Structural distortion and electron redistribution in dual-emitting gold nanoclusters

The study addresses how excited-state structural transformations in atomically precise gold nanoclusters govern their photoluminescence, with the goal of achieving controllable dual emissions for sensing applications. The authors hypothesize that photoexcitation induces a structural distortion coupled to electron redistribution within bi-tetrahedral metal(0) cores, producing two emissive states with distinct dynamics and environmental sensitivities. Understanding and controlling this process is important for designing luminescent nanoclusters as ratiometric probes responsive to polarity, viscosity, temperature, and pressure.

The work builds on concepts of photoinduced structural dynamics and charge-transfer phenomena known in molecules such as twisted intramolecular charge-transfer (TICT) systems, and prior studies of excited-state dynamics in transition-metal complexes and metal nanoclusters. Previous reports have shown luminescence tunability via shell rigidification and heavy-atom effects in nanoclusters, and dual emissions in certain Au/Ag clusters. However, direct identification of excited-state structural distortion coupled with electron redistribution within metal cores had not been fully unraveled. The authors reference time-resolved spectroscopies (fs-TA, TCSPC), Stark spectroscopy as a probe of charge redistribution, and earlier structural determinations of relevant bi-tetrahedral clusters.

- Samples: Three bi-tetrahedral nanoclusters were studied: Au24(S-TBBM)20 (Au24), Au14Cd(S-Adm)12 (Au14Cd), and Au24(S-PET)20 (Au24'). Syntheses followed published procedures. - Steady-state spectroscopy: UV-Vis absorption measured with a Hewlett Packard 8543 diode array spectrophotometer. Photoluminescence (PL) spectra recorded with Edinburgh FS-5 (200–870 nm and 850–1600 nm detectors), Horiba Nanolog Hybrid Fluorimeter with Ocean Optics 65000FL CCD (400–1100 nm), and Photon Technology International QM 40 with InGaAs detector (500–1700 nm). Optical densities maintained at ~0.1 at excitation (500 nm for Au24, 550 nm for Au14Cd, 750 nm for Au24'). - Time-resolved PL: Time-correlated single-photon counting (TCSPC) with a femtosecond laser (515 nm excitation). Decays fitted with bi-exponential functions. Selected detection windows: PL I (e.g., 550–750 nm), PL II (e.g., 900–1000 nm), or specific wavelengths (e.g., 850 nm for Au14Cd). - Temperature-dependent PL: Measurements from 298 K to 80 K using a Fluorolog-3 spectrofluorometer coupled with an Optistat DN cryostat and ITC temperature controller; samples in 2-methyltetrahydrofuran. - Femtosecond transient absorption (fs-TA): Yb:KGW laser output (1030 nm, 220 fs, 100 kHz) split for pump/probe. Pump generated by NOPA (520 nm for Au24, 550 nm for Au14Cd, 750 nm for Au24'); white-light continuum probe generated in a YAG plate. Probe windows: 500–900 nm and 1050–1300 nm (with a scattering gap between 900–1050 nm). Pump–probe overlap at <10°; transmitted probe collected by linear CCD. Solvents: DCM or hexane in 1 mm cuvettes. Transient signals computed as ΔT/T = [T(pump-on) − T(pump-off)]/T(pump-off). - Stark spectroscopy: Electroabsorption (EA) and electrofluorescence (EF) in a home-built setup at 77 K (low-temperature glass). White Xe lamp source, Acton monochromators, photodiode or PMT detection, lock-in detection at harmonics of an oscillating field (EA at second harmonic 441 Hz; EF at twice 75 Hz). EA/EF spectra fitted as linear combinations of zeroth, first, and second derivatives of absorption to extract changes in transition moment, polarizability (Δα), and dipole moment (Δμ) between ground and excited states. - High-pressure PL: Au24 in toluene loaded into a diamond anvil cell (500 μm culet). Stainless-steel gasket pre-indented to 400 μm with 200 μm hole; pressure calibrated by ruby fluorescence. PL collected in a Horiba XploRA+ confocal Raman setup (532 nm excitation, 600 groove/mm grating). Pressures up to 3.6 GPa investigated. - Environmental effects: Systematic PL and TCSPC across solvents of varying polarity/viscosity (e.g., DCM, toluene, hexane, butanol, octanol), solid state (crystal/film), and liquid nitrogen conditions; viscosity tuning by adding 1,2-dichlorobenzene to DCM.

- Dual emission and dynamics in Au24(S-TBBM)20: - Two PL bands: PL I ~650 nm and PL II ~1050 nm with distinct excitation spectra. TCSPC components: PL I lifetimes τ1 = 1.6 ns (82%), τ2 = 77.1 ns (18%); PL II lifetimes τ1 = 66 ns (77%), τ2 = 265 ns (23%). Early-time dynamics show PL II rise correlated with PL I decay, indicating state-to-state transfer on nanosecond timescales. - fs-TA reveals ground-state bleach near absorption features and multiple ESAs: ESA bands spanning 600–850 nm and 750–1200 nm; kinetics consistent with sequential relaxation from an electron-redistribution state (emitting PL I) to a distorted electron-redistribution state (emitting PL II) over ~ns. - Environmental sensitivity and sensing capability: - Solvent polarity: More polar solvents red-shift both PL I and PL II and strengthen PL II, consistent with charge-redistribution character. - Viscosity: Higher viscosity (butanol, octanol) increases quantum yield (QY) and PL I/PL II ratio; addition of viscous 1,2-dichlorobenzene to DCM slows the nanosecond state transition. - Across environments: QY increases from ~2% (DCM) to 20% (butanol), ~25–30% in solid state, and up to ~80% at liquid nitrogen temperatures; PL I blue-shifts from ~650 nm to ~600 nm with cooling/rigidification; PL II becomes negligible relative to PL I under rigid/high-viscosity conditions. - High pressure (up to 3.6 GPa in toluene): PL I intensifies and blue-shifts, PL II is suppressed (detector up to 1000 nm); attributed to pressure-induced viscosity increase hindering structural distortion and state transition—enabling ratiometric pressure sensing. - Stark spectroscopy: - For Au24 PL I, EF analysis yields a dipole moment change Δμ ≈ 1.07 D, indicating partial charge-transfer-like electron redistribution within the Au8 bi-tetrahedral core. EF/EA spectral range limited to ≤850 nm prevented evaluation of PL II. - Bi-tetrahedral series comparison and structural control: - Au14Cd(S-Adm)12: Dual emission with PL I ~770 nm, PL II ~800–900 nm. TCSPC at 850 nm shows τ1 = 334 ps (93.8%), τ2 = 86.6 ns (6.2%). fs-TA after 550 nm pump shows evolving ESA bands (e.g., ~730→~670 nm within 100 fs–1 ps; 1–10 ps shift to 700–950 nm ESA; subsequent evolution over 10–400 ps). Solvent (hexane) slows transfer to the distorted state (~500 ps) and suppresses PL II, accelerating PL I decay to ground state. Energy gap ΔE between states ~0.24 eV; transition time ~200–360 ps. - Au24(S-PET)20 (Au24'): Dual emission at ~815 nm (PL I) and ~970 nm (PL II); state transition on ~700 ps timescale. ΔE ~0.25 eV. - Au24(S-TBBM)20: Larger ΔE ~0.68 eV correlates with slower transition (~1.6 ns). Structural interpretation: face-to-face bi-tetrahedra in Au24 have greater freedom for photoinduced distortion; edge-sharing geometry in Au14Cd restrains motion, reducing the structural difference between states and accelerating the transition. In multi-tetrahedral fcc-like cores, motions are interlocked, suppressing such distortions and electron-redistribution signatures. - Mechanistic conclusion: Photoexcitation generates an electron-redistribution state that can undergo a slow, environment-sensitive structural distortion within the bi-tetrahedral metal core, yielding a second emissive state. This underlies the observed near-IR dual emission and enables ratiometric sensing of multiple parameters.

The findings confirm the central hypothesis that photoexcited bi-tetrahedral gold nanoclusters undergo an excited-state structural distortion coupled with electron redistribution within the metal core, producing two emissive states with distinct spectra and lifetimes. Time-resolved spectroscopies (TCSPC and fs-TA) map the sequential relaxation from an initial electron-redistribution state (PL I) to a distorted electron-redistribution state (PL II), with transition rates governed by the energy gap and core connectivity. Stark spectroscopy provides direct evidence of dipole change for PL I, consistent with charge-redistribution behavior. The strong sensitivity of the PL I/PL II ratio, peak positions, and quantum yields to polarity, viscosity, temperature, and pressure demonstrates that this dual-emission system functions intrinsically as a ratiometric sensor. Structural comparisons across Au24, Au14Cd, and Au24' highlight design rules: weaker inter-tetrahedral coupling (face-to-face) allows larger distortions and slower transitions, while edge-sharing or cross-joint geometries restrict motion, decrease energy gaps, and accelerate transitions. These insights enable rational tuning of excited-state dynamics for sensing, probing, and switching applications.

This work identifies and elucidates a photo-induced structural distortion and electron redistribution in bi-tetrahedral gold nanoclusters, establishing the atomic-scale origin of their near-IR dual emission. The sequential transition between two emissive excited states is slow and highly environment-sensitive, enabling versatile, self-calibrated ratiometric sensing of solvent polarity, viscosity, temperature, and pressure. By tailoring bi-tetrahedral core connectivity, the transition dynamics and energy gaps can be controlled, offering a general strategy to design next-generation luminescent nano-probes, sensors, and switches. Future research could extend Stark and ultrafast measurements to longer wavelengths to directly probe PL II, explore broader ligand chemistries and core connectivities, and integrate these nanoclusters into device architectures for practical multi-parameter sensing.

- Stark spectroscopy and electrofluorescence measurements were limited to ≤850 nm, preventing direct assessment of dipole changes for the PL II state (~1050 nm). - fs-TA probe was limited to >500 nm; an ESA below 600 nm overlapped with the ground-state bleach, complicating analysis. - High-pressure PL detection was limited to ≤1000 nm, so PL II changes beyond this range could not be fully monitored. - The heavy-atom (Se) analogue was excluded from optical analysis due to strong heavy-element effects, limiting generalization across chalcogenide ligands. - Photoinduced structural distortion was not observed in multi-tetrahedral fcc-like cores due to interlocking; thus, conclusions primarily apply to bi-tetrahedral architectures.