Ultrafast photoluminescence and multiscale light amplification in nanoplasmonic cavity glass

The study addresses how exciton–plasmon coupling can be realized in a solid, bulk material to enhance and accelerate photoluminescence. Prior work has established that localized surface plasmon resonances (LSPR) in metallic nanoparticles amplify local electromagnetic fields, enhancing fluorescence/PL, nonlinear optics, and Raman scattering. Plexcitonic systems, where emitters (e.g., quantum dots or molecules) couple to plasmonic nanoparticles, can yield amplified emission, spasers, coherent radiation, and room-temperature strong coupling, enabling applications in low-threshold lasers, sensing, and quantum technologies. Quantum dots are attractive emitters owing to their stability and size-tunable emission, but their spectral resolution (typical FWHM ~20–30 nm) limits applications. Most demonstrations of exciton–plasmon coupling have been in colloids or surface-assembled systems; bulk counterparts have been underexplored. The authors aim to fill this gap by creating a bulk glass nanocomposite co-doped with CdTe QDs and Ag nanoparticles using a Nanoparticle Direct Doping (NPDD) method that preserves nanoparticle identity and distribution, and by demonstrating ultranarrow, ultrafast PL at room temperature with evidence of coherent light amplification.

The work builds on extensive literature showing LSPR-driven enhancement of optical processes, including fluorescence and SERS, and the development of plexcitonic devices such as spasers and plasmonic nanolasers. Prior systems typically use surface architectures (e.g., metallic films with nanoparticle gaps) or colloidal assemblies to achieve Purcell enhancement and strong coupling. Quantum dots have been engineered for displays and photodetectors, but their bandwidths remain ~20–30 nm in commercial materials. Bulk matrices (polymers and glasses) have been explored for nanoparticle incorporation; polymers face aggregation and matrix separation issues. Historically, semiconductor nanocrystals in glass were formed by thermal precipitation from precursors, yielding inhomogeneous sizes and limited control over shape and co-doping. The NPDD method overcomes these limitations by directly doping pre-formed nanoparticles or QDs into glass, enabling control of size, composition, and co-doping (e.g., Ag/Au plasmonic elements with QDs). Comparisons in the discussion reference reported FWHM and PL lifetimes in related plasmonic-cavity systems, highlighting that the present bulk system achieves among the narrowest bandwidths at room temperature.

Materials and fabrication: Sodium borophosphate (Na3B2P3O13, NBP) glass was prepared from Na2CO3, NH4H2PO4, and H3BO3 (molar ratio 5:6:4), melted at 920 °C, then ground to powder. For the plasmon–exciton glass, commercially available CdTe QDs (Amax = 510 nm, diameter ~1.5 nm) and Ag nanoparticles (nAg, 20–30 nm; average ~25 nm by TEM/XRD) were directly mixed into the NBP glass powder at 0.3 wt% (CdTe) and 0.4 wt% (nAg), respectively, forming NBP:CdTe(0.3)nAg(0.4). A reference NBP:CdTe(0.3) was also prepared. The NPDD process used a micro-pulling down (μ-PD) apparatus: the powder mixture was loaded into an alumina crucible, inductively heated by an iridium tube until a melt formed at the shaper exit; convection aided mixing. Rods were pulled down in nitrogen at 1–2 mm/min. Characterization: Photoluminescence (PL) spectra were recorded under 405 nm laser diode or 488 nm CW Ar* laser excitation using a Jobin Yvon HR460 monochromator (1200 L/mm) with lock-in detection; signals were collected with cooled PMTs. Temperature-dependent PL was measured in a closed-cycle cryostat from 8 K to 300 K. PL lifetimes were measured with a Hamamatsu streak camera using a Coherent Libra femtosecond laser (1 mJ, 89 fs) with OPerA-Solo OPA tuned to 450 nm; PL collection at 480 or 625 nm. Fluorescence lifetime imaging microscopy (FLIM) mapped QD distribution. Numerical simulations: Finite-difference time-domain (FDTD) modeling evaluated PL emission enhancement (γems) and total decay enhancement (Purcell factor γtot) for a CdTe QD placed (i) at distance z from a single Ag nanoparticle and (ii) within a dimer nanocavity (two Ag particles, gap 2z). Excitation wavelength 477 nm and emission 513 nm matched QD absorption/emission. The PL in the weak-excitation regime used γems = γexc·η/η0 with η0 ≈ 0.25, where η = γrad / (γrad + γnr + (1−η0)/η0), and γtot = γrad + γnr. Distances approached z ≈ 1 nm (gap 2 nm) to probe maximal enhancements. Power-dependent measurements: PL intensity vs. excitation power was collected to identify thresholds for coherent amplification and saturation. Extinction spectra were also measured spatially to identify dips coincident with the narrow PL band, indicative of coupling features.

• The NBP:CdTe,nAg glass exhibits a strong, ultranarrow exciton PL at ~503 nm with FWHM = 13 nm at room temperature under CW excitation, significantly narrower than CdTe QDs in water (FWHM 34 nm at 505 nm) and narrower than typical Cd-based QDs (~20–30 nm).
• The exciton PL intensity in NBP:CdTe,nAg increases by ~50× compared to the non-plasmonic NBP:CdTe glass, while the broadband defect-related emission (530–800 nm) is strongly suppressed.
• PL lifetimes in NBP:CdTe,nAg are exceptionally fast, τ ≈ 60–90 ps, nearly three orders of magnitude shorter than QDs in water (≈30 ns), consistent with Purcell enhancement.
• FDTD simulations indicate that single Ag nanoparticles yield limited PL enhancement (γems < 5), whereas QDs located in Ag dimer nanocavities can experience γems ≈ 7×10^4. The total decay enhancement (Purcell factor) reaches ≈500 near a single particle and ≈10,000 in a 2 nm gap dimer, explaining the observed intensity increase and lifetime shortening. The authors note these are optimistic upper bounds due to idealized emitter placement.
• Temperature-dependent PL deconvolutes excitonic (X) and donor–acceptor (D–A) contributions. The QD narrow emission peak shows a thermal coefficient ΔE ≈ 0.495 meV/K, larger than bulk CdTe (≈0.3396 meV/K), consistent with quantum confinement. In the plasmonic glass, at T < 100 K, QD emission increases strongly while surface-defect (SD) emission diminishes, suggesting preferential energy transfer to QD states aided by plasmonic enhancement.
• Electron–phonon coupling analysis using the configurational coordinate model shows for the X line S ≈ 0.7 (weak coupling; figure indicates Ephon ≈ 17 meV), while the QD total peak FWHM fits S ≈ 40 with Ephon ≈ 43 meV, implying strong coupling with lattice distortion relative to typical CdTe phonon energies (~21.7 meV).
• Power-dependent PL exhibits a change in slope above ~100 mW, consistent with coherent light amplification; saturation appears above ~165 mW. Spatially resolved extinction shows a dip coincident with the narrow PL band in regions of high density, a signature of plasmon-driven lasing and intermediate coupling in QD–plasmon nanocavities.
• The material demonstrates ultrafast, narrowband, room-temperature spontaneous emission under CW pumping in a bulk, easily fabricated glass, enabling integration and scalability.

The findings confirm that embedding CdTe quantum dots within a plasmonic nanocavity framework in bulk glass substantially enhances spontaneous emission intensity while dramatically shortening radiative lifetimes, addressing the challenge of achieving high-performance plexcitonic behavior in solid volumetric materials. Simulations and experiments together indicate that dimer nanocavities, not single nanoparticles, dominate the observed enhancements by providing extreme field confinement and spectral alignment with QD emission. Temperature-dependent behavior supports the assignment of the narrow band to QD excitons and suggests plasmonically facilitated energy funneling toward QD states. The observed threshold-like power dependence and extinction dips corroborate coherent amplification consistent with plasmon-driven lasing in regions of high nanocavity density. Compared to previous surface or colloidal architectures, this bulk glass achieves among the narrowest room-temperature QD emission bandwidths while retaining ultrafast decay, showing potential for integrated, high-speed photonic components.

This work demonstrates a scalable, bulk nanoplasmonic glass (NBP:CdTe,nAg) fabricated by NPDD that delivers ultranarrow (13 nm FWHM at ~503 nm), ultrafast (60–90 ps) room-temperature photoluminescence under CW excitation, with strong suppression of defect emission and evidence of coherent amplification above ~100 mW. FDTD modeling attributes the enhancements to QDs residing in Ag dimer nanocavities that provide large excitation and Purcell factors. The approach enables practical, volumetric plexcitonic materials compatible with integrated devices. Future directions include further narrowing via coupling to high-Q resonators (e.g., whispering gallery modes), deliberate control of nanocavity geometries and gaps, multi-emitter or multi-plasmonic co-doping for tailored spectra, and application-specific engineering for displays, telecommunications, amplifiers, and biosensing. The strong coupling indications and fast decays also point to prospects for high-rate single-photon emission and spintronic/quantum information applications.

• The numerical enhancements represent optimistic upper bounds due to idealized emitter placement; in real materials, QDs are unlikely to be positioned at ~1 nm from Ag surfaces or in perfectly symmetric 2 nm gaps, leading to a distribution of enhancements.
• The nanocavity geometry and emitter–cavity alignment are not explicitly controlled but arise from random self-assembly during NPDD, introducing sample-to-sample variability.
• Power-dependent measurements indicate saturation above ~165 mW, and while coherent amplification is inferred from slope changes and extinction dips, a full laser characterization (e.g., linewidth narrowing vs. power, photon statistics) is not reported.
• The study focuses on a specific composition (0.3 wt% CdTe, 0.4 wt% Ag) and emitter/NP sizes; generalization to other sizes/compositions and long-term stability under continuous operation are not detailed.