betaFeedback

WorksPhysics and Astronomy

Ultrafast Formation and Annihilation of Strongly Bound, Anisotropic Excitons

Abstract

Van der Waals (vdW) layered materials with long-range magnetic order have the potential to enable novel optoelectronic and spintronic applications. Among these, CrSBr is an air-stable, direct band gap semiconductor that hosts interlayer antiferromagnetic order, a highly anisotropic electronic structure, and strongly bound excitons. In particular, excitons in CrSBr have been shown to inherit the quasi-one-dimensional nature of the material and also couple to the underlying spinorder. However, mechanisms of exciton formation, dissociation, and interaction with free carriers remain largely unexplored, despite being crucial for spintronic and optoelectronic applications. Here, we employ time- and angle-resolved photoemission spectroscopy to map the electronic structure and excited state dynamics in CrSBr. We directly resolve an exceptionally large exciton binding energy (~800 meV) and a highly anisotropic momentum space distribution of the exciton, revealing its quasi-1D real-space character. We observe an excitation-density-dependent interconversion between bound excitons and quasi-free carriers on sub- to few-picosecond timescales, indicating that many-body effects govern the excited-state dynamics and optical properties during the initial stages of relaxation. Our work highlights the strongly bound, anisotropic character of excitons in CrSBr, as well as the microscopic interactions steering relaxation pathways after photoexcitation in elevated density regimes relevant for future device applications.

Introduction

Two-dimensional, van der Waals (vdW) materials feature emergent properties, new degrees of freedom, and strong many-body effects that may enable next-generation devices based on engineered functionality in the atomic limit[1, 2, 3]. In 2D semiconductors such as transition-metal dichalcogenides (TMDs, e.g. MoS2_2, WSe2_2), the reduced dielectric screening and layer confinement lead to Coulomb-bound electron-hole pairs, or excitons, with binding energies of ∼100\sim100s meV[4, 5, 6, 7]. These excitons dominate the optical properties and, in turn, affect device function and performance in atomically thin optoelectronic applications[8, 9]. vdW materials supporting long-range magnetic order provide additional functionality, namely for exploiting the spin degree of freedom in next-generation opto-spintronic applications built around two-dimensional vdW layers or heterostructures[10, 11, 12] and offer the opportunity to explore the interplay of many-body physics and magnetism in reduced dimensions.

Among these, CrSBr is an air-stable magnetic semiconductor hosting a direct electronic band gap, highly anisotropic optical and electronic properties, and pronounced many-body exciton physics[13, 14, 15, 16]. Below the Néel temperature (TN∼132T_\mathrm{N} \sim132 K), CrSBr is an A-type antiferromagnet, with electron spins ferromagnetically oriented in-plane within a layer and antiferromagnetically aligned between adjacent layers[17, 18, 19]. In particular, excitons in CrSBr display pronounced magnetic order-dependent interlayer hybridization[20, 21, 22, 23, 24, 25, 26], providing a basis for the optical read-out and control of the underlying spin order in a magnetic semiconductor. These excitons have also been proposed to display a quasi-1D character with large binding energy[15, 26], inheriting the structural and electronic anisotropy of the material. Previous work has observed the anisotropic conduction band dispersion and revealed rich exciton physics including, for example, exciton-polaritons[14, 27, 28] and exciton-magnon coupling[21, 22, 23, 29]. However, the role of free carriers and the nature of exciton formation and relaxation in the early stages after photoexcitation remains largely unexplored[30, 31]. Disentangling these contributions to the excited state dynamics and resulting optical and electronic properties is critical to realize new device functionalities.

Here, we employ time- and angle-resolved photoemission spectroscopy (trARPES) to directly probe the momentum-resolved excitonic signatures and photoinduced excited state dynamics in bulk CrSBr. Leveraging momentum microscopy, we image the two-dimensional momentum-space distribution of the lowest-energy exciton after photoexcitation and thereby extract the corresponding real-space exciton wavefunction, confirming its highly anisotropic nature. By resolving both the exciton and single-particle conduction band minimum simultaneously, we directly observe a large exciton binding energy (EbE_b) of ∼800\sim800 meV at room temperature as well as below the Néel temperature. Performing time-resolved measurements with a wide range of excitation densities and wavelengths, we uncover an ultrafast exciton decay channel, exciton-exciton annihilation, which dominates at elevated excitation fluences and leads to a competition between bound excitons and free carriers on a sub- to few-picosecond timescale. Our results provide novel insight into the underlying physics governing the ultrafast optical response and the anisotropic, quasi-1D excitons in CrSBr. In particular, understanding the excited state dynamics and exciton physics under elevated excitation densities is also crucial for practical device applications seeking to leverage the magneto-exciton coupling of CrSBr in next-generation opto-spintronics.

Results and Discussion

Ultrafast Momentum Microscopy

To map the electronic band structure and photoinduced dynamics of bulk CrSBr, we perform trARPES measurements utilizing a high-repetition-rate high-harmonic generation (HHG) extreme ultraviolet (XUV) laser source[32, 33, 34]. Here, tunable infrared femtosecond pump pulses polarized along the bb-axis of the crystal, corresponding to the direction of the dipole-allowed optical transitions[20], excite the sample and an XUV probe (∼21.7\sim21.7 eV) pulse photoemits electrons from the sample surface after a variable time delay[35]. With momentum microscopy, we simultaneously acquire the electron binding energy in addition to both in-plane momenta, obtaining a global view of the 2D electronic band structure and exciton dynamics in a single experiment (Figure 1a)[36]. In Figure 1b, a two-dimensional momentum map shows the photoemission intensity I(kx,ky)I(k_x, k_y) at constant energy near the valence band maximum (VBM). The central Γ\Gamma point (kx=ky=0k_x = k_y = 0 Å−1^{-1}) at the top of the valence band shows suppressed intensity compared to the Γ\Gamma points of the second Brillouin zones at the edges of the map[37, 38], which we use to define the zero energy reference (E−EVBM=0E - E_{\mathrm{VBM}} = 0 eV, Fig. S1).

Excitons and single-particle excitations in CrSBr

Figure 1. Excitons and single-particle excitations in CrSBr. a, trARPES of bulk CrSBr probes the dynamic momentum-space occupation I(kx,ky,E)I(k_x, k_y, E) as a function of pump-probe time delay, Δt\Delta t. Momentum maps I(kx,ky)I(k_x, k_y) at constant energy b, near the valence band maximum (VBM), and c, of the conduction band minimum (CBM) and d, of the lowest-energy exciton state after photoexcitation (hν=1.55h\nu= 1.55 eV) at T=120T = 120 K. e, The real-space exciton wavefunction is retrieved by Fourier transformation of the exciton momentum map, showing highly anisotropic character with rms radii of ∼0.13\sim0.13 nm and ∼0.48\sim0.48 nm along the crystal aa- and bb-axis, respectively. E=0E = 0 eV is referenced to the VBM at the second Γ\Gamma point. Data shown are integrated over positive delay times (Δt=0\Delta t = 0 to Δt=1000\Delta t = 1000 fs) with an energy window of 200 meV.

We photoexcite (hν=1.55h\nu= 1.55 eV, ∼\sim0.9 mJ/cm2^{2}, pulse duration ∼\sim40 fs, s-polarized) bulk CrSBr crystals, cleaved in ultrahigh vacuum, at T=120T = 120 K, below TNT_{\mathrm{N}}. In these conditions, we excite above the lowest-lying bright exciton resonance[39, 40] and estimate an exciton density of ∼3×1013\sim3 \times10^{13}/cm2^{2} or ∼\sim0.05 excitations per unit cell (see the Methods). After photoexcitation, we observe two distinct features around E ∼\sim 1.3 eV and E ∼\sim 2.1 eV relative to the VBM, each exhibiting an anisotropic momentum space distribution and quasi-flat dispersion along Γ−X\Gamma- X at low temperature (Figure 1a,Figure 1c,Figure 1d) as well as at room temperature (Figure 2a). Such an anisotropic dispersion of the conduction band has been predicted theoretically[15, 20] and recently observed in equilibrium ARPES of few-layer CrSBr in contact with metallic substrates[38] and after alkali metal dosing of bulk CrSBr[41].

Anisotropic dispersion and ultrafast dynamics in CrSBr

Figure 2. Anisotropic dispersion and ultrafast dynamics in CrSBr. a, Delay-integrated excited state signal along Γ−X\Gamma- X (ky=0k_y = 0 Å−1^{-1}) showing the flat dispersion of both the conduction band minimum (CBM) and the exciton. The energetic assignment of these states is extracted from a fit to the momentum-integrated EDC (violet) and the energy difference corresponds to the exciton binding energy, Eb∼800E_b \sim800 meV. b, Time traces at Γ\Gamma show that the CBM (red) exhibits a delayed rise compared to the exciton (black) after photoexcitation (hν=1.55h\nu= 1.55 eV). The time traces are both individually normalized and integrated in an energy and momentum window of 200 meV and 0.5 Å−1^{-1} in both kxk_x and kyk_y, respectively. Data is acquired at T = 300 K.

While initial experimental studies estimated the electronic band gap of CrSBr to be around 1.5 eV[17], recent theoretical and experimental ARPES works[37, 41, 42, 43] have suggested a larger band gap of roughly 2 eV. Optical spectroscopy measurements have shown the lowest-energy exciton resonance at around ∼\sim1.35 eV. Therefore, we assign these two features to the lowest-energy bound exciton state and the single-particle conduction band minimum (CBM) of CrSBr, respectively. To more precisely determine the energies of these two features observed in our trARPES measurements, we fit the momentum-integrated energy distribution curves (EDCs) to a sum of two Gaussian functions (Figure 2a), obtaining a peak energies of 1.25 eV and 2.04 eV for the exciton and CBM, respectively, at room temperature. Our assignment of the exciton and CBM is further supported by time-resolved experiments with pump photon energies above the band gap, as discussed below where we also expand on the origin of the CBM after pumping below the band gap. Therefore, we confirm the flat dispersion of the conduction band along Γ−X\Gamma- X and experimentally reveal the theoretically predicted quasi-1D character of the exciton. Additional momentum maps and trARPES spectra are displayed in Figures S2-S6.

By resolving both the exciton and CBM signatures in a single measurement, we directly determine the exciton binding energy from the energy difference of these photoemission features[44, 45], extracting an exceedingly large exciton binding energy of Eb=792±4E_{\mathrm{b}} = 792 \pm4 meV at room temperature (Figure 2a, S7). Momentum-dependence of the exciton binding energy is displayed in Figure S8. Below the Néel temperature at T=120T = 120 K, the band gap slightly increases, with the exciton and CBM energies found to be 1.34 eV and 2.14 eV, respectively (Figure S9). Here, we observe a similar binding energy of Eb=793±11E_{\mathrm{b}} = 793 \pm11 meV. A recent work reported an increased exciton binding energy (∼\sim20 meV) in the antiferromagnetic phase below TNT_{\mathrm{N}} when excitons are confined within a single layer, leading to a surface-confined exciton[46, 47]. We note that due to the surface selectivity of ARPES experiments, excitons probed on the surface of a bulk material experience reduced screening compared to those in the interior. Our measurement of the exciton binding energy in CrSBr is nonetheless almost one order of magnitude larger than what has been previously observed in analogous experiments on other bulk 2D vdW systems, such as the A exciton in bulk TMDs[45]. Therefore, the high exciton binding energy (∼\sim800 meV) that we observe is reflective of the intrinsic properties of CrSBr.

By resolving both in-plane momenta simultaneously, momentum microscopy provides powerful insights into the orbital and real-space character of excitons in many molecular and inorganic semiconductor systems[36, 48]. Excitons in CrSBr are predicted to be spatially anisotropic with a ratio of ∼\sim1:3 between the radius of the exciton along the crystal aa- and bb-axes with characters in between ideal limits of site-localized Frenkel and delocalized Wannier-Mott characters[20, 15, 49, 50]. Leveraging the two-dimensional momentum space information, we retrieve the real-space wavefunction envelope of the lowest-energy exciton in CrSBr by Fourier transformation of the momentum maps corresponding to the exciton[45, 51] (see the Supplementary Information). Following its anisotropic momentum distribution (Figure 1d), the exciton wavefunction displays an elongated distribution in real-space extended along the b-axis of the crystal (Figure 1e) and localized along the weakly coupled, quasi-1D Cr-S chains[15, 41]. From this, we determine the exciton’s root mean square (rms) radii (Fig. S10) as ∼0.13\sim0.13 nm and ∼0.55\sim0.55 nm along the crystal a- and b-axis, respectively. At room temperature (Fig. S11), excitons are even more strongly localized along the b-axis (∼0.48\sim0.48 nm) and delocalized between vdW layers. These values support a predominantly Frenkel-like character of the lower lying exciton[26] and are consistent with the large binding energy we observe, as well as recent estimates[52]. Our observations provide direct confirmation of the strongly bound (Eb∼800E_b \sim800 meV) and quasi-1D nature of the exciton in CrSBr.

The observation of the CBM feature at E ∼2\sim2 eV is striking, considering the pump excitation energy of 1.55 eV is well below our assignment of the electronic band gap. Figure 2 shows the room-temperature dispersion along Γ−X\Gamma- X as well as the ultrafast dynamics of the exciton and CBM features at Γ\Gamma. While there is a rapid rise of the exciton signal near zero time delay, the CBM exhibits a delayed rise after photoexcitation, pointing to a dynamical process that populates this state on a few hundreds of femtosecond timescale.

Time-resolved measurements with varying excitation density and photon energy

To better understand the dynamics leading to the population of CBM after photoexcitation, we perform further trARPES measurements with varying excitation fluence and energy at room temperature. For these data, we integrate the signal along Γ−X\Gamma- X and subtract a constant background due to the presence of a finite signal above the VBM overlapping with the exciton and CBM features (Figs. S12, S13). Figure 3a,Figure 3b shows dynamics of the exciton and CBM features integrated along Γ−X\Gamma- X with excitation close to the exciton resonance (hν=1.36h\nu= 1.36 eV, ∼\sim50 fs) for varying incident excitation fluences (nehn_{\mathrm{eh}}) between ∼\sim0.1 – 1.5 mJ/cm2^2. Under these conditions, we estimate photoexcited exciton densities of neh∼0.2−3×1013n_{\mathrm{eh}} \sim0.2 - 3 \times10^{13} cm−2^{-2} (see Methods). Similar to the 1.55 eV excitation, we observe a delayed rise of the CBM signal relative to the exciton. Interestingly, we also observe a strong dependence on excitation fluence in both the initial decay dynamics of the exciton signal (Figure 3a) and the rise time and relative intensity of the CBM feature (Figure 3b,Figure 3c). As the excitation density increases, the exciton signal features an increasingly faster decay component in the first few hundred femtoseconds. At the same time, the CBM signal rises more quickly, from a ∼\sim200 fs to ∼\sim 50 fs timescale, with the relative CBM signal intensity approaching ∼\sim50% of the exciton.

Pump excitation energy and fluence-dependent dynamics

Figure 3. Pump excitation energy and fluence-dependent dynamics. Dynamics for the a exciton and b CBM features integrated along Γ−X\Gamma- X for near-resonant exciton excitation (hν=1.36h\nu= 1.36 eV) with varying excitation densities (nehn_{\mathrm{eh}}) between 0.2 – 2.9×10132.9 \times10^{13} cm−2^{-2}. c, Fluence-dependent rise time and relative maximum intensity of the CBM to the exciton feature. Dynamics of the exciton (black) and CBM (red) features for d hν=1.94h\nu= 1.94 eV and e hν=3.10h\nu= 3.10 eV. f, Excitation energy-dependent rise time of the exciton signal. The time traces are integrated over 0.1 eV and normalized to the respective exciton (a, b) or CBM (d, e) signal after a constant background subtraction of the negative time delay signal.

For excitation with increasingly higher photon energies, we observe a reversal in the early-time dynamics and relative intensities of the CBM and exciton such that the CBM state is transiently populated first (Figure 3d,Figure 3e), followed by a rise of the exciton signature on a few-hundred femtosecond timescale (Figure 3f, S14). This observation is most prominent with 3.10 eV excitation, which populates electrons with excess energy far above the conduction band minimum and in which excitons form within a ∼300\sim300 fs timescale. Temperature-dependent dynamics on similar few-hundred femtosecond timescales have been attributed to electron-phonon coupling and carrier cooling after above-gap excitation[31]. Notably, the exciton formation and CBM dynamics display minimal excitation fluence dependence (Fig. S15), which may indicate that excitons form partially through a geminate process directly from photoexcitation, as opposed to from thermalized free carriers at the band edge[53, 54, 55]. After the initial hot electron relaxation dynamics to the CBM, bound excitons and conduction band electrons at the CBM coexist. Such a dynamic and robust exciton formation in CrSBr is likely due to the very high exciton binding energy that we observe in this material, similar to exciton formation in monolayer TMDs[56, 53]. For the excitation energies used in this study, 1.94 eV excitation features the sharpest rise of the CBM population, consistent with our conduction band assignment near ∼2\sim2 eV as well as recent estimates of the electronic band gap by previous ARPES measurements[37, 42, 41]. We note that a second excitonic signature with high oscillator strength at 1.77 eV has also been reported[57], in addition to a prominent optically dark exciton at 1.46 eV[58, 59, 60]. Additional data below TNT_{\mathrm{N}} is (Fig. S16) exhibits similar behavior.

The fluence-dependent decay dynamics after near-resonant exciton excitation (Figure 3a,Figure 3b) suggest that a multi-exciton scattering mechanism is responsible for the sub-picosecond dynamics under these excitation densities. At elevated quasiparticle densities, the electron-hole Coulomb interaction can become screened, reducing the binding energy and leading to a renormalization of the electronic bandgap, termed band gap renormalization (BGR)[61, 62]. At a critical threshold, NMN_{\mathrm{M}}, the exciton gas can undergo a phase transition to an ionized plasma of quasi-free carriers, referred to as the excitonic Mott transition. The nature of the Mott transition has been a subject of intense research[61, 63, 64, 65] and is critical to understand semiconductor materials such as 2D TMDs, whose excitons largely determine the optical response and electronic properties[8]. Considering the extracted effective exciton radius, we estimate the Mott threshold as NM=1/aeff2≈4×1014N_{\mathrm{M}} = 1/a_{\mathrm{eff}}^{2} \approx4 \times10^{14}, an order of magnitude greater than the fluences used in these experiments. Indeed, we observe a reduction of the exciton binding energy of only 58 meV ±\pm 4 meV at our highest fluence, indicating bound excitons remain stable in this regime (Figs. S21,S22).

Based on these considerations, an excitonic Mott transition is unlikely to be the primary driving force for the observed features and dynamics we observe. However, a separate mechanism, namely exciton-exciton annihilation (EEA), may also occur at elevated excitation densities below the Mott transition[66, 67]. EEA is an Auger-type, two-exciton nonradiative decay pathway and acts as a prominent loss channel for excitonic materials at high excitation densities. EEA has been widely studied in nanomaterials including TMDs[68, 69, 70], and also in quasi-1D anisotropic systems such as carbon nanotubes[71, 72, 73] and black phosphorus[74]. During EEA, one exciton non-radiatively recombines and leads to the dissociation of another exciton, promoting its constituent carriers higher and lower into the single-particle conduction and valence bands with excess energy, respectively. Experimentally, EEA manifests as second-order kinetics with respect to the excitation density, with an increasingly faster decay component dominating with elevated excitation fluence.

Global fitting the dynamics to an EEA-based rate-equation model

To gain further insight into the observed dynamics and test the role of EEA and exciton formation in CrSBr, we employ a coupled rate-equation model (Eqs. (1)-(3)) considering populations of excitons (NXN_{\mathrm{X}}), conduction band electrons at the CBM (NCBMN_{\mathrm{CBM}}), and hot carriers high above the CBM (NhcN_{\mathrm{hc}}) [75]. Exciton recombination (τr\tau_{\mathrm{r}}), two-body exciton-exciton annihilation (γEEA\gamma_{\mathrm{EEA}}), exciton formation from free carriers (τf\tau_{\mathrm{f}}), and hot carrier relaxation (τhc\tau_{\mathrm{hc}}) processes after photoexcitation are illustrated in Figure 4a. The constant FiF_i (HiH_i) is proportional to the fluence and is used to scale the excitation density in the case of exciton (above-gap) excitation, which is modelled as a Gaussian source term. We globally fit the exciton and conduction band electron dynamics for a range of excitation fluences to extract the above time and rate constants, considering both near-resonant exciton (1.36 eV) and above-gap (3.10 eV) excitation (Figure 4b-Figure 4e,S19). We include a free parameter to correct the relative amplitudes of each trace. The exciton recombination time constant is fixed at τr=20\tau_{\mathrm{r}} = 20 ps based on previous reports[30] and data acquired at longer delay times (Fig. S18).

Exciton and free carrier interconversion

Figure 4. Exciton and free carrier interconversion. a, After photoexcitation near the exciton resonance, exciton-exciton annihilation leads to quasi-free electrons near the CBM, while above-bandgap excitation populates hot carriers with excess energy which then relax to form excitons. b – e, Global fitting of the data, considering both 1.36 eV and 3.10 eV excitation, to a coupled rate-equation model reproduces the fluence- and excitation energy-dependent trends. f, Effective EEA timescales at different excitation fluences extracted from the EEA coefficient. The excitation density for 1.36 eV excitation (b – d) is 0.2, 0.8, and 2.9 × 101310^{13} cm−2^{-2}, respectively.

dNXdt=Fi×G−NXτr+NCBMτf−γEEANX2(1)\frac{dN_{\mathrm{X}}}{dt}=F_i \times G-\frac{N_{\mathrm{X}}}{\tau_{\mathrm{r}}}+\frac{N_{\mathrm{CBM}}}{\tau_{\mathrm{f}}}-\gamma_{\mathrm{EEA}}N_{\mathrm{X}}^2 \tag*{(1)}
dNCBMdt=Nhcτhc−NCBMτf+γEEANX22(2)\frac{dN_{\mathrm{CBM}}}{dt}=\frac{N_{\mathrm{hc}}}{\tau_{\mathrm{hc}}}-\frac{N_{\mathrm{CBM}}}{\tau_{\mathrm{f}}}+\gamma_{\mathrm{EEA}}\frac{N_{\mathrm{X}}^2}{2} \tag*{(2)}
dNhcdt=Hi×G−Nhcτhc(3)\frac{dN_{\mathrm{hc}}}{dt}=H_i \times G-\frac{N_{\mathrm{hc}}}{\tau_{\mathrm{hc}}} \tag*{(3)}

Overall, the EEA-based global model reproduces our observations well, including the evolution of the rise time of the CBM and the faster initial decay dynamics of the exciton with increasing excitation fluence in the case of near-resonant exciton excitation and the reversal when considering above-gap excitation. We extract time constants of τf=109±4\tau_{\mathrm{f}} = 109 \pm4 fs and τhc=383±12\tau_{\mathrm{hc}} = 383 \pm12 fs for exciton formation and hot carrier cooling, respectively. The retrieved exciton formation time τf\tau_{\mathrm{f}} is consistent with exciton formation times when considering different excitation energies near the CBM (Fig. S14). Note that in this model, we consider only exciton formation as a decay pathway for free carriers after either formation via EEA or hot carrier relaxation to the CBM. We determine an EEA rate of γEEA=0.090±0.002\gamma_{\mathrm{EEA}} = 0.090 \pm0.002 cm2^2/s, corresponding to EEA timescales of ~5.5 ps to ~370 fs for the excitation fluences used in our measurements (0.2 – 3 × 1013^{13} cm−2^{-2}). This EEA rate is similar to rates observed in monolayer TMDs such as exfoliated MoS2_2 (0.043 cm2^2/s) [69] and WS2_2 grown by chemical vapor deposition (0.089 cm2^2/s) [76], but substantially higher than 2D quantum well systems (~10−3^{-3} cm2^2/s) [70, 77]. Defects have been suggested to play a significant role in the nonradiative losses and exciton recombination dynamics, contributing to the observed annihilation rates.[68] We observe a static in-gap signal at equilibrium indicative of n-type doping (Figure S9).[78] An ultrafast defect-assisted exciton annihilation mechanism, as has also previously been in TMDs[79, 80], may therefore also be relevant in CrSBr. Strong exciton-phonon coupling in CrSBr[81] may also lead to enhanced phonon-assisted Auger-type decay pathways[82, 83]. Nevertheless, this EEA process therefore acts as a prominent nonradiative decay pathway for photoexcited excitons at elevated excitation densities, leading to a rapid decay of the total excited state population (Fig. S17). Together, the processes of ultrafast EEA and robust exciton formation lead to a strong competition and interconversion between quasi-free conduction band electrons and bound excitons on few hundred femtosecond timescales.

Sequential absorption of photons within the pump pulse may also be responsible for the transient CBM population after near-resonant exciton excitation, as previously reported for ReSe2\mathrm{ReSe}_{2}[84]. Exciton photoionization by sequential absorption is expected to be population conserving. However, we observe a stark fluence-dependent dynamics in the total excited state population persisting after the pulse duration (~50 fs), which is not reproduced by a model considering sequential absorption in place of EEA (Fig. S20), suggesting it is not the sole origin of the dynamics observed in our experiments.

We note that this EEA-based rate-equation model neglects effects originating from, for example, coherent electronic polarization dynamics or many-body effects which may be especially relevant at short time delays during, during and directly after optical excitation.[85] In addition, our model does not capture the complex correlations that arise from interactions with other degrees of freedom, such as the spin or lattice through coupling to magnons or phonons, and which influence the optical properties and excited state dynamics, especially in the magnetically ordered phase.

In CrSBr, the quasi-1D character may modify the exciton dynamics and EEA process. In a confined 1D system, excitons interact via their nearest neighbors, leading to a time-dependent EEA rate.[86] Excitons in CrSBr inherit the quasi-1D character and material anisotropy and are confined along the crystal b-axis. However, at elevated excitation densities the effect of this confinement on the EEA behavior may be reduced, as observed for black phosphorus[74], as the inter-exciton distance is reduced and interactions along the a-axis are non-negligible.

In this work, we provide for the first time a direct, momentum-resolved view of the strongly bound, anisotropic excitons in the bulk van der Waals magnetic semiconductor CrSBr. Using time-resolved momentum microscopy, we have extracted an exceedingly large exciton binding energy of ~800 meV for excitons probed on the surface of bulk CrSBr, and we retrieved their anisotropic real-space excitonic distribution, with radii of ~0.13 nm and ~0.55 nm along the crystalline a- and b-axes at T=120T = 120 K. These results support a picture of excitons confined along the quasi-1D Cr-S chains and extending along the crystal b-axis, with a weak dependence of key exciton properties on the magnetic state of the material. Upon near-resonant excitation of the lowest-energy exciton, a many-body decay channel, namely exciton-exciton annihilation, dominates with increasing excitation density above neh∼1013n_{eh} \sim10^{13} carriers/cm2^{2}, leading to the generation of quasi-free carriers and a rapid decay of the total excited state population on timescales of sub- to few-picoseconds. With substantially higher, above-gap excitation of quasi-free electrons, we observe the relaxation of hot electrons to the band edge before the formation of strongly bound excitons, which appear extremely robust in this system. The competition between the annihilation and formation of excitons drives the femtosecond and picosecond excited state dynamics after optical excitation in CrSBr, dominating over excitation-induced weakening of the exciton binding energy in described by the excitonic Mott transition. Our work provides important insight into the many-body interactions and physics underlying the strongly bound, quasi-1D excitons in CrSBr, critical for understanding and steering functionalities in next-generation opto-spintronic devices.

Methods

CrSBr synthesis

CrSBr bulk crystal was synthesized by a chemical vapor transport method. Elemental chromium (99.99%, -60 mesh, Chemsavers, USA), sulfur (99.9999%, 1 - 6 mm Wuhan Xinrong New Materials Co., China), and bromine (99.9999%, Merck, Czech Republic) were placed with a stoichiometry of 1:1:1 in a quartz ampoule (60x250mm). The amounts corresponded to 30g of CrSBr using 2 at.% excess of Br and S. The ampoule, cooled with liquid nitrogen, was melt sealed under high vacuum using an oil diffusion pump with liquid nitrogen trap. The ampoule was then heated in a crucible furnace at 400 °C for 50 hours, at 500°C for 50 hours and at 600°C for 50 hours. The top part of the ampoule was kept under 100 °C to avoid excessive pressure formation in the ampoule. The reacted ampoule was subsequently placed in a two zone furnace, where first the source zone was heated to 700 °C and the growth zone to 900 °C. After two days the thermal gradient was reversed and the source zone was kept at 900 °C and growth zone at 800 °C for 10 days. Finally, the ampoule was cooled to room temperature and opened inside an argon-filled glovebox.

ARPES Measurements

ARPES measurements were performed using a home-built, tabletop high-harmonic generation (HHG) source operating at ∼500\sim500 kHz and delivering probe pulses at ∼21.7\sim21.7 eV for photoemission[32, 33]. For momentum-microscopy-based trARPES experiments, the output of an optical parametric chirped pulse amplification (OPCPA) was used to drive HHG by tight focusing of its second harmonic (400 nm) in an argon gas jet in vacuum. A portion of the OPCPA output (800 nm, ∼40\sim40 fs) was used for pump pulses in time-resolved momentum microscopy experiments with 1.55 eV excitation. For other trARPES experiments, HHG was driven with 515 nm from the second harmonic of a portion of the 1030 nm output of a frontend laser (Carbide, Light Conversion). This 1030 nm output was also used to pump an OPA (Orpheus, Light Conversion) for tunable pump pulses (640-940 nm, ∼50\sim50 fs). With the OPCPA-based setup, the driving beam (400 nm) was separated from the generated XUV by an aperture in the far-field while in the second case, the XUV was isolated from the driving beam (515 nm) by reflection on a SiC substrate at Brewster’s angle. The HHG order of interest near ∼21.7\sim21.7 eV was isolated by a multilayer XUV mirror focusing the XUV onto the sample, as well as a 400-nm thick transmissive Sn filter. Before the measurements, bulk CrSBr crystals were mechanically cleaved in ultrahigh vacuum in base pressures better than ∼10−10\sim10^{-10} mbar. A six-axis manipulator (Carving, SPECS GmbH) is used to control the alignment of the sample to either a time-of-flight momentum microscope (METIS1000, SPECS GmbH) or a hemispherical electron energy analyzer (PHOIBOS150, SPECS GmbH)[34]. In all measurements, the XUV pulses were p-polarized while the pump pulses were s-polarized. The excitation density is calculated[87] considering the sample dielectric function[15] and pump wavelength, polarization, angle of incidence, and spot size (∼110 μm\sim110\,\mu\mathrm{m} full-width at half-maximum) on the sample. In our measurements, we estimate the temporal resolution from the pump-probe cross correlation to be ∼40\sim40 fs (400 nm HHG driver) and ∼80\sim80 fs (515 nm HHG driver). The experimental energy resolution is ∼120\sim120 meV (400 nm HHG driver) and ∼40\sim40 meV (515 nm HHG driver), dominated by the bandwidth of the XUV probe pulses.

Data Availability

The datasets generated and analyzed in this study are made available on a public data repository:

Acknowledgements

This work was supported by the Deutsche Forschungsgemeinschaft (DFG) within the Transregio TRR 227 Ultrafast Spin Dynamics (Project B07) and through the priority program SPP2244 (Project No. 443366970), the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (H2020-FETOPEN-2018-2019-2020-01, OPTOLogic – Grant Agreement No. 899794), and the Max Planck Society. L.T.L. acknowledges financial support from the Alexander von Humboldt Foundation. A.D.V. acknowledges support from BERLIN QUANTUM, an initiative endowed by the Innovation Promotion Fund of the city of Berlin. N.P.W., A.V.S. and J.J.F. acknowledge the German Science Foundation (DFG) for financial support via the Priority Programmes SPP 2244 “2D Materials Physics of van der Waals heterobilayer”, as well as the clusters of excellence MCQST (EXS-2111) and e-conversion (EXS-2089). Z.S. was supported by project LUAUS25268 from Ministry of Education, Youth and Sports (MEYS) and by the project Advanced Functional Nanorobots (reg. No. CZ.02.1.01/0.0/0.0/15_003/0000444 financed by the EFRR). The authors thank Florian Dirnberger and Philip Hofmann for fruitful discussions and Tim Sanetra for assistance in preliminary analysis. The CrSBr visualization in Figure 1 was produced using the VESTA software[88].

Author Contributions

R.E., A.V.S., T.P. and L.T.L. initiated the project. L.T.L., T.P., M.A.W, A.D.V., T.H.L.G.C., and A.N. performed ARPES experiments. F.M., N.P.W., K.M., and Z.S. provided bulk crystal samples. M.W., R.E., L.R., and J.J.F provided supervision and experimental infrastructure. L.T.L. analyzed the data and wrote the manuscript with input from all authors.

Competing Interests

The authors declare no competing interests.

Fig. 1: Excitons and single-particle excitations in CrSBr. a, trARPES of bulk CrSBr probes the dynamic momentum-space occupation I(kx,ky,E)I(k_x, k_y, E) as a function of pump-probe time delay, Δt\Delta t. Momentum maps I(kx,ky)I(k_x, k_y) at constant energy b, near the valence band maximum (VBM), and c, of the conduction band minimum (CBM) and d, of the lowest-energy exciton state after photoexcitation (hν=1.55h\nu= 1.55 eV) at T = 120 K. e, The real-space exciton wavefunction is retrieved by Fourier transformation of the exciton momentum map, showing highly anisotropic character with rms radii of ~0.13 nm and ~0.48 nm along the crystal aa- and bb-axis, respectively. E=0E = 0 eV is referenced to the VBM at the second Γ\Gamma point. Data shown are integrated over positive delay times (Δt=0\Delta t = 0 to Δt=1000\Delta t = 1000 fs) with an energy window of 200 meV.

Fig. 2: Anisotropic dispersion and ultrafast dynamics in CrSBr. a, Delay-integrated excited state signal along Γ−X\Gamma- X (ky=0k_y = 0 Å−1^{-1}) showing the flat dispersion of both the conduction band minimum (CBM) and the exciton. The energetic assignment of these states is extracted from a fit to the momentum-integrated EDC (violet) and the energy difference corresponds to the exciton binding energy, Eb∼800E_b \sim800 meV. b, Time traces at Γ\Gamma show that the CBM (red) exhibits a delayed rise compared to the exciton (black) after photoexcitation (hν=1.55h\nu= 1.55 eV). The time traces are both individually normalized and integrated in an energy and momentum window of 200 meV and 0.5 Å−1^{-1} in both kxk_x and kyk_y, respectively. Data is acquired at T = 300 K.

Fig. 3: Pump excitation energy and fluence-dependent dynamics. Dynamics for the a exciton and b CBM features integrated along Γ−X\Gamma- X for near-resonant exciton excitation (hν=1.36h\nu= 1.36 eV) with varying excitation densities (nehn_{\mathrm{eh}}) between 0.2 – 2.9 × 101310^{13} cm−2^{-2}. c, Fluence-dependent rise time and relative maximum intensity of the CBM to the exciton feature. Dynamics of the exciton (black) and CBM (red) features for d hν=1.94h\nu= 1.94 eV and e hν=3.10h\nu= 3.10 eV. f, Excitation energy-dependent rise time of the exciton signal. The time traces are integrated over 0.1 eV and normalized to the respective exciton (a, b) or CBM (d, e) signal after a constant background subtraction of the negative time delay signal.

Fig. 4: Exciton and free carrier interconversion. a, After photoexcitation near the exciton resonance, exciton-exciton annihilation leads to quasi-free electrons near the CBM, while above-bandgap excitation populates hot carriers with excess energy which then relax to form excitons. b – e, Global fitting of the data, considering both 1.36 eV and 3.10 eV excitation, to a coupled rate-equation model reproduces the fluence- and excitation energy-dependent trends. f, Effective EEA timescales at different excitation fluences extracted from the EEA coefficient. The excitation density for 1.36 eV excitation (b – d) is 0.2, 0.8, and 2.9 × 101310^{13} cm−2^{-2}, respectively.

References

  1. [1]Geim, A. K. & Grigorieva, I. V. Van der Waals heterostructures. Nature 499, 419–425 (2013).DOI
  2. [2]Schaibley, J. R. et al. Valleytronics in 2D materials. Nature Reviews Materials 1, 16055 (2016).DOI
  3. [3]Wilson, N. P., Yao, W., Shan, J. & Xu, X. Excitons and emergent quantum phenomena in stacked 2D semiconductors. Nature 599, 383–392 (2021).DOI
  4. [4]Chernikov, A. et al. Exciton binding energy and nonhydrogenic Rydberg series in monolayer WS₂. Phys Rev Lett 113, 076802 (2014).DOI
  5. [5]He, K. et al. Tightly Bound Excitons in Monolayer WSe₂. Phys. Rev. Lett. 113, 026803 (2014).DOI
  6. [6]Wang, G. et al. Colloquium : Excitons in atomically thin transition metal dichalcogenides. Reviews of Modern Physics 90, (2018).DOI
  7. [7]Goryca, M. et al. Revealing exciton masses and dielectric properties of monolayer semiconductors with high magnetic fields. Nat Commun 10, 4172 (2019).DOI
  8. [8]Mak, K. F. & Shan, J. Photonics and optoelectronics of 2D semiconductor transition metal dichalcogenides. Nature Photonics 10, 216–226 (2016).DOI
  9. [9]Mueller, T. & Malic, E. Exciton physics and device application of two-dimensional transition metal dichalcogenide semiconductors. npj 2D Materials and Applications 2, 29 (2018).DOI
  10. [10]Burch, K. S., Mandrus, D. & Park, J.-G. Magnetism in two-dimensional van der Waals materials. Nature 563, 47–52 (2018).DOI
  11. [11]Mak, K. F., Shan, J. & Ralph, D. C. Probing and controlling magnetic states in 2D layered magnetic materials. Nat Rev Phys 1, 646–661 (2019).DOI
  12. [12]Gong, C. & Zhang, X. Two-dimensional magnetic crystals and emergent heterostructure devices. Science 363, eaav4450 (2019).DOI
  13. [13]Wu, F. et al. Quasi-1D Electronic Transport in a 2D Magnetic Semiconductor. Advanced Materials 34, 2109759 (2022).DOI
  14. [14]Dirnberger, F. et al. Magneto-optics in a van der Waals magnet tuned by self-hybridized polaritons. Nature 620, 533–537 (2023).DOI
  15. [15]Klein, J. et al. The Bulk van der Waals Layered Magnet CrSBr is a Quasi-1D Material. ACS Nano 17, 5316–5328 (2023).DOI
  16. [16]Ziebel, M. E. et al. CrSBr: An Air-Stable, Two-Dimensional Magnetic Semiconductor. Nano Lett. 24, 4319–4329 (2024).DOI
  17. [17]Telford, E. J. et al. Layered Antiferromagnetism Induces Large Negative Magnetoresistance in the van der Waals Semiconductor CrSBr. Advanced Materials 32, 2003240 (2020).
  18. [18]Göser, O., Paul, W. & Kahle, H. G. Magnetic properties of CrSBr. Journal of Magnetism and Magnetic Materials 92, 129–136 (1990).DOI
  19. [19]Pei, F. et al. Surface-Sensitive Detection of Magnetic Phase Transition in Van Der Waals Magnet CrSBr. Advanced Functional Materials 34, 2309335 (2024).DOI
  20. [20]Wilson, N. P. et al. Interlayer electronic coupling on demand in a 2D magnetic semiconductor. Nat. Mater. 20, 1657–1662 (2021).DOI
  21. [21]Bae, Y. J. et al. Exciton-coupled coherent magnons in a 2D semiconductor. Nature 609, 282–286 (2022).DOI
  22. [22]Diederich, G. M. et al. Tunable interaction between excitons and hybridized magnons in a layered semiconductor. Nat. Nanotechnol. 1–6 (2022) doi:10.1038/s41565-022-01259-1.DOI
  23. [23]Sun, Y. et al. Dipolar spin wave packet transport in a van der Waals antiferromagnet. Nat. Phys. 1–7 (2024) doi:10.1038/s41567-024-02387-2.DOI
  24. [24]Heißenbüttel, M.-C. et al. Quadratic optical response to a magnetic field in the layered magnet CrSBr. Phys. Rev. B 111, 075107 (2025).DOI
  25. [25]Brennan, N. J., Noble, C. A., Tang, J., Ziebel, M. E. & Bae, Y. J. Important Elements of Spin-Exciton and Magnon-Exciton Coupling. ACS Phys. Chem Au 4, 322–327 (2024).DOI
  26. [26]Šmier­tka, M. et al. Distinct magneto-optical response of Frenkel and Wannier excitons in CrSBr. Nat Commun https://doi.org/10.1038/s41467-026-68482-5 (2026) doi:10.1038/s41467-026-68482-5.
  27. [27]Wang, T. et al. Magnetically-dressed CrSBr exciton-polaritons in ultrastrong coupling regime. Nat Commun 14, 5966 (2023).arxiv.org/abs/2302.07616
  28. [28]Ruta, F. L. et al. Hyperbolic exciton polaritons in a van der Waals magnet. Nat Commun 14, 8261 (2023).DOI
  29. [29]Datta, B. et al. Magnon-mediated exciton–exciton interaction in a van der Waals antiferromagnet. Nat. Mater. 24, 1027–1033 (2025).DOI
  30. [30]Meineke, C. et al. Ultrafast Exciton Dynamics in the Atomically Thin van der Waals Magnet CrSBr. Nano Lett. 24, 4101–4107 (2024).DOI
  31. [31]Jiang, P. et al. Ultrafast Electron Dynamics Dominated by Electron–Phonon Coupling in CrSBr Revealed by Photoemission Electron Microscopy. J. Phys. Chem. C https://doi.org/10.1021/acs.jpcc.4c07231 (2024) doi:10.1021/acs.jpcc.4c07231.
  32. [32]Puppin, M. et al. 500 kHz OPCPA delivering tunable sub-20 fs pulses with 15 W average power based on an all-ytterbium laser. Opt. Express 23, 1491 (2015).DOI
  33. [33]Puppin, M. et al. Time- and angle-resolved photoemission spectroscopy of solids in the extreme ultraviolet at 500 kHz repetition rate. Review of Scientific Instruments 90, 023104 (2019).
  34. [34]Maklar, J. et al. A quantitative comparison of time-of-flight momentum microscopes and hemispherical analyzers for time- and angle-resolved photoemission spectroscopy experiments. Review of Scientific Instruments 91, 123112 (2020).
  35. [35]Zhang, H. et al. Angle-resolved photoemission spectroscopy. Nat Rev Methods Primers 2, 54 (2022).
  36. [36]Karni, O., Esin, I. & Dani, K. M. Through the Lens of a Momentum Microscope: Viewing Light-Induced Quantum Phenomena in 2D Materials. Advanced Materials 35, 2204120 (2023).DOI
  37. [37]Bianchi, M. et al. Paramagnetic Electronic Structure of CrSBr: Comparison between Ab Initio GW Theory and Angle-Resolved Photoemission Spectroscopy. Phys. Rev. B 107, 235107 (2023).DOI
  38. [38]Bianchi, M. et al. Charge transfer induced Lifshitz transition and magnetic symmetry breaking in ultrathin CrSBr crystals. Phys. Rev. B 108, 195410 (2023).DOI
  39. [39]Linhart, W. M. et al. Optical markers of magnetic phase transition in CrSBr. J. Mater. Chem. C 11, 8423–8430 (2023).arxiv.org/abs/2303.17926
  40. [40]Lin, K. et al. Probing the Band Splitting near the Γ Point in the van der Waals Magnetic Semiconductor CrSBr. J. Phys. Chem. Lett. 15, 6010–6016 (2024).DOI
  41. [41]Smolenski, S. et al. Large exciton binding energy in a bulk van der Waals magnet from quasi-1D electronic localization. Nat Commun 16, 1134 (2025).
  42. [42]Watson, M. D. et al. Giant exchange splitting in the electronic structure of A-type 2D antiferromagnet CrSBr. npj 2D Mater Appl 8, 54 (2024).DOI
  43. [43]Ramasubramaniam, A., Pérez, D. H., Junquera, J. & Camarasa-Gómez, M. Accurate electronic and optical properties of bulk antiferromagnet CrSBr via a tuned hybrid density functional with on-site corrections. Preprint at https://doi.org/10.48550/arXiv.2603.13494 (2026).
  44. [44]Christiansen, D., Selig, M., Malic, E., Ernstorfer, R. & Knorr, A. Theory of exciton dynamics in time-resolved ARPES: Intra- and intervalley scattering in two-dimensional semiconductors. Physical Review B 100, (2019).DOI
  45. [45]Dong, S. et al. Direct measurement of key exciton properties: Energy, dynamics, and spatial distribution of the wave function. Natural Sciences 1, e10010 (2021).
  46. [46]Shao, Y. et al. Magnetically confined surface and bulk excitons in a layered antiferromagnet. Nat. Mater. 1–8 (2025) doi:10.1038/s41563-025-02129-6.DOI
  47. [47]Liebich, M. et al. Controlling Coulomb correlations and fine structure of quasi-one-dimensional excitons by magnetic order. Nat. Mater. 1–7 (2025) doi:10.1038/s41563-025-02120-1.DOI
  48. [48]Reutzel, M., Jansen, G. S. M. & Mathias, S. Probing excitons with time-resolved momentum microscopy. Advances in Physics: X 9, 2378722 (2024).DOI
  49. [49]Qian, T.-X., Zhou, J., Cai, T.-Y. & Ju, S. Anisotropic electron-hole excitation and large linear dichroism in the two-dimensional ferromagnet CrSBr with in-plane magnetization. Phys. Rev. Res. 5, 033143 (2023).DOI
  50. [50]Semina, M. A. et al. Excitons and trions in CrSBr bilayers. Phys. Rev. B 111, 205301 (2025).DOI
  51. [51]Man, M. K. L. et al. Experimental measurement of the intrinsic excitonic wave function. Sci. Adv. 7, eabg0192 (2021).DOI
  52. [52]Li, Q. et al. Two-Dimensional Magnetic Exciton Polariton with Strongly Coupled Atomic and Photonic Anisotropies. Phys. Rev. Lett. 133, 266901 (2024).DOI
  53. [53]Trovatello, C. et al. The ultrafast onset of exciton formation in 2D semiconductors. Nat Commun 11, 5277 (2020).DOI
  54. [54]Han, X. et al. Ultrafast Exciton Formation in Perovskite Quantum Rods. Nano Lett. 25, 4913–4920 (2025).DOI
  55. [55]Mourzidis, K. et al. Exciton Formation in Two-Dimensional Semiconductors. Phys. Rev. X 15, 031078 (2025).arxiv.org/abs/2509.06543
  56. [56]Steinleitner, P. et al. Direct Observation of Ultrafast Exciton Formation in a Monolayer of WSe2. Nano Lett. 17, 1455–1460 (2017).
  57. [57]Komar, R. et al. Colossal magneto-excitonic effects in 2D van der Waals magnetic semiconductor CrSBr. Preprint at http://arxiv.org/abs/2409.00187 (2024).
  58. [58]Sears, J. et al. Observation of Anisotropic Dispersive Dark-Exciton Dynamics in CrSBr. Phys. Rev. Lett. 135, 146503 (2025).DOI
  59. [59]Meng, Z. et al. Flat-band mechanism for strongly bound dark excitons in two-dimensional magnetic semiconductors. npj Comput Mater 12, 80 (2026).DOI
  60. [60]Bork, S. et al. Excitonic optical interface for GHz–THz collective excitations in a van der Waals magnet.
  61. [61]Chernikov, A., Ruppert, C., Hill, H. M., Rigosi, A. F. & Heinz, T. F. Population inversion and giant bandgap renormalization in atomically thin WS2 layers. Nature Photonics 9, 466–470 (2015).DOI
  62. [62]Pogna, E. A. et al. Photo-Induced Bandgap Renormalization Governs the Ultrafast Response of Single-Layer MoS2. ACS Nano 10, 1182–8 (2016).DOI
  63. [63]Steinhoff, A. et al. Exciton fission in monolayer transition metal dichalcogenide semiconductors. Nature Communications 8, 1166 (2017).
  64. [64]Guerci, D., Capone, M. & Fabrizio, M. Exciton Mott transition revisited. Phys. Rev. Mater. 3, 054605 (2019).
  65. [65]Siday, T. et al. Ultrafast Nanoscopy of High-Density Exciton Phases in WSe2. Nano Lett. 22, 2561–2568 (2022).DOI
  66. [66]Linardy, E. et al. Harnessing Exciton–Exciton Annihilation in Two-Dimensional Semiconductors. Nano Lett. 20, 1647–1653 (2020).DOI
  67. [67]Uddin, S. Z., Rabani, E. & Javey, A. Universal Inverse Scaling of Exciton–Exciton Annihilation Coefficient with Exciton Lifetime. Nano Lett. 21, 424–429 (2021).DOI
  68. [68]Steinhoff, A., Jahnke, F. & Florian, M. Microscopic theory of exciton-exciton annihilation in two-dimensional semiconductors. Phys. Rev. B 104, 155416 (2021).DOI
  69. [69]Sun, D. et al. Observation of rapid exciton-exciton annihilation in monolayer molybdenum disulfide. Nano Lett 14, 5625–9 (2014).DOI
  70. [70]Yuan, L., Wang, T., Zhu, T., Zhou, M. & Huang, L. Exciton Dynamics, Transport, and Annihilation in Atomically Thin Two-Dimensional Semiconductors. J. Phys. Chem. Lett. 8, 3371–3379 (2017).DOI
  71. [71]Ma, Y.-Z., Valkunas, L., Dexheimer, S. L., Bachilo, S. M. & Fleming, G. R. Femtosecond Spectroscopy of Optical Excitations in Single-Walled Carbon Nanotubes: Evidence for Exciton-Exciton Annihilation. Phys. Rev. Lett. 94, 157402 (2005).pubmed.ncbi.nlm.nih.gov/15904186
  72. [72]Valkunas, L., Ma, Y.-Z. & Fleming, G. R. Exciton-exciton annihilation in single-walled carbon nanotubes. Phys. Rev. B 73, 115432 (2006).DOI
  73. [73]Wang, F., Wu, Y., Hybertsen, M. S. & Heinz, T. F. Auger recombination of excitons in one-dimensional systems. Phys. Rev. B 73, 245424 (2006).DOI
  74. [74]Pareek, V., Madéo, J. & Dani, K. M. Ultrafast Control of the Dimensionality of Exciton-Exciton Annihilation in Atomically Thin Black Phosphorus. Phys. Rev. Lett. 124, 057403 (2020).DOI
  75. [75]Burgos-Caminal, A., Socie, E., Bouduban, M. E. F. & Moser, J.-E. Exciton and Carrier Dynamics in Two-Dimensional Perovskites. J. Phys. Chem. Lett. 11, 7692–7701 (2020).
  76. [76]Cunningham, P. D., McCreary, K. M. & Jonker, B. T. Auger Recombination in Chemical Vapor Deposition-Grown Monolayer WS2. J. Phys. Chem. Lett. 7, 5242–5246 (2016).pubmed.ncbi.nlm.nih.gov/27973899
  77. [77]Taylor, R. A., Adams, R. A., Ryan, J. F. & Park, R. M. Exciton recombination dynamics in ZnCdSeZnSe<math><mtext>ZnCdSe</mtext><mtext>ZnSe</mtext></math> quantum wells. Journal of Crystal Growth 159, 822–825 (1996).
  78. [78]Biktagirov, T. et al. Intrinsic defects as a source of n-type conductivity in CrSBr. npj 2D Mater Appl 9, 106 (2025).DOI
  79. [79]Wang, H., Zhang, C. & Rana, F. Ultrafast Dynamics of Defect-Assisted Electron–Hole Recombination in Monolayer MoS₂. Nano Lett. 15, 339–345 (2015).DOI
  80. [80]Wang, H. et al. Fast exciton annihilation by capture of electrons or holes by defects via Auger scattering in monolayer metal dichalcogenides. Phys. Rev. B 91, 165411 (2015).DOI
  81. [81]Lin, K. et al. Strong Exciton–Phonon Coupling as a Fingerprint of Magnetic Ordering in van der Waals Layered CrSBr. ACS Nano 18, 2898–2905 (2024).DOI
  82. [82]Bushick, K. & Kioupakis, E. Phonon-Assisted Auger-Meitner Recombination in Silicon from First Principles. Phys. Rev. Lett. 131, 076902 (2023).DOI
  83. [83]Scharf, B. & Perebeinos, V. Phonon-Assisted Auger Decay of Excitons in Doped Transition Metal Dichalcogenide Monolayers. Preprint at http://arxiv.org/abs/2408.00097 (2024).
  84. [84]Guislain, B. G. et al. Isolating Exciton Dissociation Pathways in ReSe$_{ \text{2}}$. Preprint at https://doi.org/10.48550/ARXIV.2604.11906 (2026).
  85. [85]Luo, L. et al. Ultrafast terahertz snapshots of excitonic Rydberg states and electronic coherence in an organometal halide perovskite. Nat Commun 8, 15565 (2017).DOI
  86. [86]Srivastava, A. Diffusion-limited exciton-exciton annihilation in single-walled carbon nanotubes: A time-dependent analysis. Phys. Rev. B 79, (2009).DOI
  87. [87]Dendzik, M. et al. Observation of an Excitonic Mott Transition Through Ultrafast Core-cum-Conduction Photoemission Spectroscopy. Phys. Rev. Lett. 125, 096401 (2020).
  88. [88]Momma, K. & Izumi, F. VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data. Journal of Applied Crystallography 44, 1272–1276 (2011).DOI

Paper details

vlnc:W_axr2etzz
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents