the manuscript <ref:2607. p. 008355#pg2>.

Competing Interests and Acknowledgments

The authors declare no competing interests <ref:2607.08355#pg2>. This work was authored by the National Laboratory of the Rockies for the U. p. 0S. Department of Energy (DOE) under Contract No. p. 0 DE-AC36-08GO28308 <ref:2607. p. 008355#pg2>. S.A., J.M., J.C.J., and J.L.B acknowledge funding from the Laboratory Discretionary Research and Development (LDRD) program of the National Laboratory of the Rockies <ref:2607. p. 008355#pg2>. For M.v.S. and D.P., funding was provided by the Computational Chemical Sciences program within the Office of Basic Energy Sciences, U. p. 9S. Department of Energy <ref:2607. p. 008355#pg2>. S.A acknowledges the use of computational resources sponsored by the Department of Energy’s Office of Energy Efficiency and Renewable Energy and located at the National Laboratory of the Rockies <ref:2607. p. 008355#pg2>. S.A acknowledges the use of the National Energy Research Scientific Computing Center, under Contract No. DE-AC02-05CH11231 using NERSC award BES-ERCAP0021783 for a portion of the work <ref:2607. p. 008355#pg2>. The views expressed in the article do not necessarily represent the views of the DOE or the U. p. 9S. Government <ref:2607. p. 008355#pg2>.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI can accurately predict exciton brightness by calculating an interference figure of merit I, which is defined as the ratio of the coherent and incoherent sums of the per-channel contributions (Equation 9). This allows for distinguishing between bright and dark partners based on internal phase effects rather than just population differences.

  2. Improved AI can differentiate between Frenkel-like and Wannier-Mott interpretations by analyzing the diagonal D2 and cross X2 weights, which are a diagnostic of whether the exciton is dominated by same-branch or cross-branch transitions (Equation 10). This enables a system to classify an exciton as being a symmetry-adapted, ligand-hybridized object rather than a simple molecular excitation.

  3. Improved AI can predict experimentally observable spectral features by calculating the optical amplitude D alpha S and oscillator strength f alpha S using the BSE eigenvector and transition density vectors. This allows the system to simulate how bright and dark partners are symmetry-adapted coherent superpositions of ligand-field-like and charge-transfer-like band transitions.

  4. Improved AI can identify subtle but crucial experimental signatures by calculating k space participation ratios (kPR) for excitons, which are used to compare near-partner states. This helps predict whether the same exciton hierarchy is not controlled by onsite d-d weight alone and identifies states that are nearly identical k-space regions.

Related papers