betaFeedback

WorksPhysics and Astronomy

Phonon-modulated Kerr nonlinearity in ultrathin 2H-MoTe2

Abstract

Controlling nonequilibrium responses in optically driven quantum materials is essential for advancing applications in energy conversion, ultrafast electronics, and quantum computation. Nonlinear optical spectroscopy serves as a powerful tool to investigate ultrafast electron and phonon dynamics in these systems; however, conventional nonlinear approaches often require intense laser pulses (> 10 GW/cm2) and typically encounter a strong background. Here, we introduce a phase-sensitive nonlinear spectroscopic technique that operates at low laser powers (~ 10 kW/cm2, pulse energies ~ 10 pJ) and enables real-time monitoring and active control of coherent phonons in a few-layer (three to five) thick 2H-MoTe2. Upon excitation with ultrashort (~ 10 fs) pump pulses, we achieve displacive excitation of coherent phonons, which periodically modulate the Kerr nonlinearity of the material, leading to cross-phase modulation (XPM) of a delayed probe pulse. This phase modulation induces spectral broadening and oscillations in the center of mass (COM) of the probe spectrum in time, enabling the detection of subtle nonlinear optical responses in a background-free manner. The nonlinear response can be selectively amplified or attenuated by adjusting the strength of the pump pulse, which controls the distribution of photoexcited carriers in the electronic bands. By combining two-color nondegenerate pump-probe measurements and time-dependent density-functional theory (TDDFT) calculations, we directly resolve the coupled nonequilibrium electronic and phonon dynamics. A dual-pump pulse scheme enables precise control of phonon oscillations, allowing selective activation or suppression of specific phonon modes and correspondingly the modulation of the Kerr nonlinearity.

Ultrafast optical control of quantum materials provides a powerful approach to access nonequilibrium states of matter[1, 2], such as light-induced superconductivity[3, 4], magnetism[5], and topological phases[6, 7]. Understanding these transient phenomena requires insights into the complex interplay between electronic excitations and coherent phonon motions, yet, state-of-the-art techniques struggle to disentangle these dynamics at their natural timescales while providing phase-sensitive information.

Nonlinear optical processes, e.g., the Kerr effect—a third-order nonlinear (χ(3)\chi^{(3)}) response leading to intensity-dependent modulation of the refractive index—are crucial in ultrafast signal processing and in the investigation of complex light-matter interactions[8, 9, 10, 11]. In particular, cross-phase modulation (XPM), where a ‘pump’ pulse-induced nonlinearity modulates the phase of a time-delayed ‘probe’ pulse via the nonlinear Kerr effect, offers a highly phase-sensitive approach for tracking both coherent and incoherent quasiparticle dynamics[12]. In contrast to self-phase modulation (SPM), which entangles various nonlinear interactions within the material[13], XPM provides a phase-locked detection tool of dynamics evolving in real-time. In this process, the intensity, polarization, and the frequency of the probe pulse can be dynamically modulated via interactions with the material pre-excited by an ultrafast pump pulse[14, 15, 16].

Leveraging XPM for ultrafast material control is particularly compelling in layered transition metal dichalcogenides (TMDs), which exhibit pronounced Kerr nonlinearities alongside exceptional optical and electronic properties[17, 18, 19]. Among them, MoTe2_{2} is distinguished by its rich nonequilibrium states and structural phase transitions under intense optical pulse excitation[20, 21, 22, 23], making it a promising platform for all-optical modulation and control.

Phonon-mediated cross-phase modulation

In this work, we introduce a phase-sensitive XPM spectroscopic technique that directly resolves and controls femtosecond electron and phonon dynamics in a few-layer thick 2H2H-MoTe2_2. Ultrashort laser pulses generated by an ultra-broadband Ti:Sapphire oscillator were split into two spectrally separated ‘pump’ and ‘probe’ pulses using an 800 nm shortpass dichroic mirror (DMSP), Figure 1a. The pump (λ\lambda ∼\sim800–950 nm, ∼\sim10 fs long) and probe pulses (λ\lambda ∼\sim700–750 nm, ∼\sim20 fs long) were further spectrally filtered using bandpass (BP) filters to avoid any spectral overlap and interference artifacts between them, their respective spectra are shown in Figure 1b. The probe pulse was delayed using a precision motorized linear stage before being recombined with the pump pulse. The collinearly propagating pump and probe pulses were then focused onto the sample using a reflective objective (numerical aperture ∼\sim 0.5). The light emanating from the sample was collected using the same objective and passed through a 690 nm shortpass filter (SP) before being guided into a spectrometer.

Phase-sensitive nonlinear spectroscopy of ultrathin 2H-MoTe2

Figure 1. Phase-sensitive nonlinear spectroscopy of ultrathin 2H-MoTe₂. a, Schematic illustration of phase-sensitive nonlinear pump-probe spectroscopy in a chemical vapor deposition (CVD) grown few-layer (three to five) thick 2H-MoTe₂ on a quartz substrate. Collinearly propagating nondegenerate pump (~ 800-950 nm, ~ 10 fs long) and probe (~ 700-750 nm, ~ 20 fs long) laser pulses were focused onto the sample using a reflective objective. The light emanating from the sample was collected using the same objective, filtered through a ~ 690 nm shortpass filter to remove residual incident laser components, before being focused into the spectrometer. BP: Bandpass filter, SP: Shortpass filter, DMSP: Shortpass dichroic mirror, DMLP: Longpass dichroic mirror. b, Spectra of the incoming pump (red filed curve) and the probe (orange filed curve) pulses, shown on the right side of the dashed black curve, together with the measured spectral broadening of the probe pulse induced by cross-phase modulation (XPM, green filled curve) and the four-wave mixing (FWM, blue filled curved) processes on the left side of the dashed black curve. Left inset: schematic of the FWM contribution to the measured spectra, resulting from interaction of two photons from the probe pulse (upward transitions) ensued by one photon interaction from the pump pulse (downward transition). Right inset: Spectral-domain illustration of the self-phase modulation (SPM), XPM, and the FWM processes. Incoming pump and probe pulses undergo SPM as well as XPM, resulting in broadened spectra. Spectral components present on the left side of the dashed black curve were measured in the experiments, comprising of SPM of the probe pulse, XPM, and the FWM signals. SPM of the pump pulse was not measured as it lies outside the measured spectral window.

A probe pulse propagating through the 2H2H-MoTe2_2 sample experiences a time-varying refractive index[24] owing to the Kerr effect, thus, inducing a time-dependent phase shift across the pulse envelope[25]:

Δϕ(t)=2πλn2I(t)L(1)\Delta\phi(t) = \frac{2\pi}{\lambda} n_2 I(t)L \tag*{(1)}

where λ\lambda is the wavelength of incident light, n2n_2 is the nonlinear refractive index coefficient of 2H2H-MoTe2_2, I(t)I(t) is the instantaneous intensity of the probe pulse, and LL is the effective interaction length within the material. This phase shift leads to spectral broadening (see Fig. S1), characterized by the instantaneous frequency change:

δω(t)=−∂Δϕ(t)∂t(2)\delta\omega(t) = -\frac{\partial\Delta\phi(t)}{\partial t} \tag*{(2)}

When both the pump and probe pulses co-propagate through the nonlinear medium, the phase shift experienced by the probe pulse is influenced by the combined intensity of both pulses. Consequently, the time-dependent phase shift acquired by the probe pulse becomes:

Δϕ(t)=2πλn2(Iprobe(t)+κIpump(t))L(3)\Delta\phi(t) = \frac{2\pi}{\lambda} n_2 \left(I_{\mathrm{probe}}(t)+\kappa I_{\mathrm{pump}}(t)\right)L \tag*{(3)}

where κ\kappa is a coefficient, often close to 2 in different materials[25, 26, 27], indicating that the pump pulse exerts a stronger influence on the probe pulse, than the probe pulse does on itself (SPM). This interaction gives rise to an intense XPM signal at the zero time delay between the pulses, as shown by the green-shaded spectral region in Figure 1b and the inset (right) sketch. Moreover, when the pump and probe pulses temporally overlap, four-wave mixing (FWM) processes also occur. One prominent case involving two photons from the probe pulse and one photon from the pump pulse results in the blue-shaded spectrum in Figure 1b, as depicted in the inset (left).

The nonlinear XPM and FWM responses were characterized by varying the delay between the pump and the probe pulses. Figure 2a shows a series of pump-probe spectra, in which the individual spectra were subtracted with a spectrum measured at a far negative delay of -1 ps between the pulses, effectively removing the SPM signal generated by the probe pulse alone. Owing to the broad spectral separation between the pump and probe pulses, SPM induced solely by the pump pulse is not detected. At zero delay, pronounced nonlinear signals emerge, corresponding to XPM (~ 670–690 nm) and FWM (~ 620–670 nm). Following the initial pump-probe overlap, the XPM signal exhibits a sharp drop in its amplitude followed by periodic oscillations lasting up to ~ 10 ps. The initial decrease and recovery of the XPM signal at longer delays is attributed to an ultrafast electronic response, where depletion of the carriers in the valence band upon photoexcitation with the pump pulses reduces the Kerr nonlinearity[28].

Figure 2. Ultrafast modulation of Kerr nonlinearity via coherent phonons. a, A series of spectra measured as a function of the delay between the pump and the probe pulses. All the spectra were subtracted by the spectrum recorded at -1 ps delay, in order to remove the contribution of the SPM of the probe pulse, to generate a purely XPM spectral trace. SPM of the pump pulse was not measured as it lies relatively far from the edge of the measured spectral window. Inset shows the close up of the measurement after the zero time delay. b, Temporal cross-cut of the XPM spectra at ∼680\sim680 nm from the measurement shown in a. Pump and probe laser powers were set to be ∼2.1\sim2.1 mW (26 pJ) and ∼3.0\sim3.0 mW (37.5 pJ), respectively. Inset: simulation (yellow) of the experimental time trace (orange) by summing up two damped oscillations, representing the A1g\mathrm{A}_{1g} (red) and A1b\mathrm{A}_{1b} (blue) phonon modes. c, Top panel: Fast Fourier transform (FFT) of the time trace in b in the delay range of 0.1 ps to 10 ps. The FFT spectrum reveals a strong peak at ∼172\sim172 cm−1^{-1}, corresponding to the A1g\mathrm{A}_{1g} phonon mode of 2H2H-MoTe2\mathrm{MoTe}_2, and a broad peak at ∼480\sim480 cm−1^{-1}, attributed to the A1b\mathrm{A}_{1b} phonon mode of the quartz substrate. Bottom panel: Raman spectrum of the 2H2H-MoTe2\mathrm{MoTe}_2 sample on quartz, measured with a 633 nm CW laser excitation source. A laser power of 0.89 mW with an acquisition time of 20 s was used in the measurement. d, Simulated pump-probe spectra incorporating both XPM and FWM contributions. Inset shows the close-up of the simulation after the zero delay. e, Schematic illustration of the displacive excitation of the coherent phonons. Upon excitation with the pump pulse (red), the potential energy surface (PES) of the system shifts within the Born–Oppenheimer approximation, resulting into a new equilibrium position (Q0′Q_0') of the nuclei. This sudden shift in the PES exerts a fully symmetrical force, FF, on the lattice, initiating coherent phonon oscillations (yellow). The PES (orange) and equilibrium position of the nuclei (purple) evolve in time.

To analyze the oscillatory behavior in detail, the temporal crosscut of the pump-probe XPM spectra at ~ 680 nm is plotted in Figure 2b. The time-trace exhibits a strong peak at the zero delay, followed by an abrupt reduction and a subsequent recovery within ~1 ps, superimposed with long-lived oscillations. Fast Fourier transform (FFT) analysis of this time-domain signal reveals a pronounced peak at ~ 172 cm−1^{-1} (upper panel in Figure 2c), corresponding to the out-of-plane A1gA_{1g} phonon mode of 2H2H-MoTe2_2. This mode exhibits a long coherence lifetime (~ 5.0 ps), consistent with previous transient reflection measurements[29]. In addition, our data also reveals a broad peak centered at ~ 480 cm−1^{-1}, indicating a short-lived phonon mode (~ 0.4 ps, see also the time-frequency analysis of the pump-probe traces in Fig. S2 in SI). This spectral peak is attributed to the A1bA_{1b} mode of the quartz substrate, a ring Obridge_{\text{bridge}}-breathing phonon mode[30].

Apart from these dominant phonon modes, we also observed a weak spectral peak at ~ 202 cm−1^{-1} (Figure 2c) that matches the A1rA_{1r} (rotation mode of SiO4_4 tetrahedra) mode of quartz[30]. These mode attributions were further confirmed by Raman spectroscopic characterization performed on the same sample (lower-panel in Figure 2c). Notably, few-layer 2H2H-MoTe2_2 transferred onto a Si3_3N4_4 substrate shows neither the A1bA_{1b} nor the A1rA_{1r} mode (see Fig. S3 in SI). The presence of the phonon modes of the quartz substrate in the time-resolved measurements corroborates the exceptional phase sensitivity of the present XPM-based technique.

A cosine-like oscillation can be assigned to both the A1gA_{1g} mode of 2H2H-MoTe2_2 and A1bA_{1b} mode of quartz by fitting the time-domain signal backwards to the zero delay with phases of (0.02 ± 0.07)π, and (1.02 ± 0.05)π, respectively, as shown in the inset of Figure 2b. The absolute maximum of the FWM signal in the measurements enables clocking of the zero time-delay between the pump and the probe pulses. The opposite phases at the zero-delay of the two phonon modes likely results from their distinct eigenvector orientations and differences in the underlying band structures[30, 31]. Notably, the strongest Raman-active E2gE_{2g} mode of 2H2H-MoTe2_2 at ~240 cm−1^{-1} is absent in the time-resolved XPM measurements (Figure 2c). A similar behavior was observed in the few-layer thick 1T1T-MoTe2_2, where the AgA_g mode at ~ 253 cm−1^{-1} is observed without the presence of the corresponding E2gE_{2g} mode (see Fig. S3 in SI). Coherent phonon oscillations corresponding to the AgA_g mode were also measured for a monolayer 2H2H-MoTe2_2 mechanically exfoliated onto a z-cut sapphire substrate (see Fig. S4 in SI). The selective mode excitation of fully symmetric AgA_g modes and their cosine-like oscillatory behavior suggest a displacive excitation mechanism of phonons in MoTe2_2 and the underlying quartz substrate[32], as schematically shown in Figure 2e. Illumination with ultrafast laser pulses induces an instantaneous redistribution of electronic charge in 2H2H-MoTe2_2. This rapid electronic redistribution alters the potential energy surface (PES), UU, on attosecond timescales[30], causing a shift in the equilibrium position (Q0Q_0) of the lattice within the adiabatic Born–Oppenheimer approximation. As a result, atoms are impulsively displaced from their original equilibrium positions and subsequently oscillate coherently around the new equilibrium position (Q0′Q'_0) defined by the transient electronic state[31].

The oscillations in the XPM signal at the phonon frequencies of 2H2H-MoTe2_2 and quartz indicates that the pump pulses excite coherent phonons, which are subsequently detected by the probe-pulse via phase-sensitive nonlinear interactions. This interpretation is further attested by time-dependent density functional theory (TDDFT) calculations, see Fig. S5 and SI for details. The probe pulse on interacting with the coherently oscillating lattice acquires a periodic phase modulation, leading to a periodic shift in the COM of its spectral weight. This spectral shift manifests experimentally as oscillations in the broadened spectral intensity of the probe-pulse[12]. By incorporating both FWM and XPM contributions into the simulations, the key features in the pump-probe spectra can be reasonably reproduced, as shown in Figure 2d. The transient modulation of the Kerr nonlinearity by the pump pulse can be attributed to photoexcitation of the carriers and the resulting modifications of the electronic band structure[9, 28], which will be discussed later in the text. In the simulation shown in Figure 2d, the ultrafast electronic depletion is modelled using a step-like response followed by an exponential decay[9, 28, 32], see methods for details.

Presence of extensive dynamical features in the pump-probe traces demonstrate that the probe pulse is highly sensitive to pump-induced modifications of the dielectric function of the material[33]. Compared to other nonlinear techniques, such as high-harmonic generation[30, 34], transient reflection and transmission spectroscopy[35], this phase-sensitive approach operates at substantially lower pulse energies (<10< 10 pJ) while providing superior phase sensitivity. As a result, it offers a powerful tool for resolving quasiparticle interactions, exciton formation, and their relaxation dynamics in quantum materials.

Displacive excitation of symmetric phonon modes

To elucidate the mechanism underlying the coherent phonon excitation, we performed fluence-dependent measurements. First, we varied the fluence of the probe pulses while keeping the pump fluence fixed. The spectral intensities of the retrieved phonon modes, A1gA_{1g} and A1bA_{1b}, exhibit a quadratic dependence on the probe fluence, as shown in Figure 3a. Similarly, the amplitude of the depleted XPM signal at the zero pump-probe delay, also exhibits a quadratic dependence on the probe fluence (blue curve, Figure 3b). The retrieved quadratic scaling of the intensity of the phonon modes rules out the FWM (or coherent anti-Stokes Raman (CARS)) process involving two photons from the pump pulse and one photon from the probe pulse as the origin of the vibrational coherence (see also Fig. S6)[36, 37]. It is worth mentioning that the anti-Stokes Raman signal would be fully suppressed in our experimental configuration (see SI for details). The observed FWM process at ∼650\sim650 nm, involving two photons from the probe pulse and one photon from the pump pulse, exhibits the expected quadratic scaling on probe fluence (see Fig. S7). The nonlinear dependence of both the XPM response and phonon amplitudes on probe fluence strongly supports the nonlinear XPM mechanism. It is worth noting that the decay constants associated with the relaxation times of the photocarriers exhibit a decreasing trend on increasing probe fluence (red curve in Figure 3b), reflecting the role of enhanced carrier-carrier interactions at higher photoexcitation intensities on the retrieved relaxation times.

Six-panel plots showing the evolution of Kerr nonlinearity with probe and pump power

Figure 3. Evolution of Kerr nonlinearity with the fluence of the driving pulses and displacive excitation of coherent phonons. a, Variation of the spectral intensities of the A1gA_{1g} (red dots) and A1bA_{1b} (blue dots) phonon modes on increasing the fluence of the probe pulse. Phonon amplitudes were obtained from the FFT analysis of the time-resolved pump-probe traces (see Fig. S8 in Supplementary Information). Red and blue curves represent fits in the dual logarithmic plots, with fit parameters annotated in the respective legends. The power of the pump pulse was kept at 2.2 mW, and an acquisition time of 0.5 s per delay step was used for all the spectra. b, Variation of the amplitudes of the XPM signal at the zero delay between the pump and probe pulses (blue dots), and the decay time constants (red dots) on increasing the fluence of the probe pulse, extracted from the time traces shown in Fig. S8. c, A series of time-resolved XPM measurements with increasing fluence of the pump pulse, the XPM signal was enhanced (red area) at low pump fluence but becomes depleted (blue area) at high pump fluence. The fluence of the probe pulse was fixed at 2.1 mW. Pump fluences are annotated to each time trace, which are vertically shifted for clarity. d, Variation of the amplitudes of the A1gA_{1g} (red dots) and A1bA_{1b} (blue dots) phonon modes as a function of the increasing fluence of the pump pulse; retrieved from the FFT analysis of the measurements shown in c. Red and blue curves display the nonlinear fits in the dual logarithmic plots, with the fit parameters displayed in

The excitation of coherent phonons on interaction with ultrashort laser pulses typically arises from two plausible mechanisms: (1) impulsive stimulated Raman scattering (ISRS) or (2) displacive excitation of coherent phonons (DECP)[29]. To further distinguish between them, we performed time-resolved XPM measurements by varying the pump fluence, while keeping the fluence of the probe pulse fixed. The temporal crosscuts at ∼680\sim680 nm from the pump-probe spectra are shown in Figure 3c. Compared to the probe fluence dependence measurement, the time-resolved traces show markedly more complex behavior (see time traces in Fig. S8). At low fluence of the pump pulse, the XPM signal gradually increases after the zero time delay (bottom curves in Figure 3c), in contrast with the abrupt reduction observed in Figure 2a and Figure 2b.

As the pump fluence increases, this gradual rise in the XPM signal, after the zero time delay, transitions into a pronounced depletion (top curves in Figure 3c). This amplification or attenuation of the XPM signal directly relates to an enhanced or reduced optical Kerr nonlinearity (n2n_2) on varying the pump fluence[38]. While this behavior complicates the extraction of the decay constants of the electronic response, phonon amplitudes can still be robustly quantified via the FFT analysis of the time-domain traces. The extracted amplitudes of both the A1gA_{1g} and A1bA_{1b} modes exhibit virtually a sub-linear dependence on the pump fluence (Figure 3d), which further substantiates the DECP mechanism[32].

Within the DECP framework, the uniform photoexcitation of electrons creates a symmetric driving force:

F(t)=−∂U(t)∂Q,F(t)=-\frac{\partial U(t)}{\partial Q},

leading to the selective excitation of fully symmetric vibrational modes, such as the A1gA_{1g} mode in MoTe2_2, whose atomic displacements match the symmetry of the perturbation. In contrast, low-symmetry modes, such as doubly degenerate EE-type modes, which involve antisymmetric or degenerate atomic motions, are not efficiently excited, as their displacements do not align with the symmetry of the driving force[39]. This symmetry-selective excitation mechanism explains the exclusive observation of fully symmetric AA modes in both 2H-MoTe2_2 and quartz (a similar trend is also observed in 1T-MoTe2_2, see SI for details).

Amplifying and attenuating Kerr nonlinear response

We now focus on the amplification and attenuation behavior of the Kerr nonlinear response induced by the nonlinear interaction of the probe pulse ensuing excitation by the pump pulse, as highlighted by the red and blue areas in Figure 3c. The interband contribution to the imaginary part of the third-order nonlinear susceptibility, χinter(3)\chi_{\mathrm{inter}}^{(3)}, can be expressed in terms of the electronic distribution in different bands as[40]:

Im⁡χinter(3)∝∑i,j∫d3k4π3∣Pij(k)∣4×fi(k)(1−fj(k))×δ(Ej(k)−Ei(k)−ℏω)(4)\operatorname{Im}\chi_{\mathrm{inter}}^{(3)} \propto\sum_{i,j}\int\frac{\mathrm{d}^{3}k}{4\pi^{3}} \lvert P_{ij}(\mathbf{k})\rvert^{4} \times f_i(\mathbf{k})\left(1-f_j(\mathbf{k})\right) \times\delta\left(E_j(\mathbf{k})-E_i(\mathbf{k})-\hbar\omega\right) \tag*{(4)}

where PijP_{ij} denotes the matrix element of the momentum operator, and fif_i and fjf_j are the occupation probabilities of the initial (ii) state and the final (jj) states. Thus, photoexcitation directly modifies the nonlinear refractive index n2n_2 via changes in the electronic population. In the present case, we focus on the K1K_1 and K2K_2 points in the band structure, where the pump pulse (∼1.3−1.55\sim1.3 - 1.55 eV) resonantly excites electrons from the valence band to the conduction band, as indicated by the vertical red arrows in Figure 3e.

In this case, the pump-induced modification in Kerr nonlinearity mainly arises from electronic populations at the K1K_1 and K2K_2 points. At the K1K_1 point, the valence band states deplete under resonant pump excitation (blue area on the left side of Figure 3e), nonetheless, the remaining electrons at the K1K_1 point can still participate in XPM process via resonant interband transitions (solid green arrow on the left side of Figure 3e). However, this transition probability would be reduced on increasing the pump fluence, because a fraction of valence-band electrons would have already been promoted to the conduction band. This results into the reduction in the χ(3)\chi^{(3)} contribution and leads to a suppression of the nonlinear signal, consistent with the reduced XPM response observed in Figure 2a and Figure 2b.

Nevertheless, photoexcited electrons in the conduction band, unoccupied in equilibrium, can amplify the nonlinear response. This amplification occurs predominantly at the K2K_2 point, where pump-excited carriers (red area on the right side of Figure 3e) enable resonant probe-induced χ(3)\chi^{(3)} processes involving higher conduction band states (solid green arrow on the right side of Figure 3e). This contribution effectively increases the Kerr nonlinearity and the associated XPM signal.

The interplay between the above two processes gives rise to the non-monotonic fluence dependence recorded in Figure 3c. Upon weak pump pulse excitation, photoexcited carriers at the K2K_2 point (nK2n_{K2}) enhance the nonlinear response experienced by the probe pulse, whereas the carrier population in the conduction band near K1K_1, remains largely unchanged. On increasing the pump fluence, resonant depletion of valence band states near K1K_1 point (nK1n_{K1}) dominates, leading to a net reduction of the XPM signal. This imbalance in the carrier population, Δn=nK2−nK1\Delta n = n_{K2} - n_{K1}, as captured by the TDDFT calculations, which increases at low pump fluence (red area) and decreases at higher fluence (blue area), is shown by the black squares in Figure 3f.

In contrast to the complex nonlinear electronic response, we find that the phonon amplitude increases virtually linearly with pump fluence, as shown by the red dots in Figure 3f, in agreement with the experimentally extracted phonon amplitudes shown in Figure 3d. These results suggest that electronic and lattice dynamics can be independently and sensitively controlled using even weak optical gating pulse[30, 32, 39]. Such low-power control of optical nonlinearity is highly desirable for ultrafast modulation, signal processing and photonic switching applications.

Coherent control of phonon oscillations via a dual pump scheme

Having established the excitation and detection mechanisms of the coherent phonons, we now focus on their controlled manipulation, which is crucial in shaping the physical properties of materials on the terahertz timescales. Controlled phonon dynamics is particularly relevant in phenomena such as charge density wave formation and electron pairing in superconductors. To achieve this, we employed two identical pump pulses, hereafter referred to as ‘pump-1’ and ‘pump-2’, each capable of individually exciting coherent phonons. By precisely tuning the delay between these two pump pulses, phonon oscillations can be selectively enhanced or suppressed through constructive or destructive interference of the launched phonon wavepackets, as illustrated in Figure 4a. The phase-sensitive modulation of the probe pulse enables real-time monitoring of this interference of the phonon wavepackets.

All-optical coherent control of Kerr nonlinearity via a dual-pump pulse scheme

Figure 4. All-optical coherent control of Kerr nonlinearity via a dual-pump pulse scheme. a, Schematic illustration of coherent control phonon dynamics using two pump pulses (‘pump-1’ and ‘pump-2’), which launch coherent phonon wavepackets Q1(t)Q_1(t) (red) and Q2(t)Q_2(t) (magenta), respectively. The relative delay TT between two pump pulses controls the phase (Φ(T)\Phi(T)) and amplitude (Q(T)Q(T)) of the launched phonons and, in turn, modulates the phonon-driven Kerr nonlinearity of 2H2H-MoTe2_2 probed at a later delay τ\tau by the probe pulse. When T=nTphT=nT_{\mathrm{ph}}, the phonon response is enhanced; when T=(n+12)TphT=(n+\frac{1}{2})T_{\mathrm{ph}}, it is suppressed, where TphT_{\mathrm{ph}} is the phonon oscillation period. b, A series of time-resolved traces measured as a function of the delay between the pump-1 (and pump-2) and probe pulses for various delays between the pump-1 and pump-2 pulses (TT), as annotated along the y-axis. The temporal crosscuts were extracted at ~680 nm from the dual-pump probe measurements. The pump-1 and pump-2 pulse powers were set at 3.4 mW (43 pJ), and the probe pulse power was set at 2.0 mW (25 pJ). The acquisition time was 0.5 s per delay step for all the spectra. c, FFT spectra of the time-resolved traces in b. d, Comparison of the time-resolved traces measured at two fixed delays between pump-1 and pump-2 pulses (TT at -0.42 ps and -0.49 ps) as indicated by the red and purple arrows in b, demonstrating the intensity control of the A1gA_{1g} phonon mode by varying the delay between the two pump pulses. Inset: FFT spectra of the two time-resolved traces show phonon oscillations being completely activated (T∼2TphT\sim2T_{\mathrm{ph}}) or suppressed (T∼2.5TphT\sim2.5T_{\mathrm{ph}}). e, Comparison of the time-resolved traces measured at two fixed delays between the pump-1 and pump-2 (TT

Figure 4b demonstrates phonon control by varying the delay, TT, between the pump-1 and pump-2 pulses. In each time trace, the delay between these two pump pulses were fixed as labelled along the vertical y-axis, whereas the delay between the two pump pulses and the probe pulse was varied along the horizontal x-axis. The individual time traces in Figure 4b represent temporal crosscuts at ∼680\sim680 nm from the pump-probe measurements. The amplitudes of the phonon oscillations exhibit a clear periodic modulation as a function of TT. The FFT spectra (Figure 4c) of these time traces reveal that the amplitude of the A1gA_{1g} mode oscillates with a period of Tph∼194T_{\mathrm{ph}} \sim194 fs, corresponding to the A1gA_{1g} phonon mode frequency (∼172\sim172 cm−1^{-1}). At specific pump-pump delays, the oscillations are largely enhanced (T∼2TphT \sim2T_{\mathrm{ph}}, red arrows in Figure 4b), whereas at other delays, they are almost completely suppressed (T∼2.5TphT \sim2.5T_{\mathrm{ph}}, purple arrows in Figure 4b). The representative time traces and their FFT spectra (inset) are shown in Figure 4d.

In addition to the amplitude control, the phase of the phonon oscillations can also be tuned by adjusting the delay between the pump-1 and pump-2 pulses, allowing the oscillations to be driven completely in- or out-of-phase, as shown in Figure 4e. This behavior is extracted from the time traces indicated by the orange and yellow arrows in Figure 4b. The interference of the phonon wavepackets leading to variation in the amplitude (red dots) and phase (blue dots) of the A1gA_{1g} phonon mode can be modeled using two driven damped harmonic oscillators[41]:

Qν(t,T)=Qν,0cos⁡(ωνt)e−tτp+Qν,0cos⁡(ων(t−T))e−(t−T)τp(5)Q_\nu(t,T)=Q_{\nu,0}\cos(\omega_\nu t)e^{-\frac{t}{\tau_p}}+Q_{\nu,0}\cos(\omega_\nu(t-T))e^{-\frac{(t-T)}{\tau_p}} \tag*{(5)}

where Qν,0Q_{\nu,0} is the initial amplitude set by DECP excitation, ων\omega_\nu is the phonon oscillation frequency, and τp\tau_p is the phonon decay time constant. The resulting phase of the phonon oscillation is given by:

Φν(T)=−arctan⁡(sin⁡(ωνT)eTτpcos⁡(ωνT)eTτp+1)(6)\Phi_\nu(T)=-\arctan\left(\frac{\sin(\omega_\nu T)e^{\frac{T}{\tau_p}}}{\cos(\omega_\nu T)e^{\frac{T}{\tau_p}}+1}\right) \tag*{(6)}

Using this model, we accurately reproduce the experimentally observed amplitude and phase modulation, as shown by the solid curves in Figure 4f. The phonon control as experimentally measured is also independently confirmed by TDDFT calculations (see SI for details).

This phonon control offers a novel approach for manipulating dynamics in materials. By selectively turning phonon oscillations on or off, the electronic response can also be modulated, as observed by the gain in the XPM signal—indicating an enhanced Kerr nonlinearity—observed at shorter delays between the pump-1 and pump-2 pulses in Figure 4b. The relatively slow modulation of the Kerr nonlinear response leads to the emergence of a very low-frequency mode in the FFT spectra in Figure 4c. Such a precise control over phonons holds promise for manipulating nonequilibrium phases in MoTe2_2, such as phonon-mediated superconductivity and phase transitions[22, 42].

Concluding Remarks

In this work, we demonstrated a phase-sensitive nonlinear optical spectroscopic technique capable of providing direct insights into the ultrafast electronic and phononic dynamics. We showed that the optically excited coherent phonons modulate the Kerr nonlinearity of materials, inducing cross-phase modulation on a co-propagating probe pulse. This approach operates with minimal pulse power, is applicable to various materials, and is highly sensitive to transient phase transitions and excitonic dynamics in a nondestructive manner. Coherent control of phonon oscillations was achieved using a dual-pump pulse scheme, enabling precise real-time manipulation of phonon dynamics. This capability lays the foundation for further exploration of phonon-mediated control in quantum materials, such as charge density waves and superconductivity. Our findings open new avenues for phonon-engineered light-matter interactions on femtosecond timescales, and highlight the potential of ultrafast nonlinear optical modulations in advancing next-generation quantum and optoelectronic devices.

Methods

Experimental setup

Ultrafast pump (λ∼800\lambda\sim800–950950 nm, ∼10\sim10 fs long) and probe pulses (λ∼700\lambda\sim700–750750 nm, ∼20\sim20 fs long) were employed for the phase-sensitive nonlinear spectroscopy measurements. The pump and probe pulses were spectrally separated from the ultra-broadband output of a Ti:Sapphire oscillator (Element 2) using an 800 nm shortpass dichroic mirror (DMSP) and were pre-compressed with a pair of chirped mirrors.

The collinearly propagating pump and probe pulses were focused onto the sample (mounted onto a motorized 3D stage) using a 40 ×\times reflective objective with a numerical aperture of 0.5. The light emanating from the sample was then collected using the same objective and passed through a 690 nm shortpass filter (SP) before being directed into a spectrometer for analysis. A precision motorized linear stage (PI Instruments) was used to adjust the delay between the pump and probe pulses. For all spectra, an acquisition time of 0.5 s per delay step was used unless otherwise specified.

For pump-probe experiments, the power and polarization of both pump and probe pulses could be independently controlled using two motorized neutral density (ND) filters in their respective beam paths. Raman spectroscopy was performed on a commercial Raman spectrometer from WiTec, and a 633 nm CW laser was used for the measurements. The laser power was kept below 1 mW to avoid sample damage. All the experiments were conducted at room temperature.

Samples preparation

The few-layer (three to five) thick 2H2H-MoTe2_2 samples, grown by chemical vapor deposition (CVD) on a quartz substrate, were purchased from 2D Semiconductors. The 2H2H-MoTe2_2 and 1T1T-MoTe2_2 flakes (in Fig. S3), were mechanically exfoliated from their bulk crystals (bought from HQ Graphene) and transferred onto the Si3_3N4_4 substrate within a glove box. The monolayer 2H2H-MoTe2_2 were mechanically exfoliated and transferred onto a sapphire substrate, and the thicknesses were characterized using Raman spectroscopy (as shown in Fig. S4).

Pump-probe spectra simulation

In the pump-probe experiments, nonlinear cross-phase modulation (XPM) and four-wave mixing (FWM) processes occur as the pump and probe pulses interact in the nonlinear medium. For the XPM spectra simulation in Figure 2d, the Kerr nonlinearity response was separated into phonon and electronic contributions[32]. The phase response modulated by phonon oscillations was modeled as a damped oscillator:

Φp,ν(t,τ)=a⋅cos⁡(ων(t+τ)+φν)e−t+ττp\Phi_{p,\nu}(t,\tau) = a \cdot\cos(\omega_\nu(t+\tau)+\varphi_\nu)e^{-\frac{t+\tau}{\tau_p}}

where aa is a perfector, ων\omega_\nu is the phonon oscillation frequency, φν\varphi_\nu is the phase of the phonon oscillation, and τp\tau_p is the phonon decay time constant determined from the phonon decay analysis in Fig. S2.

The phase response contributed by electronic components was modeled as a simple step-decay exponential function upon the pump pulse excitation:

Φe(t,τ)=b⋅u(τ)e−t+ττe\Phi_e(t,\tau)=b\cdot u(\tau)e^{-\frac{t+\tau}{\tau_e}}

where bb is a prefactor, u(τ)u(\tau) (u(τ≤0)=0,u(τ>0)=1u(\tau\le0)=0,u(\tau>0)=1) is a step function, and τe\tau_e is the electronic response decay time constant determined from the exponential fit in Figure 2b. In the simulation, aa and bb are the only free fit parameters, which are related to the pump laser power.

FWM involves the interaction of the fields with nonlinear polarization of the medium, which results in the generation of new optical fields. In the simulations, the FWM processes that could generate frequencies larger than 700 nm were considered, as shown in Fig. S6. The nonlinear polarization in the medium responsible for the FWM processes could be expressed in terms of the third-order susceptibility χ(3)\chi^{(3)}, which is given by:

P=∑j,k,lϵ0χijkl(3)EjEkElP=\sum_{j,k,l}\epsilon_0\chi_{ijkl}^{(3)}E_jE_kE_l

where ii, jj, and kk represent the pump and probe pulse indices, respectively. The generated field at ωi=±ωj±ωk±ωl\omega_i=\pm\omega_j\pm\omega_k\pm\omega_l is determined by the pump and probe field frequencies.

By adjusting the free parameters aa and bb in the XPM simulation, both the FWM and phonon oscillations in Figure 2d could be reproduced by the model described above.

Time-dependent density-functional theory (TDDFT)

The excited state dynamics can be simulated via a real-time density functional theory framework (rt-TDDFT), which we implemented in the QUANTUM ESPRESSO package[43, 44, 45].

Under the theorem of TDDFT, the time evolution of electron wave functions is governed by the time dependent Kohn-Sham (TDKS) equations[46]:

i∂∂tψi,k(r,t)=[12m(p−ecA)2+V(r,t)]ψi,k(r,t),\mathrm{i}\frac{\partial}{\partial t}\psi_{i,\mathbf{k}}(\mathbf{r},t)=\left[\frac{1}{2m}\left(\mathbf{p}-\frac{e}{c}\mathbf{A}\right)^2+V(\mathbf{r},t)\right]\psi_{i,\mathbf{k}}(\mathbf{r},t),

where velocity gauge is used and the external field appears in the kinetic term in the form of vector potential A(t)\mathbf{A}(t). The propagation of TDKS orbitals is implemented on the adiabatic basis ϕn,k(r,t)\phi_{n,\mathbf{k}}(\mathbf{r},t):

∣ψi,k(r,t)⟩=∑ncink(t)∣ϕn,k(r,t)⟩,|\psi_{i,\mathbf{k}}(\mathbf{r},t)\rangle=\sum_n c_{in\mathbf{k}}(t)|\phi_{n,\mathbf{k}}(\mathbf{r},t)\rangle,

where ii and nn denote the band index, k\mathbf{k} refers to the reciprocal momentum index and cink(t)c_{in\mathbf{k}}(t) the time dependent coefficients. The adiabatic basis is calculated on the fly at each ionic step by diagonalizing the Hamiltonian:

Hk(r,t)∣ϕn,k(r,t)⟩=εn,k(t)∣ϕn,k(r,t)⟩H_{\mathbf{k}}(\mathbf{r},t)|\phi_{n,\mathbf{k}}(\mathbf{r},t)\rangle=\varepsilon_{n,\mathbf{k}}(t)|\phi_{n,\mathbf{k}}(\mathbf{r},t)\rangle

where εn,k(t)\varepsilon_{n,\mathbf{k}}(t) is the eigenvalue. The charge density can be calculated with cink(t)c_{in\mathbf{k}}(t) and ϕn,k(r,t)\phi_{n,\mathbf{k}}(\mathbf{r},t):

ρ(r,t)=∑k∑nqn,k(t)∣⟨ϕn,k(r,t)∣ϕn,k(r,t)⟩∣2,\rho(\mathbf{r},t)=\sum_{\mathbf{k}}\sum_n q_{n,\mathbf{k}}(t)\left|\langle\phi_{n,\mathbf{k}}(\mathbf{r},t)|\phi_{n,\mathbf{k}}(\mathbf{r},t)\rangle\right|^2,

where

qn,k(t)=∑i∣cink(t)∣2q_{n,k}(t) = \sum_i \lvert c_{ink}(t) \rvert^2

is the population on the adiabatic basis.

For ions that are much heavier than electrons, their motions are treated classically on an averaged potential energy surface determined by the electronic distribution according to the Ehrenfest theorem. The nuclear positions are updated following the Newton's second law[47]:

Mαd2Rαdt2=∑i⟨ψi∣∇Rα(12m(p−ecA)2+V(r,t))∣ψi⟩M_\alpha\frac{d^2\mathbf{R}_\alpha}{dt^2} = \sum_i \langle\psi_i \lvert\nabla_{\mathbf{R}_\alpha} \left( \frac{1}{2m} \left( \mathbf{p} - \frac{e}{c}\mathbf{A} \right)^2 + V(\mathbf{r},t) \right) \rvert\psi_i \rangle

where MαM_\alpha and Rα\mathbf{R}_\alpha are the mass and position of the α\alphath ion.

Data availability

The data that supports the findings of this study are available from the corresponding authors on request.

Code availability

The details needed to reproduce the computations have been provided in the “Methods” section and Supplementary Information file.

Acknowledgments

We thank Wolfgang Stiepany and Marko Memmler for technical support.

Contributions

S.S., Y.L., Sa.S., M.B., A.A. and M.G. built the experimental setup, performed the experiments and analyzed the experimental data. C.W., Y.W. and S.M. designed and performed the theoretical calculations and analyzed the theoretical data. M.G. conceived the project and designed the experiments. All authors interpreted the results and contributed to the preparation of the manuscript.

Competing interests

The authors declare no competing interests.

the respective legends. e, Density functional theory (DFT)-calculated band structure of 2H2H-MoTe2_2. Under pump pulse excitation, electrons at the K1K_1 and K2K_2 points are resonantly promoted to the conduction band (red arrows). The contribution of the probe pulse on XPM arises primarily from resonant interband transitions (solid green arrows). f, TDDFT calculated evolution of the population difference of the photoexcited electrons at the K1K_1 and K2K_2 points, Δn=nK2−nK1\Delta n = n_{K_2} - n_{K_1} (black square, left y-axis), as a function of increasing fluence of the pump pulses. The evaluated A1g_{1g} phonon amplitudes (red dots, right y-axis) as a function of the increasing fluence of the pump pulses. The double-logarithmic fit (solid red-curve) of the evaluated phonon amplitudes yields a slope of ~ 0.74, indicating an approximately linear dependence on the pump fluence.

at -0.31 ps and -0.47 ps) pulses as indicated by the orange and yellow arrows in b, demonstrating phase control of the A1gA_{1g} phonon mode by adjusting the delay between the two pump pulses. Inset: A close-up of the phonon oscillations, highlighting the out-of-phase oscillation behavior. f, The amplitude (Q(T)Q(T), upper panel, red dots) and phase (Φ(T)\Phi(T), lower panel, blue dots) of the A1gA_{1g} phonon mode in 2H2H-MoTe2_2 as retrieved from the measurement shown in b as a function of TT. The solid red and blue curves represent simulations based on two damped harmonic oscillators driven by a pair of pump pulses.

References

  1. [1]Mankowsky, R., Först, M. & Cavalleri, A. Non-equilibrium control of complex solids by nonlinear phononics. Rep. Prog. Phys. 79, 064503 (2016).DOI
  2. [2]Disa, A. S., Nova, T. F. & Cavalleri, A. Engineering crystal structures with light. Nat. Phys. 17, 1087–1092 (2021).DOI
  3. [3]Fausti, D. et al. Light-Induced Superconductivity in a Stripe-Ordered Cuprate. Science 331, 189–191 (2011).DOI
  4. [4]Mitrano, M. et al. Possible light-induced superconductivity in K₃C₆₀ at high temperature. Nature 530, 461–464 (2016).DOI
  5. [5]Gerber, S. et al. Direct characterization of photoinduced lattice dynamics in BaFe₂As₂. Nature Comm. 6, 7377 (2015).DOI
  6. [6]Liu, H., Sun, J. T., Cheng, C., Liu, F. & Meng, S. Photoinduced Nonequilibrium Topological States in Strained Black Phosphorus. Phys. Rev. Lett. 120, 237403 (2018).DOI
  7. [7]McIver, J. W., Hsieh, D., Steinberg, H., Jarillo-Herrero, P. & Gedik, N. Control over topological insulator photocurrents with light polarization. Nat. Nanotechnol. 7, 96–100 (2012).DOI
  8. [8]Price, M. B. et al. Hot-carrier cooling and photoinduced refractive index changes in organic-inorganic lead halide perovskites. Nat. Commun. 6, 1–8 (2015).DOI
  9. [9]Johnson, A. S. et al. All-optical seeding of a light-induced phase transition with correlated disorder. Nat. Phys. 20, 970–975 (2024).DOI
  10. [10]Venanzi, T. et al. Ultrafast switching of trions in 2D materials by terahertz photons. Nat. Photonics 18, 1344–1349 (2024).DOI
  11. [11]Ginsberg, J. S. et al. Phonon-enhanced nonlinearities in hexagonal boron nitride. Nat. Commun. 14, 7685 (2023).DOI
  12. [12]Soffer, Y., Raanan, D. & Oron, D. Low Frequency Collinear Pre-Resonant Impulsive Stimulated Raman Microspectroscopy. ACS Photonics 7, 3481–3488 (2020).DOI
  13. [13]Wu, Y. et al. Emergence of electron coherence and two-color all-optical switching in MoS₂ based on spatial self-phase modulation. Proc. Natl. Acad. Sci. U. S. A. 112, 11800–11805 (2015).DOI
  14. [14]Dhar, L., Rogers, J. A. & Nelson, K. A. Time-Resolved Vibrational Spectroscopy in the Impulsive Limit. Chem. Rev. 94, 157–193 (1994).DOI
  15. [15]Shivkumar, S. et al. Selective Detection in Impulsive Low-Frequency Raman Imaging Using Shaped Probe Pulses. Phys. Rev. Appl. 19, 054075 (2023).DOI
  16. [16]Nicholls, L. H. et al. Ultrafast synthesis and switching of light polarization in nonlinear anisotropic metamaterials. Nat. Photonics 11, 628–633 (2017).DOI
  17. [17]Tyulnev, I. et al. Valleytronics in bulk MoS₂ with a topologic optical field. Nature 628, 746–751 (2024).DOI
  18. [18]Bauer, J. M. et al. Excitonic resonances control the temporal dynamics of nonlinear optical wave mixing in monolayer semiconductors. Nat. Photonics 16, 777–783 (2022).
  19. [19]Wen, X., Gong, Z. & Li, D. Nonlinear optics of two-dimensional transition metal dichalcogenides. InfoMat 1, 317–337 (2019).DOI
  20. [20]Cho, S. et al. Phase patterning for ohmic homojunction contact in MoTe₂. Science (1979). 349, 625–628 (2015).DOI
  21. [21]Zhang, M. Y. et al. Light-Induced Subpicosecond Lattice Symmetry Switch in MoTe₂. Phys. Rev. X 9, 021036 (2019).DOI
  22. [22]Kim, S., Lei, S., Schoop, L. M., Cava, R. J. & Ong, N. P. Edge supercurrent reveals competition between condensates in a Weyl superconductor. Nat. Phys. 20, 261–268 (2024).DOI
  23. [23]Shi, J. et al. Intrinsic 1T′ phase induced in atomically thin 2H-MoTe₂ by a single terahertz pulse. Nat. Commun. 14, 5905 (2023).DOI
  24. [24]Hu, L., Sun, F., Zhao, H. & Zhao, J. Nonlinear optical response spatial self-phase modulation in MoTe₂: correlations between χ(3) and mobility or effective mass. Opt. Lett. 44, 5214 (2019).DOI
  25. [25]Thomas Schneider. Nonlinear Optics In Telecommunications. Springer (2004).DOI
  26. [26]Alfano, R. R., Baldeck, P. L. & Ho, P. P. Cross-phase Modulation and Induced-focusing of Optical Nonlinearities in Optical Fibers and Bulk Materials. J. Opt. Soc. Am. B 6, 824–829 (1989).DOI
  27. [27]Cruz, C., Stephenson, J. & Wahlstrand, J. Phase-sensitive pump-probe measurement of the complex nonlinear susceptibility of silicon across the direct band edge. Optica 11, 2334-2536 (2024).
  28. [28]Fukuda, T. et al. Coherent optical response driven by non-equilibrium electron-phonon dynamics in a layered transition-metal dichalcogenide. APL Mater. 12, 021102 (2024).DOI
  29. [29]Sayers, C. J. et al. Strong Coupling of Coherent Phonons to Excitons in Semiconducting Monolayer MoTe₂. Nano Lett. 23, 9235–9242 (2023).DOI
  30. [30]Zhang, J. et al. High-harmonic spectroscopy probes lattice dynamics. Nat. Photonics 18, 792–798 (2024).arxiv.org/abs/2403.14906
  31. [31]Wang, C., Chen, D., Wang, Y. & Meng, S. Directional Pumping of Coherent Phonons and Quasiparticle Renormalization in a Dirac Nodal-Line Semimetal. Phys. Rev. X 15, 21053 (2025).DOI
  32. [32]Zeiger, H. J. et al. Theory for displacive excitation of coherent phonons. Phys. Rev. B 45, 768–778 (1992).pubmed.ncbi.nlm.nih.gov/10001117
  33. [33]Vialla, F. & Fatti, N. Del. Time-domain investigations of coherent phonons in van der waals thin films. Nanomaterials 10, 2543 (2020).DOI
  34. [34]Heide, C., Kobayashi, Y., Haque, S. R. U. & Ghimire, S. Ultrafast high-harmonic spectroscopy of solids. Nat. Phys. 20, 1546–1557 (2024).DOI
  35. [35]Attar, A. et al. Simultaneous observation of carrier-specific redistribution and coherent lattice dynamics in 2H-MoTe₂ with femtosecond core-level spectroscopy. ACS Nano 14, 15829-15840 (2020).DOI
  36. [36]Luo, Y. et al. Imaging and controlling coherent phonon wave packets in single graphene nanoribbons. Nat. Commun. 14, 3784 (2023).DOI
  37. [37]Virga, A. et al. Coherent anti-Stokes Raman spectroscopy of single and multi-layer graphene. Nat. Commun. 10, 3658 (2019).DOI
  38. [38]Wu, L. et al. Recent Advances of Spatial Self-Phase Modulation in 2D Materials and Passive Photonic Device Applications. Small 16, 2002252 (2020).DOI
  39. [39]Huang, Y. et al. Ultrafast Measurements of Mode-Specific Deformation Potentials of Bi₂Te₃ and Bi₂Se₃. Phys. Rev. X 13, 41050 (2023).
  40. [40]Hache, F., Ricard, D., Flytzanis, C. & Kreibig, U. The Optical Kerr Effect in Small Metal Particles: The Case of Gold Metal. Appl. Phys. A 47, 347-357 (1988).
  41. [41]Utikal, T., Stockman, M. I., Heberle, A. P., Lippitz, M. & Giessen, H. All-optical control of the ultrafast dynamics of a hybrid plasmonic system. Phys. Rev. Lett. 104, 113903 (2010).pubmed.ncbi.nlm.nih.gov/20366478
  42. [42]Wall, S. et al. Ultrafast changes in lattice symmetry probed by coherent phonons. Nat. Commun. 3, 721 (2012).DOI
  43. [43]Meng, S. & Kaxiras, E. Real-time, local basis-set implementation of time-dependent density functional theory for excited state dynamics simulations. J. Chem. Phys. 129, 054110 (2008).DOI
  44. [44]Lian, C., Zhang, S. J., Hu, S. Q., Guan, M. X. & Meng, S. Ultrafast charge ordering by self-amplified exciton–phonon dynamics in TiSe₂. Nat. Commun. 11, 43 (2020).DOI
  45. [45]Lian, C., Guan, M., Hu, S., Zhang, J. & Meng, S. Photoexcitation in Solids: First-Principles Quantum Simulations by Real-Time TDDFT. Adv. Theory Simul. 1, 1800055 (2018).DOI
  46. [46]Runge, E. & Gross, E. K. U. Density-functional theory for time-dependent systems. Phys. Rev. Lett. 52, 997–1000 (1984).DOI
  47. [47]Kapral, R. Mixed quantum-classical dynamics. J. Chem. Phys. 110, 8919–8929 (1999).DOI

Paper details

vlnc:W_fx8qdb2t
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