Fig. 4: Ultrafast recombination dynamics and coexistence of different excitonic species.
a, Maximum of the pump-induced change in the MIR electric field \(\Delta {E}_{{\rm{MIR}}}^{\max }\) as a function of the pump–probe delay time tpp for a fixed electro-optic sampling time tEOS (light-coloured lines; ΦNIR = 225 μJ cm–2) recorded at different temperatures T. The data on a logarithmic scale are vertically offset for clarity. The darker-coloured lines represent mono- and biexponential decay fits. For clarity, a ten-point (thirty-point) moving average was applied for delay times smaller (larger) than 20 ps. The dashed lines indicate the isolated fast- and slow-decay components of the fit and highlight the deviation from monoexponential decay at elevated temperatures. Top: density fraction of intralayer and interlayer excitons as a function of tpp at T = 200 K (dashed lines). b, Decay times of intralayer (τ1; blue circles) and interlayer (τ2; red triangles) excitons as a function of T obtained from the fit in a. τ2 rises close to TN (dashed line). c, τ1 and τ2 as a function of T with an applied magnetic bias (Bstat = 200 mT) along the b axis, obtained from the fit in Extended Data Fig. 6. The slower decay component (τ2) emerges at T′ ≈ 80 K (grey dashed line). d,e, Density fraction of intra- and interlayer excitons as a function of T without (d) and with (e) an applied magnetic field. The extraction procedure is described in Supplementary Section 4. The error bars represent the 95% confidence interval of the fitting routine. For the temperatures featuring only the decay component τ1 in b and c and no error bars in d and e, a monoexponential decay was sufficient to describe the experimental data reasonably well. Panels b–e share the same colour code. The inset in a shows the level scheme of interlayer-like (Xinter) and intralayer-like (Xintra) excitons relative to the continuum (shaded region) and the ground state (GS). Upon excitation by the NIR pulse (yellow arrow), both exciton species can form (dashed arrows) and decay with rates τ1 and τ2. Elevated lattice temperatures facilitate the population of Xinter (purple arrow), resulting in a decrease (increase) in the fraction of the short-lived (long-lived) exciton species.
