Localization of coherent light into photons in a single-crystalline material

Abstract: The absorption of light by materials is one of the most fundamental processes in optics and condensed-matter physics. Here we investigate whether laser light is absorbed by a crystalline material as an electromagnetic wave or as localized photon energies. We excite the first-order phase transition of vanadium dioxide with laser pulses of sufficient frequency to overcome the band gap but with insufficient pulse energy to overcome the latent heat. According to Maxwell's equations and Bloch theory, no transition should occur, because nowhere in the material is enough energy. Nevertheless, we observe with ultrafast electron diffraction for short times a disordered crystal geometry with nanometer-sized spots of switched material. Their amount matches approximately to the number of photons in the absorbed laser wave. Two optical experiments confirm this phenomenon, and simulations of single absorbed photons reproduce all measurements results. Although laser light and Bloch electrons are extended quantum objects, the energy of the individual photons is localized into nanometer dimensions, enabling local consequences at substantially higher energy than average.

• The paper demonstrates that energy from coherent light is localized as individual photons, inducing nm-scale phase domains in VO2.
• The paper uses femtosecond ultrafast electron diffraction and mid-IR pump-probe techniques to distinguish localized photon absorption from thermal heating.
• The paper establishes a quantitative correspondence between photon count and induced domain formation, challenging traditional wave-based energy deposition models.

Photon Localization in Coherent Light Absorption by Single-Crystalline VO$_2$

Introduction

This study addresses a longstanding question in condensed matter optics: does a macroscopic, coherent electromagnetic wave deposit its energy into a crystal as a delocalized field or as localized quantum photons? The authors employ femtosecond-scale experiments on single-crystalline vanadium dioxide (VO$_2$), exploiting its well-characterized first-order monoclinic-to-rutile phase transition with a significant latent heat and ultrafast insulator-metal transition. By judiciously selecting laser fluence such that individual photons have sufficient energy to bridge the band gap but the overall pulse does not supply enough energy to thermally drive the entire transition, the work discriminates between wave-like and photon-like energy deposition mechanisms.

Experimental Design and Methodology

The experimental core involves single-crystal VO$_2$ lamellae probed using ultrafast electron diffraction (UED) following laser excitation ($\lambda = 1030$ nm, $E_{\rm ph} = 1.2$ eV, $\tau = 250$ fs). Pulse energy is set far below the threshold needed to overcome the latent heat across the illuminated volume, thereby ruling out classical nucleation and thermal domain growth. Precise spatial and temporal alignment ensures highly homogeneous single-domain behavior and robust exclusion of polymorphic or defect-induced artifacts, supported by comprehensive materials characterization (EDX, ICP-OES, calorimetry, and additional electron/mid-IR measurements).

Complementary all-optical mid-infrared pump-probe and thermal emission experiments track the ultrafast evolution of electronic and phononic observables in the same regime.

Key Observations

The central experimental result is the transitory, spatially localized breakdown and reversal of crystal translational symmetry within $\lesssim 1$ ps post-excitation, as revealed by:

• UED Analysis: There is a transient $\sim 7$\% reduction in monoclinic diffraction intensity and a concomitant increase in rutile peaks and diffuse scattering. The spatial inhomogeneity and magnitude of the diffuse component are incompatible with uniform heating or phononic disorder. Instead, they indicate nanometer-scale disordered regions (estimated typical domains $\sim 0.4$ nm$^3$), with a number density matched to the absorbed photon density (3.5–4.5\% per rutile unit cell).
• Mid-IR Reflectivity: Ultrafast reflectivity changes below the phase transition threshold scale linearly with photon density, exhibit rapid decay on $_2$0ps timescales, and are well reproduced by simulation assuming photon-localized energy deposition.
• Thermal Radiation: Absence of measurable high-$_2$1 emission at low fluence, transitioning to a linear-in-fluence signal (with constant spectral shape) above threshold. This demonstrates that sub-threshold excitations do not produce extended hot electron or lattice regions; rather, more photons generate more localized hot spots, not higher local temperatures.

Monte Carlo simulations with single-photon-localized energy depositions, including Ising-type cooperative flipping and realistic heat dissipation, quantitatively reproduce the experimental UED and optical traces.

Theoretical Implications

The key theoretical assertion is that laser absorption in a coherent, defect-free crystal occurs via the localization of individual photons, with the spatial extent of the initial excitation determined by the energy required to nucleate a phase transition cluster (latent heat, heating costs, interfacial energy), rather than by the delocalized electromagnetic field or a statistical distribution of electron-phonon interactions. This breaks the translational symmetries imposed by Maxwell's and Bloch's equations and requires a symmetry-breaking mechanism, which the authors attribute to decoherence mediated by the enormous phase space of electron, phonon, and quasiparticle states in the bulk.

This mechanism is distinguished from purely electron- or phonon-mediated symmetry breaking and is argued to reflect a measurement-like quantum collapse, in which the material effectively measures and localizes the photonic state.

Strong Numerical Results and Claims

The work establishes several robust quantitative correspondences:

• The temporal width of the disordering and its recovery agrees with thermalization timescales in VO$_2$2 (sub-5 ps regime).
• The cluster energy budget per domain is $_2$3 eV, consistent with the single-photon energy ($_2$4 eV), but ≫ latent heat/unit cell or V–V dimer breaking energy.
• The number of induced domains is approximately equal to the number of absorbed photons, a linear relationship confirmed for both structure and reflectance observables.
• Simulations with delocalized (wave-like) energy fail to produce these observables, supporting the necessity of photon localization.

The authors explicitly reject the classical homogeneous heating/nucleation picture by showing that the measured domain dynamics, scaling laws, and recovery kinetics are irreconcilable with these models.

Practical and Theoretical Implications

Practically, these results indicate that under coherent excitation, materials can exhibit transient local energy densities far exceeding the average input fluence, which is critical for ultrafast optical switching, photonic control of phase transitions, laser processing, and nanoscale energy concentration in optoelectronic and photovoltaic contexts.

Theoretically, this sheds new light on the quantum-to-classical crossover in condensed matter, demonstrating that even in large, perfect crystals, quantum measurement and decoherence are responsible for symmetry breaking necessary for photonic absorption. It suggests methodologies for probing many-body decoherence rates, localization scales, and the emergence of classical disorder in complex solids with high temporal and spatial resolution.

The findings challenge classical, mean-field treatments of light-matter interaction by highlighting the necessity of a quantum, photon-localized paradigm, with broader implications for the modeling of photoinduced transitions, heat management at nanoscales, and the interpretation of ultrafast phenomena in correlated electron systems.

Prospects for Future AI and Materials Science

AI-driven material design and analysis frameworks could leverage these insights to predict and optimize ultrafast and nanoscale responses, including the controlled generation of localized hot spots for logic, memory, or energy applications. The ability to simulate photon-localized dynamics at atomic resolution will become increasingly relevant for next-generation materials-by-design pipelines and for elucidating the role of quantum measurement in solid-state physics.

Conclusion

The study offers rigorous experimental and numerical evidence that the absorption of coherent laser light by single-crystalline VO$_2$5 occurs through the real-space localization of individual photons, breaking the macroscopic translational symmetry inherent to both light and electron Bloch states. This localization leads to transient formation of nm-scale domains undergoing phase transition, with the number and structure of the domains directly determined by the photon count and energy, and the dynamics governed by intrinsic decoherence and local thermodynamic constraints. The results have substantial implications for our understanding of quantum-classical transition in solids, for the technology of ultrafast optics and photonics, and for the development of AI tools aimed at predictive modeling of nanoscale light-matter interactions.