Picosecond time-resolved photon antibunching measures nanoscale exciton motion and the true number of chromophores

Multichromophoric nanoparticles can host multiple excitons that interact via diffusion and singlet–singlet annihilation (SSA), affecting brightness, photoluminescence lifetime, harvesting efficiency, and photostability—key for optoelectronic performance. Conventional photon antibunching (Hanbury Brown–Twiss) can count chromophores only if SSA is negligible or, conversely, estimate SSA rates if the number of chromophores is known—conditions rarely met simultaneously. The research question is how to decouple and quantify both the true number of chromophores and the time-dependent exciton diffusion/annihilation dynamics from the same measurement. The authors introduce picosecond time-resolved antibunching (psTRAB), which leverages the microtime-resolved photon stream under pulsed excitation to reveal early-time statistics reflecting the physical chromophore count and later-time evolution governed by diffusion-assisted annihilation.

Prior work has used photon antibunching to count emitters and to probe multi-exciton processes in diverse systems (conjugated polymers, quantum dots, nanocrystals, light-harvesting complexes). Reports of single-photon emission from large multichromophoric systems have been interpreted as evidence of long-range interactions, but such analyses lose information about the true chromophore number when exciton diffusion and SSA are present. Ensemble photoluminescence analyses typically infer diffusion/annihilation from non-exponential decays at high excitation densities, far from device-relevant conditions, and cannot access rare double-excitation events at low fluence. Time-gated photon correlations have been used in specific contexts but not to concurrently extract chromophore counts and exciton transport parameters broadly across deterministic multichromophoric structures and polymer aggregates.

psTRAB uses time-correlated single-photon counting (TCSPC) with a Hanbury Brown–Twiss (HBT) setup to record absolute photon arrival times on two detectors: (i) macrotime (pulse index) and (ii) microtime (delay after excitation). The photon stream is binned by microtime windows after each laser pulse, and cross-correlation histograms are constructed between detectors. For each microtime bin, the ratio of central (Δt = 0) to lateral (Δt ≠ 0) coincidence peaks, N0/Ne, yields the instantaneous number of independent chromophores, n, synchronized to excitation; early bins report the physical chromophore count before annihilation occurs, while later bins reflect exciton diffusion and SSA reducing n. Analytical model: For two-photon events from the same pulse, probabilities P(t) and P′(t+Δt) are derived assuming single-exponential annihilation with rate kSSA. The ratio N0/Ne connects to the number of dyes, dyes, and kSSA, leading to n(t) = {y0 [A exp(−kSSA t)]}−1 for simple pairwise annihilation, and a biexponential generalization n(t) = {y0 [A1 exp(−kSSA,1 t) + A2 exp(−kSSA,2 t)]}−1 when near- and next-nearest interactions both contribute. The physical dye count is extracted from the asymptote via ndyes = (y0 − A)−1 (monoexponential) or ndyes = (y0 − (A1 + A2))−1 (biexponential). Importantly, the N0/Ne ratio is independent of radiative/nonradiative decay rates and of singlet–triplet or other dark-state interactions. DNA-origami model system: A 12-helix-bundle (225 nm) with five labeling sites (∼3 nm spacing) was constructed; seven designs varied dye number and spacing (ATTO647N under ROXS). Confocal microscopy with pulsed excitation (636 nm, ∼80 ps FWHM, operated at 40 MHz) recorded TCSPC data. PL decays and psTRAB histograms were computed in 200 ps microtime bins; data from 54–98 individual origami structures per design were accumulated; only the first 5 s per trace were analyzed with ≤10% PL intensity drift to minimize photobleaching/blinking effects. Conjugated polymer aggregates: Deterministic aggregates from two PPEB polymers produced ordered H-type (PPEB-1) and disordered J-type (PPEB-2) coupling via solvent vapor annealing. Isolated aggregates embedded in PMMA were measured on a confocal microscope using femtosecond excitation (to avoid double excitation within lifetime): 405 nm for PPEB-1 and 440 nm for PPEB-2. Large datasets were acquired: 631 H-type aggregates (∼54 chains each) and 705 J-type aggregates (∼9 chains). psTRAB n(t) was computed with microtime bins from 3 ps at early times up to 80–160 ps at later times. For diffusion-controlled SSA, a second-order rate formulation for small-number reactants leads to d n/dt = −kSSA n(n−1). Transforming yields a linear relation: ln(n/(n−1)) Vagg = y t + const, where Vagg is the aggregate volume and y is the bimolecular annihilation rate. Thus, plotting ln(n/(n−1)) Vagg versus time gives y from the slope; time dependence of y indicates dimensionality and regime of diffusion. Aggregate volumes were estimated from chain counts and material density (details in Supplementary Information). Diffusion lengths were inferred at late times (y → 0) from the number of independent chromophores remaining and aggregate volume, assuming spherical 3D exploration volume per exciton.

DNA origami chromophore counting and annihilation: • Five-dye structure (3 nm spacing): PL lifetime ∼4.2 ns (single exponential). psTRAB gives n ≈ 4.8 in the first 200 ps (close to 5), dropping to ∼2.8 (200–400 ps) and ∼1.1 (6400–6600 ps), demonstrating time-resolved SSA. • One-dye control: n ≈ 1.02 after the first two points (early fast effects due to multiple excitation within an ∼80 ps pulse). • Two dyes at 12 nm: n = 1.85 ± 0.01 (constant), indicating negligible SSA. • Two dyes at 3 nm: kSSA = 1.72 ± 0.06 ns−1; ndyes = 1.8 ± 0.03. • Two dyes at 6 nm: kSSA = 0.06 ± 0.01 ns−1; ndyes = 1.8 ± 0.03. • Three dyes at 6 nm: kSSA = 0.06 ± 0.01 ns−1; ndyes = 2.7 ± 0.1. • Three dyes at 3 nm: biexponential SSA with amplitude-weighted average kSSA = 0.98 ± 0.09 ns−1; ndyes = 2.9 ± 0.1. • Five dyes at 3 nm: biexponential behavior with average kSSA = 0.72 ± 0.07 ns−1; ndyes = 4.7 ± 0.2. • For all cases with nonzero average kSSA, n → 1 at long microtimes, evidencing exciton hopping/diffusion enabling annihilation even when distant dyes cannot annihilate directly. Conjugated polymer aggregates (psTRAB diffusion analysis): • H-type (ordered) vs J-type (disordered) aggregates show distinct n(t) dynamics. H-type: rapid drop in n over first ∼250 ps, then continued decay leveling at ∼2000 ps. J-type: smaller fast drop, then slower near-linear decay, leveling slightly higher than H-type at >2000 ps. • ln(n/(n−1)) Vagg vs time identifies three regimes: (i) Early (<250 ps): curvature indicates time-dependent y and 1D to <2D diffusion (along-chain and interchain π-stack for H-type; predominantly along-chain for J-type). (ii) Intermediate (250–2000 ps): linear behavior indicates time-independent y and 3D diffusion; y is about an order of magnitude larger in H-type than J-type; typical y values in 10−9–10−10 cm3 s−1 range, consistent with literature. (iii) Late (>2000 ps): y → 0 as exciton density becomes too low for further annihilation. • Aggregate-size/disorder effect: In J-type aggregates, a smaller 6-chain aggregate shows nearly twofold higher y than a larger ∼9-chain aggregate in the 3D regime, consistent with increased order and stronger interchain coupling in smaller aggregates. • Lower limits for 3D exciton diffusion lengths from late-time regime: L3D ≥ 8.97 ± 0.1 nm (H-type) and L3D ≥ 5.2 ± 0.1 nm (J-type), in line with typical conjugated-polymer values. • Measurement conditions: Results are robust at low excitation densities (independent of excitation intensity within the examined regime) and sensitive to rare double-excitation events (e.g., ∼300 ppm coincidences in H-type aggregates).

psTRAB resolves, in a single experiment, the true chromophore count and the time evolution of exciton diffusion and annihilation. Early-time photon statistics faithfully report the physical number of emitters before annihilation, overcoming a key limitation of conventional antibunching analysis. Time-resolved changes in n quantify SSA rates and reveal transport dimensionality: initial 1D or <2D motion (intra-chain and early interchain hopping) transitioning to 3D diffusion at later times. The method differentiates ordered H-type from disordered J-type polymer aggregates through an order-of-magnitude difference in diffusion-controlled annihilation rates and distinct temporal regimes, reflecting interchain coupling and nanoscale disorder. Unlike ensemble PL-decay analyses that demand high exciton densities and deconvolution of excited-state decay, psTRAB directly accesses annihilation dynamics at low fluence by conditioning on two-photon events. This enables precise extraction of annihilation kinetics, diffusion dimensionality, and lower bounds on diffusion lengths relevant to the design of bright emitters and efficient optoelectronic materials.

The study introduces picosecond time-resolved antibunching as a powerful single-particle tool to count chromophores and quantify exciton–exciton annihilation and diffusion. On DNA-origami structures with controlled spacing, psTRAB accurately recovers the true number of dyes and distance-dependent SSA rates. Applied to conjugated-polymer aggregates, it distinguishes H- and J-type coupling, identifies time-dependent dimensionality of diffusion, and yields lower limits to 3D diffusion lengths. The approach provides a route to decode nanoscale organization and excitonic coupling in complex multichromophoric materials. Future work could extend psTRAB to higher-order photon correlations to separate multi-exciton interaction orders, apply to other nanomaterials and device-relevant environments (solid-state matrices with fixed dipole orientations), and integrate with structural probes to correlate morphology with exciton transport.

- Time resolution constraints: Fast early-time 1D intrachain motion in J-type aggregates may exceed the experimental temporal resolution, making precise quantification challenging. - Orientation effects: DNA-origami dyes in solution are assumed freely rotating; in fixed or solid matrices, fixed dipole orientations could significantly affect SSA rates and psTRAB dynamics. - Photon budget and background: Sensitivity to rare two-photon events requires long integration and is ultimately limited by detector dark counts and background. - Microtime binning trade-off: Choice of bin width (e.g., 200 ps for DNA-origami) balances timing precision against noise in correlation histograms. - Aggregate volume estimation: Derivation of y and diffusion lengths relies on volume estimates from mass/density and chain counts, introducing systematic uncertainty. - Photobleaching/blinking: Mitigated by analyzing only the first 5 s with stable intensity, but residual effects can slightly affect absolute antibunching contrast (not the temporal decay of n).