SCIENCE ADVANCES | RESEARCH ARTICLE MATERIALS SCIENCE Copyright © 2021 The Authors, some Long-lived charge separation following rights reserved; exclusive licensee pump-wavelength–dependent ultrafast charge American Association for the Advancement transfer in graphene/WS2 heterostructures of Science. No claim to original U.S. Government Works. Distributed Shuai Fu1, Indy du Fossé2, Xiaoyu Jia1, Jingyin Xu1,3, Xiaoqing Yu1, Heng Zhang1, under a Creative Wenhao Zheng1, Sven Krasel1, Zongping Chen4, Zhiming M. Wang3, Klaas-Jan Tielrooij5, Commons Attribution Mischa Bonn1, Arjan J. Houtepen2, Hai I. Wang1* NonCommercial License 4.0 (CC BY-NC). Van der Waals heterostructures consisting of graphene and transition metal dichalcogenides have shown great promise for optoelectronic applications. However, an in-depth understanding of the critical processes for de- vice operation, namely, interfacial charge transfer (CT) and recombination, has so far remained elusive. Here, we investigate these processes in graphene-WS2 heterostructures by complementarily probing the ultrafast tera- hertz photoconductivity in graphene and the transient absorption dynamics in WS2 following photoexcitation. We observe that separated charges in the heterostructure following CT live extremely long: beyond 1 ns, in contrast to ~1 ps charge separation reported in previous studies. This leads to efficient photogating of graphene. Further- more, for the CT process across graphene-WS2 interfaces, we find that it occurs via photo-­thermionic emission for sub-A-exciton excitations and direct hole transfer from WS2 to the valence band of graphene for above-A-exciton excitations. These findings provide insights to further optimize the perform­ance of optoelectronic devices, in par- ticular photodetection. INTRODUCTION g-based hybrid photodetectors, e.g., g-PbS QD photodetectors (9), Atomically thin layers, including graphene and monolayer transition an extremely long interfacial charge separation time (~20 ms) has metal dichalcogenides (TMDCs), represent a fascinating material class been attributed to carrier trapping in the QDs. The long-lived inter- for electronic and optoelectronic applications. As a bond-free strategy, facial charge separation establishes an electric field at vdW interfaces, stacking these two-dimensional (2D) layers allows the production leading to photoconductive gain and, thus, a high Rph in the photo- of artificial van der Waals (vdW) heterostructures, which offers the detectors (the so-called “photogating effect,” in analogy to field-effect prospect of discovering new synergetic electronic, optical, and mag- gating). As the photoconductive gain (G) is linearly proportional to netic phenomena (1–4). Thanks to the development of precise con- the charge separation lifetime (CS) (13), it remains unclear how ef- trol over composition, layer numbers, stacking angles, and sequences ficient photodetectors can be realized in g-TMDC heterostructures of the atomic layers, the past decade has witnessed the blossoming (5, 14, 15) with a reported CS of ~1 ps. of novel concepts and high-performance devices based on vdW hetero- Along with the controversy on the charge separation lifetime, the structures (5–7). One notable example is that the integration of second puzzle regarding the (interfacial) charge carrier dynamics in graphene and monolayer TMDCs enables sensitive photodetectors g-TMDC heterostructures lies in the efficiency and underlying with high photoresponsivity (Rph) up to 107 A W−1 at room tem- mechanisms of charge transfer (CT) processes. Immediately follow- perature (8), comparable to state-of-the-art photodetectors based ing photoexcitation of bare graphene, the photogenerated, nonther- on graphene–quantum dot (g-QD) hybrid system with Rph up to malized hot carriers can efficiently transfer their excess energy to 108 A W−1 (9). other charge carriers within tens of femtoseconds via carrier-carrier There has been a great and successful effort to increase the effi- scattering (16–19). This thermalization process leads to the forma- ciency of graphene-TMDC (g-TMDC) photodetectors, but the un- tion of thermalized hot carriers with a well-defined electron tem- derstanding of the fundamental photophysics of these devices has perature (Te) following the Fermi-Dirac distribution (20, 21). The remained elusive. For instance, recent spectroscopic studies (10–12) thermalized hot carriers undergo a cooling process within a few have reported a very fast (~1 ps) charge recombination (via back picoseconds via electron-phonon scattering (22–26). In g-TMDC electron transfer process) at g-WS2 interfaces. Such a short charge vdW heterostructures, TMDCs can serve as transport channels to separation lifetime seems in contradiction to the large Rph reported harvest hot carriers (in principle, for both nonthermalized and in the photodetectors based on g-TMDCs heterostructures. For other thermalized hot carriers) from graphene before the cooling process takes place. Pioneering device works (27, 28) have shown that hot electron transfer (HET) contributes to the photocurrent generation 1 Max Planck Institute for Polymer Research, Ackermannweg 10, D-55128 Mainz, at g-based vdW interfaces, which has been further confirmed by ul- Germany. 2Optoelectronic Materials Section, Faculty of Applied Sciences, Delft Uni- versity of Technology, Van der Maasweg 9, 2629 HZ Delft, Netherlands. 3Institute of trafast spectroscopic studies (10–12). However, the mechanism of Fundamental and Frontier Sciences, University of Electronic Science and Technol- HET across vdW interfaces, and in particular whether HET occurs ogy of China, Chengdu 610054, P. R. China. 4School of Materials Science and Engi- before or after thermalization, remains highly debated. For instance, neering, Zhejiang University, Zheda Road 38, Hangzhou 310027, China. 5Catalan in the device work by Massicotte et al. (27), the photocurrent gener- Institute of Nanoscience and Nanotechnology (ICN2), BIST and CSIC, Campus UAB, Bellaterra, 08193 Barcelona, Spain. ation in a g-WSe2–g vdW heterostructure following below-­WSe2- *Corresponding author. Email: wanghai@mpip-mainz.mpg.de bandgap excitation is attributed to photo-thermionic emission, in Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 1 of 10 SCIENCE ADVANCES | RESEARCH ARTICLE which thermalized hot electrons with energy above the interfacial (Fig. 1B) following photoexcitation with a wide range of photon energy barrier can be injected into WSe2 (with a quantum yield of ~1%). energies from 0.7 to 3.1 eV. The unique combination of THz and In sharp contrast, Chen et al. (10) recently proposed that HET at TA spectroscopies enables us to track the charge carrier dynamics g-WS2 interfaces competes with the thermalization in graphene and in both the donor and acceptor independently and to identify and shows an extremely high quantum yield (~ 50%). Furthermore, quantify the trapping process (if active) at the interfaces. We find Yuan et al. (11) suggested an alternative model for HET at g-WS2 that the charge separation lifetime following ultrafast CT is remark- interfaces, in which a direct excitation from graphene to WS2 can ably long-lived, beyond 1 ns (limited by the scan range of the setup). take place via CT states due to strong interfacial electronic coupling. This observation is in sharp contrast to the short-lived excited-state It is apparent that further studies on HET across g-TMDCs inter- dynamics (~1 ps) in WS2 observed by TA, both in this study and in faces are required to solve this debate. Furthermore, while the contri- previous works (10, 11). We rationalize the discrepancy between the bution of hot electrons to CT has been recognized, it remains unclear carrier lifetime in the electron donor (graphene) and the acceptor if the “cold” electrons, e.g., the valance band electrons in graphene, (WS2) by the presence of trap states at g-WS2 interfaces. These trap are involved in interfacial CT at g-TMDC interfaces. states can effectively capture the electrons from the excited states of Here, aiming to provide a comprehensive understanding of the WS2 within 1 ps (corresponding to the fast decay in TA dynamics) interfacial carrier dynamics in g-WS2 vdW heterostructures, we mea- and further store them for over 1 ns (corresponding to the long- sure complementarily the ultrafast photoconductivity dynamics in lived photoconductivity in graphene by THz study) before recombin- graphene by terahertz (THz) spectroscopy (Fig. 1A) and the excited-­ ing with the holes in graphene. This leads to a long-lived photogating state dynamics in TMDCs by transient absorption (TA) spectroscopy effect in graphene. On the other hand, for the forward CT process from graphene to WS2, we unveil two distinctively different CT re- gimes when exciting the heterostructure below (0.7 to 2 eV) or above the A-exciton resonance (2 to 3.1 eV) of monolayer WS2. Exciting below the WS2 A-exciton transition, we find that a relatively inefficient (~1%) HET via photo-thermionic emission governs the CT process, in which thermalized hot electrons in graphene are emitted over the energy barrier and transferred into WS2 (Fig. 1C). In the second regime where the photoexcitation is above the A-exciton resonance of WS2, we report a more efficient (up to ~5%) direct hole transfer (DHT) mechanism, which involves the photogenerated holes in the valence band of WS2 and electrons in the valence band of graphene (Fig. 1D). RESULTS The g-WS2 heterostructure used in this study was obtained com- mercially (from SixCarbon Technology, Shenzhen). It is produced by transferring a chemical vapor–deposited (CVD) graphene mono- layer onto a CVD WS2 monolayer, grown on a sapphire substrate (schematically depicted in fig. S1A). As both layers are multicrystal- line with a typical domain size of several micrometers, the dynamics that we are probing is an average response of the heterostructure with mixed twisting angles (given that the probe beam has a diameter of ~0.5 to 1 mm for THz measurements and ~0.1 mm for TA mea- surements). In an independent electrical four-point probe measure- ment of the resistivity of graphene (produced by the same method) at varied gating potentials, we find that the graphene in the hetero- structure is initially p-doped, in line with previous reports (10) for the same system. Before studying the ultrafast carrier dynamics, we have characterized the static optical and electronic properties of the Fig. 1. Investigation of the nonequilibrium hot carrier dynamics at g-WS2 heterostructure using ultraviolet-visible (UV-vis) absorption and interfaces. (A) Schematic of using ultrafast THz spectroscopy to measure the photo- Raman spectroscopy. We observe two exciton resonances at 2.0 and conductivity of graphene following photoexcitation. (B) Scheme of ultrafast TA 2.4 eV in the absorption spectra (fig. S1B), corresponding to the A- spectroscopy to measure the excited-state dynamics of WS2 following photoexci- and B-exciton transitions from the spin-split valence bands at the K tation. (C) Illustration of interfacial band alignment and photo-thermionic emission point of the Brillouin zone in monolayer WS2 (29–31), respectively for thermalized hot electrons following sub-A-exciton excitation, in which thermalized (see also the simplified band structure of WS2 in Fig. 1, C and D). hot electrons in graphene can be emitted over the energy barrier and transferred into the excited states of WS2. A- and B-exciton transitions in WS2, originating The featureless constant absorption (~2.3%) in the near-infrared from the spin-split valence bands at the K point of the Brillouin zone, are marked. range originates from the absorption of monolayer graphene (32, 33). (D) Schematic of direct hole transfer (DHT) at g-WS2 interfaces following above-A-­ Raman studies shown in fig. S1 (C and D) further confirm that both exciton excitation, in which the photogenerated holes in the valance band of WS2 the graphene and WS2 layer are monolayers in nature. On the basis recombine with valence band electrons in graphene. of the G-band position in graphene, we can estimate the Fermi level Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 2 of 10 SCIENCE ADVANCES | RESEARCH ARTICLE (EF) in graphene to be ~0.11 eV (equivalent to a free carrier density appears only in the heterostructure. We assign this positive photo- N of 7.9 × 1011 cm−2; see section S1) below the Dirac point (given conductivity to an interfacial charge-transfer process across g-WS2 the p-doped nature of graphene from electrical measurements). We interfaces (which is supported by TA measurements in Fig. 2D have further verified EF in graphene using THz-time domain spec- below). Furthermore, we find that the lifetime of the long-lived troscopy (THz-TDS; see section S2). In short, in the THz-TDS mea- photoconductivity does not depend on pump photon energy (see surements, we record the THz electric field transmitted through the fig. S2B). We attribute this positive, long-lived photoconductivity WS2/sapphire substrate configurations in the time domain, with and its independence on excitation photon energy to photogating and without graphene on the top, i.e., E(t) (for graphene/WS2/sapphire) effect in the vdW heterostructure: Following photoexcitation, elec- and E0(t) (for WS2/sapphire), respectively. The obtained time-dependent trons are first transferred from graphene to WS2 on a sub-picosecond THz fields are further converted into the frequency domain by Fourier time scale and subsequently get trapped at interfacial states where transform as E() and E0(). The THz absorption, because of the they remain for over 1 ns. For the initially p-doped graphene layer, presence of free charges in graphene, can be well described by the extracting electrons from graphene results in a shift of Fermi level Drude model, which provides microscopic transport properties in further away from the Dirac point, and thus an increase of conduc- graphene, including the charge carrier density (or Fermi level). On tivity in graphene. To confirm such a Fermi-level downshift follow- the basis of THz-TDS results, we estimate the EF in graphene to ing excitation and electron transfer from graphene to WS2, we analyze be 0.1 eV versus the Dirac point [see fig. S1 (E and F) and the asso- the frequency dependence of the complex THz photoconductivity ciated discussions in section S2], in excellent agreement with our by THz-TDS (see details in section S2). The pump-probe delay () is Raman measurements. chosen at ~100 ps to avoid the intrinsic hot carrier state in graphene, which has a lifetime of ~10 ps. As shown in Fig. 2B, we find the Ultrafast interfacial CT and long-lived charge separation frequency-resolved THz photoconductivity can be well fitted with in g-WS2 vdW heterostructures probed by THz spectroscopy the Drude model, indicating that free carriers dominate the THz We investigate the dynamics of photogenerated charge carriers in photoconductivity, with a scattering time of 69 fs (see details in sec- the g-WS2 system by optical pump-THz probe (OPTP) spectroscopy. tion S2). On the basis of the fit, EF of graphene at  = 100 ps follow- In a typical OPTP measurement, as shown in Fig. 1A, an optical ing 1.55-eV excitation is 0.14 eV below the Dirac point (equivalent pump with ~50-fs duration selectively excites either only the graphene to N = 1.1 × 1012 cm−2), indicating a 30- to 40-meV downshift of EF layer (sub-A-exciton transition in WS2, hv < 2 eV) or both layers in the in graphene following electron transfer from graphene to WS2 (given heterostructure at a fixed sample spot (hv > 2 eV). The pump-induced the initial EF of 0.1 to 0.11 eV). photoconductivity (∆) is probed by a THz pulse by monitoring the change in the transmitted electrical field (∆E = Epump − E0) following CT dynamics and interfacial charge separation time probed the optical pump as a function of pump-probe delay. The measure- by TA spectroscopy ment is based on the principle that −∆E is proportional to ∆ (34). To provide a direct spectroscopic signature of electron injection As control measurements, we first investigate the carrier dynam- from graphene to WS2, we have performed complementary mea- ics of the individual monolayer WS2 and graphene upon 1.55-eV surements using TA spectroscopy. Along with providing direct evi- excitation. As the excitation energy is still below the A-exciton res- dence for CT across the heterostructures, using TA spectroscopy is onance of monolayer WS2, we observe no photoconductivity (Fig. 2A, further motivated by noticing a large discrepancy between the charge blue line) for monolayer WS2, as expected. On the other hand, the separation lifetime obtained by our THz results (beyond 1 ns) and same excitation energy for graphene yields a transient reduction in previous TA results (reported to be ~1 ps). By probing the corre- the conductivity, i.e., negative photoconductivity (Fig. 2A, gray cir- sponding excited-state absorption changes with a pump-probe cle points). The negative photoconductivity in doped graphene has scheme, the charge occupation dynamics in WS2 can be directly ob- been widely reported previously (19, 35–40) and can be briefly un- tained with a subpicosecond time resolution. As shown in Fig. 2C, derstood as follows: Following optical excitation and the rapid ther- we observe no photobleaching (PB) signal when exciting the indi- malization in the doped graphene, the increased carrier temperature vidual monolayer WS2 with 1.55-eV pulses (below its A-exciton leads to a reduced screening of the long-range Coulomb scattering resonance), in line with our THz results (Fig. 2A, blue line). This and, thus, reduced conductivity (38). Within a few picoseconds, result also indicates that the pump fluence used in this study is suf- these thermalized hot carriers cool down to the initial thermal equi- ficiently low to avoid notable two-photon absorption in WS2. A librium via electron-phonon scattering. small, short-lived differential-like dynamic (with a lifetime of ~150 fs For g-WS2 heterostructures, one could, in the first instance, ex- as shown in Fig. 2E) is noticed and may be attributed to the co- pect the photoconductivity upon 1.55-eV excitation to be a super- herent artifact, predominantly as a result of the optical Stark effect, position of that of individual monolayer graphene and WS2, i.e., which has been widely reported previously (41). In contrast, for the exhibiting an overall negative photoconductivity with sub–10 ps g-WS2 heterostructure, we observe two PB signals at both the A- lifetime. However, we observe distinctively different dynamics for and B-exciton resonances of monolayer WS2 (2.0 and 2.4 eV, re- the heterostructure upon 1.55-eV excitation (Fig. 2A, red line): Af- spectively) for the same excitation energy and similar pump fluence, ter the short-lived negative photoconductivity, the photoconductiv- as shown in Fig. 2D. Note that, for monolayer WS2, A- and B-exciton ity turns positive within 10 ps, followed by a remarkably long-lived resonances originate from two spin-split valence bands at the K point positive photoconductivity with a lifetime extending 1 ns (without (29, 30). For the conduction bands, the spin splitting is usually small, decay in 1 ns for some cases, see statistics for the measurements of resulting in quasi-degeneracy of the two spin splitting bands (see seven different sample areas in section S3). While the short-lived Fig. 1D). Thanks to this effect, the simultaneous occurrence of PB at negative contribution can be attributed to the hot state of the both A- and B-exciton resonances provides additional evidence of graphene electronic system, the long-lived positive contribution the injection of (hot) electrons rather than holes from graphene to Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 3 of 10 SCIENCE ADVANCES | RESEARCH ARTICLE Fig. 2. Ultrafast interfacial CT and long-lived charge separation in graphene-WS2 vdW heterostructures. (A) OPTP carrier dynamics for graphene (gray circle points), WS2 (blue line), and the g-WS2 heterostructure (red line). Samples are measured with 1.55-eV excitation under dry N2 purging conditions. The absorbed photon densities are 3.8 × 1012, 3.7 × 1012, and 0.8 × 1012 cm−2 for WS2, g-WS2 heterostructure, and graphene, respectively. (B) Complex photoconductivity for g-WS2 heterostructure at a pump-probe delay of ~100 ps under 1.55-eV excitation with an absorbed photon density of 2.6× 1013 cm−2. Red and blue lines represent the Drude fit for the real and imaginary part of the complex photoconductivity, respectively. We use the conductivity unit in quantum conductance G0 = 2e2/h; 2D TA color map for (C) monolayer WS2 and (D) g-WS2 heterostructure following 1.55-eV excitation with pump fluences of 670 and 620 W, respectively. mOD, milli optical density. The white lines in (C) and (D) represent the spectral slices averaged between 0.5 and 1 ps. (E) TA carrier dynamics averaged over the A-exciton resonance (between 1.88 and 2.07 eV) for g-WS2 heterostructure (red line) and monolayer WS2 (blue line) following 1.55-eV excitation with a pump fluence of 670 and 620 W, respectively. (F) TA carrier dynamics averaged over the A-exciton resonance (between 1.88 and 2.07 eV) for g-WS2 heterostructure (red line) and monolayer WS2 (blue line) following 3.10-eV excitation with pump fluences of 248 and 250 W, respectively. a.u., arbitrary units. Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 4 of 10 SCIENCE ADVANCES | RESEARCH ARTICLE WS2 following sub-A-exciton excitation. If not, then one expects sites, which can electrostatically gate the graphene layer. By quanti- only PB at A-exciton resonance for the hole injection into WS2. fying the density of transferred charges (see the THz-TDS study in We find that the injected electrons in the excited states in WS2 the next section), we can set a lower limit for the defect density in decay within 1 ps as shown in Fig. 2E, in sharp contrast to over 1-ns our sample of 3.2 × 1011 cm−2, by assuming that the same amount of charge separation lifetime in g-WS2 heterostructures based on our charges are trapped for the photogating process. This value is con- THz results shown in Fig. 2A. Note that the short excited-state dy- sistent with the values reported in the literature (48, 49). Further namics observed here is in line with previous TA measurements studies using photo-electrochemical methods (50–52) to unveil the (10, 11). In these studies, the ultrafast depopulation process was at- energetics and the nature of the involved defects at the g-WS2 inter- tributed to back electron transfer from WS2 to graphene. Therefore, faces could shed light on this issue. By tuning the Fermi level in the large difference in the measured carrier lifetime in the donor graphene via, e.g., ionic gating, we can intentionally fill the defects (graphene) and the acceptor (WS2) suggests a new recombination in WS2 responsible for the photogating process. By monitoring the pathway. gate voltage–dependent recombination rates, one should be able to map out the density and energetics of the defect states. Interfacial photogating effect in g-WS2 heterostructure is supported by interfacial states CT rate and mechanism at g-WS2 interfaces Here, we attribute the long-lived (over ~1 ns) photoconductivity Exploiting the long-lived photogating effect, we further investigate in graphene (by THz spectroscopy) to an interfacial photogating the mechanism underlying ultrafast CT at g-WS2 vdW interfaces on mechanism mediated by interfacial states (e.g., defects) at g-WS2 early time scales. The positive photoconductivity originating from interfaces. In such a scheme, following photoexcitation, electrons in the photogating effect provides a direct measure to the number of WS2, either directly generated by excitation (with hv > 2 eV) or in- electrons transferred across the interfaces. We estimate the energy jected from graphene to the conduction band of WS2 (for hv < 2 eV), barrier for CT to be ~0.8 eV, by taking into account the energy dif- are subsequently trapped (within ~1 ps) to long-lived interfacial ference between WS2 conduction band minimum (CBM) and the states. The relatively long-lived trapped electrons can effectively Fermi level in graphene, as well as corrections including the band- gate graphene, leading to an efficient modification of the carrier gap reduction in WS2 due to dielectric screening of graphene (53), density (thus the Fermi level) in graphene (observed by THz spec- and shifts in energetics associated with CT (see details in section troscopy). Intriguingly, following the fast ~1-ps PB decay, the TA S4). We found that the estimated energy barrier for CT is in line dynamics is dominated by a much longer-lived, spectral shift (with with previous studies (10). Given this large barrier and the fact that the differential-like feature; see the dynamics and the spectral slices only graphene is excited by 1.55-eV pump photons (thermalized or in Fig. 2D). This feature can be understood by a transient electrical nonthermalized), hot carriers are necessarily involved in the inter- field–induced Stark effect, following ultrafast CT and trapping at facial CT process at g-WS2 interfaces. For most spectroscopic studies the interface. Such spectral shift due to the local electrical field built up to now, the optical excitation is limited to either below (10, 11) at the interfaces has previously been reported in heterostructures or only at the A-exciton resonance of TMDCs (12). To investigate consisting of semiconducting quantum dots (42). whether and how the excitation energy or the initial charge config- Now, we briefly discuss the possible origin of the “interfacial” uration affects CT channels and mechanisms, we monitor the CT states for photogating effect. First, as the energies of these states dynamics at various photoexcitation fluences and with a broad range lie between the conduction band of WS2 and the Fermi level of of photon energies (hv) from 0.7 eV (which only excites graphene) graphene, it seems reasonable to assume that they originate from to 3.1 eV (which excites both graphene and WS2), across the A-exciton defect states present in the WS2 layer or hybridized states at inter- resonance (2 eV) of WS2. faces. To test this hypothesis and shed light on the nature of these In Fig. 3A, we present typical fluence-dependent CT dynamics states, we have studied and compared the TA dynamics of mono- for g-WS2 heterostructure following 1.38-eV excitation. The long- layer WS2 and g-WS2 heterostructure following 3.10-eV excitation. lived positive photoconductivity increases with increasing absorbed As shown in Fig. 2F, we observe very similar decay dynamics for both photon density. To qualitatively describe the CT efficiency under samples (especially for the fast initial decay). This result strongly different excitation conditions, we extract the maximum positive suggests a universal recombination pathway at the first 1 to 2 ps, photoconductivity [∆max ~ (−E/E)max] for different pump ener- very likely due to trapping in the defect states in WS2. Note that this gies and fluences, as shown in Fig. 3B. For the pump photon energies assumption is in line with previous THz and TA dynamics reported and fluences used, we always observe positive long-lived photocon- in monolayer TMDCs, where the fast, ~1-ps decay is also assigned ductivity. This indicates that, for all cases, CT involves electron in- to charge trapping (43–45). In addition, a relatively slow process jection from graphene to WS2 (or equivalently, hole transfer from with >10-ps lifetime is also observed, which can be assigned to trap-­ WS2 to graphene). Intriguingly, for the lowest photoexcitation en- assisted recombination following previous reports (43, 46). ergy used in this study (0.7 eV), the excess energy (Eex = hv/2 − EF) According to literature, we speculate that the defect states could in the photogenerated nonthermalized hot electrons is only ~0.45 eV, originate from sulfur vacancies, which are ubiquitously present in which is way below the energy barrier for CT at g-WS2 interfaces CVD-WS2. For instance, a recent report combining ab initio GW (~0.8 eV). The fact that we nonetheless observe a positive photo- calculations and scanning tunneling spectroscopic studies has shown conductivity indicates that CT takes place following thermalization that sulfur vacancies in WS2 can generate two types of unoccupied via so-called photo-thermionic emission (27). This photo-thermionic in-gap states, which are located ~0.5 and 0.7 eV below the conduction emission scheme relies on very high electron temperatures in graphene band of WS2, respectively (47), with at least one of them lying between reached for typical incident excitation powers. Sufficiently hot elec- the Fermi level of graphene and the conduction band of monolayer trons in the valence band can be heated across the Dirac point into WS2. These empty defect states can serve as the “intermediate” trap the conduction band. Following this interband heating process (38), Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 5 of 10 SCIENCE ADVANCES | RESEARCH ARTICLE Fig. 3. CT efficiency and mechanism at graphene-WS2 interfaces. (A) Fluence-dependent CT dynamics at g-WS2 interfaces. Samples are measured using 1.38-eV excitation with a series of absorbed photon densities, from 6.6 × 1011 to 5.6 × 1012 cm−2. (B) Summary of the fluence-dependent photoconductivity maximum of −E/E [i.e., (−E/E)max] for different photon excitation energies, from hv = 0.71 to 3.10 eV. The data are described by a simple power law, namely, (−E/E)max = A·P ( ≥ 0), as described in the main text. (C) Photon energy–dependent power index  obtained by THz-OPTP, THz-TDS, and TA methods. For above-A-exciton excitations,  is close to 1 for TA and THz-TDS, while it is smaller than 1 for OPTP. We attribute this discrepancy to the different physical quantities measured by the tools (transferred charge carrier density for TA and THz-OPTP versus photoconductivity for OPTP; see details in the Supplementary Materials). (D) Quantification of photon energy–dependent CT efficiency in the heterostructure. The black dots represent the “relative” CT efficiency estimated by THz-OPTP data (the left Y axis), while the red diamonds stand for the absolute CT efficiency estimated by THz-TDS data (the right red Y axis), and the blue triangles represent the relative CT efficiency estimated by TA data in the sub-A-exciton excitation regime (the right blue Y axis). the thermalized hot electrons with sufficient energy over the energy In principle, the photoconductivity caused by the photo-thermionic barrier at g-WS2 interfaces can be injected into WS2. We calculate emission should exhibit a superlinear dependence on the pump flu- the electron temperature (Te) in graphene as a function of incident ence (27, 28) (see also the simulation result shown in fig. S3E). We fluence based on the photo-induced THz response of graphene (see therefore describe the pump fluence–dependent maximum positive details in section S5). As shown in fig. S3A, Te can reach 1500 to photoconductivity with P as shown in Fig. 3B, where P is the pump 3000 K over our fluence range. Figure S3C shows the resultant hot power, and  is the power index ( > 1 for superlinear dependence). electron distribution following the thermalization. A substantial In Fig. 3C, we show the extracted values of THz-OPTP for different fraction of thermalized hot electrons (~22.2% with Te of 2700 K) photon excitation energies. We find that THz-OPTP is bigger than reaches energies in excess of the conduction band in WS2, in good 1 when the photon excitation energy is below the A-exciton reso- agreement with our proposed photo-thermionic emission scheme. nance of WS2. With increasing photon excitation energy, THz-OPTP We conduct a similar analysis for thermalized hot holes in the gradually decreases and undergoes a transition around the A-exciton valence band in graphene. We observe a much smaller fraction (~1.6% resonance of WS2, from above 1 ( ~ 1.4 at 0.7-eV excitation, super- with Te of 2700 K) of thermalized hot holes with energy above the linear) to slightly below 1 ( ~ 0.9 for hv > 2 eV, sublinear). It is energy barrier for hole injection than that of thermalized hot elec- worth noting that in previous TA studies (10, 11), a linear fluence trons in our system (see fig. S3D), because of the larger energy dependence (i.e.,  = 1) of the signal was reported, following separation between the Fermi level in graphene and the valence sub-A-exciton excitation. Therefore, to further verify our THz band in WS2 (~1.3 eV). This result can explain our experimental results, we have repeated the fluence-dependent measurements for observation of the injection of thermalized hot electrons rather than the same sample using other complementary methods, including holes from graphene to WS2 in the study. THz-TDS and TA spectroscopy (see details of our TA study in the Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 6 of 10 SCIENCE ADVANCES | RESEARCH ARTICLE last section). The power indexes  obtained from these three meth- quasi-equilibrium following ultrafast CT, so that the carrier mobility ods are found to behave similarly, as shown in Fig. 3C (detailed  of graphene is nearly constant over time. As the photoconductivity analysis and discussions of the experimental data can be found in ∆ is proportional to the products of  and NCT(hv), ∆/Nabs at the section S6). This result shows that the change in  is independent of maximum positive photoconductivity reflects the relative change the used spectroscopic tools. in the CT efficiency under different photon excitation energies. We attribute this superlinear-to-sublinear pump fluence depen- In Fig. 3D, we plot the CT efficiencies obtained by (i) THz-TDS dence of the CT dynamics to a transition between two distinct CT (as right Y axis) and (ii) OPTP (as left Y axis) as a function of pho- regimes: HET via photo-thermionic emission for sub-A-exciton ex- ton excitation energy. For 1.55- and 3.10-eV excitations, where both citation (see Fig. 1C) and DHT for above-A-exciton excitation. The methods are applied, the CT efficiencies evaluated by these two DHT process involves the interfacial recombination of holes in the methods agree perfectly, validating our proposed methods. We ob- valence band of WS2 with valence band electrons in graphene (see serve that the CT efficiency shows a transition across the A-exciton Fig. 1D). A slightly sublinear feature in our study can be understood resonance of WS2: Below it, the CT efficiency is relatively inefficient as follows: With increasing fluence, many-body effects, for instance, with a quantum yield of <1%; above it, the CT efficiency increases exciton-exciton annihilation in TMDCs (54), can play a critical role with the increasing excitation energy and reaches ~5% for 3.10-eV in the charge carrier dynamics on the subpicosecond to picosecond excitation. One of the origins for the much-enhanced CT efficiency time scale. These many-body effects can result in a decrease in the in the DHT regime in comparison to that for the photo-thermionic hole density in TMDCs, which reduces the CT efficiency at high emission regime could be attributed to the density of electrons in fluence. Note that in the above-A-exciton excitation regime, hot graphene for CT: In the photo-thermionic emission regime, only a electrons in graphene, despite having higher energy than those limited number (on the order of 1011 to 1012 cm−2) of electrons with generated by sub-A-exciton excitation (for a given absorbed power), sufficiently high energy can be emitted over the energy barrier, contribute little to CT. This is mainly due to the much weaker ab- while in the DHT regime, the valence band electrons in graphene, sorption in graphene in comparison to that in WS2. On top of that, with a much higher density (on the order of ~1014 cm−2), can effi- we discuss other CT pathways, including photo-thermionic emis- ciently recombine with the photogenerated holes in the valence sion and hole transfer from WS2 to the photoexcited hot electrons band of WS2. Furthermore, in the DHT regime, we observe that in graphene in the above-A-exciton excitation regime (see the DHT efficiency goes up with increasing the excitation photon energy extended discussions in section S7) and argue that they cannot (hv). This result can be rationalized by partial extraction of hot holes compete with the DHT process. to graphene before cooling down to the A-exciton state in WS2. It is worth commenting that our results and the proposed micro- Our data here provide direct and strong evidence for the transi- scopic model agree well with a recent study by time-resolved angle-­ tion of the CT mechanism from HET (via photo-thermionic emis- resolved photoemission spectroscopy. In the study (12), an ultrafast sion when only graphene is excited) to DHT between the valence (sub–200 fs) loss of the valence band electrons in graphene has been bands from WS2 to graphene (in the above-A-exciton excitation directly observed in the same heterostructure by resonantly exciting regime). Last, in the sub-A-exciton excitation regime, while it is the A-exciton resonance of monolayer WS2. Furthermore, for the clear that HET via photo-thermionic emission takes place (THz- photocurrent generation in g-based heterostructures (21, 22), the OPTP > 1 for hv < 2.0 eV), it remains ambiguous if other HET photo-thermionic emission is proposed to govern the photocurrent pathways are possible. To check this aspect, we renormalize the generation for low photoexcitation energy, whereas direct interlayer maximum positive photoconductivity to the incident power (P, tunneling becomes the dominant mechanism for high photoexci- rather than Nabs as shown in Fig. 3D) and replot it versus the pump tation energy. Our results here are also in line with these macro- photon energy. As the electron temperature in graphene is indepen- scopic device results and further provide macroscopic mechanisms dent of the pump photon energy for a fixed incident power, the HET for the interfacial CT processes. Last, despite the clear difference in efficiency via pure photo-thermionic emission will be constant for CT mechanisms between the two regimes discussed, we note that different pump photon energies in such a plot (27, 28). As shown in the resultant charge configurations at the interfaces are indistin- fig. S6, for hv < 1 eV, the HET efficiency is nearly constant, in line guishable with electrons in the excited state of WS2 and holes in the with photo-thermionic emission; for 1.4 < hv < 2 eV, we observe an valence band of graphene. increase in the HET efficiency. This indicates that in the sub-A-­ To further confirm the proposed two CT regimes, we quantify exciton excitation regime, with sufficiently high pump photon energy, their efficiency at different excitation energies. Here, we use two other possible CT routes may also contribute to the CT process. approaches to determine the interfacial CT efficiency based on THz- One of the plausible scenarios is CT involving nonthermalized hot TDS and OPTP results: (i) we define the absolute CT efficiency as electrons (on top of photo-thermionic emission). Knowing the en-  = NCT(hv)/Nabs, where NCT(hv) is the number of electrons injected ergy difference between the WS2 CBM and graphene’s Dirac point from graphene to WS2 at excitation energy hv, and Nabs represents (~0.7 eV), the minimum pump energy required for nonthermalized the absorbed photon density in the donor, i.e., graphene for the hot electron injection is 2 × 0.7 eV = 1.4 eV, in line with the experi- photo-thermionic emission regime and WS2 for the DHT regime. mental observation. However, further studies are needed to rule out other NCT and Nabs are experimentally accessible by THz-TDS (see details possible CT mechanisms, such as DHT at graphene and small domains in section S2) and absorption measurements (see details in section of multilayer WS2 (with absorption onset around 1.3 eV) (21). S1), respectively; (ii) on the basis of the OPTP results shown in Fig. 3B, we can directly compare the maximum positive photocon- ductivity at a fixed absorbed photon density (approximately 5 × DISCUSSION 1012 cm−2). At the pump-probe delay where the positive photoconduc- We unveil new injection and recombination pathways in g-TMDCs tivity is maximum (~15 ps), charge carriers in our system reach and demonstrate the beneficial effect of interfacial defect states for Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 7 of 10 SCIENCE ADVANCES | RESEARCH ARTICLE promoting a long interfacial charge separation time and thus an amplifier from Light Conversion. The THz electrical field in the efficient photogating phenomenon in g-based vdW structures. Un- time domain is mapped out by electro-optic sampling method. derstanding such injection and recombination pathways is import- ant for both fundamental studies and optoelectronic applications. SUPPLEMENTARY MATERIALS For instance, the long-lived photogating effect reported here (medi- Supplementary material for this article is available at http://advances.sciencemag.org/cgi/ content/full/7/9/eabd9061/DC1 ated by ultrafast interfacial CT and interfacial trapping process) rec- onciles the reported very short charge separation time at g-TMDCs REFERENCE AND NOTES interfaces (10, 11) and, however, very efficient photodetectors based 1. Y. Liu, Y. Huang, X. Duan, Van der Waals integration before and beyond two-dimensional on these structures (5–7). While interfacial defects are beneficial in materials. Nature 567, 323–333 (2019). such circumstance, their impact on the performance of other opto- 2. D. Jariwala, T. J. Marks, M. C. Hersam, Mixed-dimensional van der Waals heterostructures. electronic devices depends strongly on the application: for instance, Nat. Mater. 16, 170–181 (2017). 3. K. S. Novoselov, A. Mishchenko, A. Carvalho, A. H. C. Neto, 2D materials and van der Waals high density of interfacial defects is troublesome for photovoltaics, heterostructures. Science 353, aac9439 (2016). where an efficient electron and hole separation toward electrodes is 4. A. K. Geim, I. V. Grigorieva, Van der Waals heterostructures. Nature 499, 419–425 required for highly performed devices. The short lifetime of inject- (2013). ed hot electrons in the conduction band of TMDCs imposes a lim- 5. F. H. L. Koppens, T. Mueller, P. Avouris, A. C. Ferrari, M. S. Vitiello, M. Polini, Photodetectors based on graphene, other two-dimensional materials and hybrid systems. ited time window for the efficient separation of charges toward Nat. Nanotechnol. 9, 780–793 (2014). electrodes. Passivation of the interfacial states or development of 6. M. Massicotte, P. Schmidt, F. Vialla, K. G. Schädler, A. Reserbat-Plantey, K. Watanabe, further extraction of injected hot electrons from TMDCs through T. Taniguchi, K. J. Tielrooij, F. H. L. Koppens, Picosecond photoresponse in van der Waals an ultrafast, subpicosecond channel is required. heterostructures. Nat. Nanotechnol. 11, 42–46 (2016). In summary, we investigate the nonequilibrium hot carrier 7. J. Shim, D.-H. Kang, Y. Kim, H. Kum, W. Kong, S.-H. Bae, I. Almansouri, K. Lee, J.-H. Park, J. Kim, Recent progress in Van der Waals (vdW) heterojunction-based electronic dynamics in g-WS2 vdW heterostructures combining ultrafast THz and optoelectronic devices. Carbon 133, 78–89 (2018). and TA spectroscopy. We report a transition in both CT efficiency 8. W. Zhang, C.-P. Chuu, J.-K. Huang, C.-H. Chen, M.-L. Tsai, Y.-H. Chang, C.-T. Liang, and mechanism by tuning the pump photon energies across the Y.-Z. Chen, Y.-L. Chueh, J.-H. He, Ultrahigh-gain photodetectors based on atomically thin A-exciton resonance of WS2. Upon excitation below the A-exciton graphene-MoS2 heterostructures. Sci. Rep. 4, 3826 (2014). 9. G. Konstantatos, M. Badioli, L. Gaudreau, J. Osmond, M. Bernechea, F. P. G. De Arquer, resonance of WS2, a relatively inefficiently HET via photo-thermionic F. Gatti, F. H. L. Koppens, Hybrid graphene–quantum dot phototransistors with ultrahigh emission governs the CT process, and only thermalized hot elec- gain. Nat. Nanotechnol. 7, 363–368 (2012). trons with sufficient energy can be injected into the excited states of 10. Y. Chen, Y. Li, Y. Zhao, H. Zhou, H. Zhu, Highly efficient hot electron harvesting WS2. In contrast, we show that highly energetic nonthermalized hot from graphene before electron-hole thermalization. Sci. Adv. 5, eaax9958 (2019). electrons in graphene do not contribute to the CT process. Rather, 11. L. Yuan, T. F. Chung, A. Kuc, Y. Wan, Y. Xu, Y. P. Chen, T. Heine, L. Huang, Photocarrier generation from interlayer charge-transfer transitions in WS2-graphene heterostructures. a relatively efficient (up to ~5%) DHT process occurs from the Sci. Adv. 4, e1700324 (2018). valence band of WS2 to the valence band of graphene. We show that 12. S. Aeschlimann, A. Rossi, M. Chávez-Cervantes, R. Krause, B. Arnoldi, B. Stadtmüller, the injected electrons only occupy the excited states of WS2 for M. Aeschlimann, S. Forti, F. Fabbri, C. Coletti, Direct evidence for efficient ultrafast ~1 ps and then get trapped and stored at vdW interfaces (probed by charge separation in epitaxial WS2/graphene heterostructures. Sci. Adv. 6, eaay0761 (2020). TA spectroscopy). This results in a long-lived photogating effect in 13. A. Rose, Concepts in Photoconductivity and Allied Problems (Interscience publishers, 1963). graphene over 1 ns (observed by THz spectroscopy). Our results 14. J. D. Mehew, S. Unal, E. Torres Alonso, G. F. Jones, S. Fadhil Ramadhan, M. F. Craciun, here provide new insights into both the CT mechanism and recom- S. Russo, Fast and highly sensitive ionic-polymer-gated WS2–graphene photodetectors. bination pathway at g-TMDCs vdW interfaces, which are critical Adv. Mater. 29, 1700222 (2017). to potential optoelectronic and energy-harvesting applications of 15. T. Chen, Y. Sheng, Y. Zhou, R. Chang, X. Wang, H. Huang, Q. Zhang, L. Hou, J. H. Warner, High photoresponsivity in ultrathin 2D lateral graphene: WS2: Graphene g-TMDCs vdW heterostructures. photodetectors using direct CVD growth. ACS Appl. Mater. Interfaces 11, 6421–6430 (2019). 16. M. Breusing, S. Kuehn, T. Winzer, E. Malić, F. Milde, N. Severin, J. P. Rabe, C. Ropers, MATERIALS AND METHODS A. Knorr, T. Elsaesser, Ultrafast nonequilibrium carrier dynamics in a single graphene layer. Phys. Rev. B. 83, 153410 (2011). Sample information 17. D. Brida, A. Tomadin, C. Manzoni, Y. J. Kim, A. Lombardo, S. Milana, R. R. Nair, The g-WS2 heterostructure on a sapphire substrate is obtained K. S. Novoselov, A. C. Ferrari, G. Cerullo, Ultrafast collinear scattering and carrier commercially (from SixCarbon Technology, Shenzhen) and pro- multiplication in graphene. Nat. Commun. 4, 1987 (2013). duced by CVD method. The heterostructure is characterized by 18. K. J. Tielrooij, L. Piatkowski, M. Massicotte, A. Woessner, Q. Ma, Y. Lee, K. S. Myhro, UV-vis spectroscopy, Raman spectroscopy, and THz-TDS to exam- C. N. Lau, P. Jarillo-Herrero, N. F. Van Hulst, F. H. L. Koppens, Generation of photovoltage in graphene on a femtosecond timescale through efficient carrier heating. ine the number of layers and EF in graphene. Nat. Nanotechnol. 10, 437–443 (2015). 19. K. J. Tielrooij, J. C. W. Song, S. A. Jensen, A. Centeno, A. Pesquera, A. Zurutuza Elorza, OPTP spectroscopy M. Bonn, L. S. Levitov, F. H. L. Koppens, Photoexcitation cascade and multiple hot-carrier To track ultrafast charge flow across the g-WS2 interfaces, we per- generation in graphene. Nat. Phys. 9, 248–252 (2013). 20. I. Gierz, J. C. Petersen, M. Mitrano, C. Cacho, I. C. E. Turcu, E. Springate, A. Stöhr, A. Köhler, form OPTP spectroscopy to study photoconductivity of the sam- U. Starke, A. Cavalleri, Snapshots of non-equilibrium Dirac carrier distributions ples. The THz spectrometer is powered by a femtosecond laser in graphene. Nat. Mater. 12, 1119–1124 (2013). (based on regenerative amplified and mode-locked Ti:sapphire) 21. Z. Mics, K.-J. Tielrooij, K. Parvez, S. A. Jensen, I. Ivanov, X. Feng, K. Müllen, M. Bonn, with a central wavelength of 800 nm, a pulse length of ~50 fs, and a D. Turchinovich, Thermodynamic picture of ultrafast charge transport in graphene. repeating frequency of 1 kHz. For optical excitations, the funda- Nat. Commun. 6, 7655 (2015). 22. J. C. W. Song, M. Y. Reizer, L. S. Levitov, Disorder-assisted electron-phonon scattering mental wavelength (800 nm) and the frequency doubled wave- and cooling pathways in graphene. Phys. Rev. Lett. 109, 106602 (2012). length (400 nm) by using a BiB3O6 crystal can be readily achieved; 23. M. W. Graham, S. F. Shi, D. C. Ralph, J. Park, P. L. McEuen, Photocurrent measurements other wavelengths are obtained by a commercial optical parametric of supercollision cooling in graphene. Nat. Phys. 9, 103–108 (2013). Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 8 of 10 SCIENCE ADVANCES | RESEARCH ARTICLE 24. A. C. Betz, S. H. Jhang, E. Pallecchi, R. Ferreira, G. Fève, J.-M. Berroir, B. Plaçais, Supercollision 49. J. Hong, Z. Hu, M. Probert, K. Li, D. Lv, X. Yang, L. Gu, N. Mao, Q. Feng, L. Xie, Exploring cooling in undoped graphene. Nat. Phys. 9, 109–112 (2013). atomic defects in molybdenum disulphide monolayers. Nat. Commun. 6, 6293 (2015). 25. M. T. Mihnev, F. Kadi, C. J. Divin, T. Winzer, S. Lee, C.-H. Liu, Z. Zhong, C. Berger, 50. S. C. Boehme, H. Wang, L. D. A. Siebbeles, D. Vanmaekelbergh, A. J. Houtepen, W. A. De Heer, E. Malic, Microscopic origins of the terahertz carrier relaxation and cooling Electrochemical charging of CdSe quantum dot films: Dependence on void size dynamics in graphene. Nat. Commun. 7, 11617 (2016). and counterion proximity. ACS Nano 7, 2500–2508 (2013). 26. T. Low, V. Perebeinos, R. Kim, M. Freitag, P. Avouris, Cooling of photoexcited carriers 51. S. C. Boehme, J. M. Azpiroz, Y. V. Aulin, F. C. Grozema, D. Vanmaekelbergh, in graphene by internal and substrate phonons. Phys. Rev. B. 86, 045413 (2012). L. D. A. Siebbeles, I. Infante, A. J. Houtepen, Density of trap states and Auger-mediated 27. M. Massicotte, P. Schmidt, F. Vialla, K. Watanabe, T. Taniguchi, K. J. Tielrooij, electron trapping in CdTe quantum-dot solids. Nano Lett. 15, 3056–3066 (2015). F. H. L. Koppens, Photo-thermionic effect in vertical graphene heterostructures. Nat. 52. S. C. Boehme, T. A. Walvis, I. Infante, F. C. Grozema, D. Vanmaekelbergh, L. D. A. Siebbeles, Commun. 7, 12174 (2016). A. J. Houtepen, Electrochemical control over photoinduced electron transfer 28. Q. Ma, T. I. Andersen, N. L. Nair, N. M. Gabor, M. Massicotte, C. H. Lui, A. F. Young, W. Fang, and trapping in CdSe-CdTe quantum-dot solids. ACS Nano 8, 7067–7077 (2014). K. Watanabe, T. Taniguchi, J. Kong, N. Gedik, F. H. L. Koppens, P. Jarillo-Herrero, Tuning 53. A. Raja, A. Chaves, J. Yu, G. Arefe, H. M. Hill, A. F. Rigosi, T. C. Berkelbach, P. Nagler, ultrafast electron thermalization pathways in a van der Waals heterostructure. Nat. Phys. C. Schüller, T. Korn, C. Nuckolls, J. Hone, L. E. Brus, T. F. Heinz, D. R. Reichman, 12, 455–459 (2016). A. Chernikov, Coulomb engineering of the bandgap and excitons in two-dimensional 29. Z. Y. Zhu, Y. C. Cheng, U. Schwingenschlögl, Giant spin-orbit-induced spin splitting materials. Nat. Commun. 8, 15251 (2017). in two-dimensional transition-metal dichalcogenide semiconductors. Phys. Rev. B. 84, 54. D. Sun, Y. Rao, G. A. Reider, G. Chen, Y. You, L. Brézin, A. R. Harutyunyan, T. F. Heinz, 153402 (2011). Observation of rapid exciton–exciton annihilation in monolayer molybdenum disulfide. 30. B. Zhu, X. Chen, X. Cui, Exciton binding energy of monolayer WS2. Sci. Rep. 5, 9218 Nano Lett. 14, 5625–5629 (2014). (2015). 55. E. del Corro, A. Botello-Méndez, Y. Gillet, A. L. Elias, H. Terrones, S. Feng, C. Fantini, 31. H. M. Hill, A. F. Rigosi, A. Raja, A. Chernikov, C. Roquelet, T. F. Heinz, Exciton broadening D. Rhodes, N. Pradhan, L. Balicas, X. Gonze, J.-C. Charlier, M. Terrones, M. A. Pimenta, in WS2/graphene heterostructures. Phys. Rev. B. 96, 205401 (2017). Atypical exciton–phonon interactions in WS2 and WSe2 monolayers revealed by 32. R. R. Nair, P. Blake, A. N. Grigorenko, K. S. Novoselov, T. J. Booth, T. Stauber, N. M. R. Peres, resonance Raman spectroscopy. Nano Lett. 16, 2363–2368 (2016). A. K. Geim, Fine structure constant defines visual transparency of graphene. Science 320, 56. J. Yan, Y. Zhang, P. Kim, A. Pinczuk, Electric field effect tuning of electron-phonon 1308 (2008). coupling in graphene. Phys. Rev. Lett. 98, 166802 (2007). 33. K. F. Mak, M. Y. Sfeir, Y. Wu, C. H. Lui, J. A. Misewich, T. F. Heinz, Measurement of the 57. A. Das, S. Pisana, B. Chakraborty, S. Piscanec, S. K. Saha, U. V. Waghmare, K. S. Novoselov, optical conductivity of graphene. Phys. Rev. Lett. 101, 196405 (2008). H. R. Krishnamurthy, A. K. Geim, A. C. Ferrari, A. K. Sood, Monitoring dopants by Raman 34. R. Ulbricht, E. Hendry, J. Shan, T. F. Heinz, M. Bonn, Carrier dynamics in semiconductors scattering in an electrochemically top-gated graphene transistor. Nat. Nanotechnol. 3, studied with time-resolved terahertz spectroscopy. Rev. Mod. Phys. 83, 543–586 (2011). 210–215 (2008). 35. S.-F. Shi, T.-T. Tang, B. Zeng, L. Ju, Q. Zhou, A. Zettl, F. Wang, Controlling graphene 58. M. Tinkham, Energy gap interpretation of experiments on infrared transmission through ultrafast hot carrier response from metal-like to semiconductor-like by electrostatic superconducting films. Phys. Rev. 104, 845–846 (1956). gating. Nano Lett. 14, 1578–1582 (2014). 59. S. H. Abedinpour, G. Vignale, A. Principi, M. Polini, W. K. Tse, A. H. Macdonald, Drude 36. S. A. Jensen, Z. Mics, I. Ivanov, H. S. Varol, D. Turchinovich, F. H. L. Koppens, M. Bonn, weight, plasmon dispersion, and ac conductivity in doped graphene sheets. Phys. Rev. B. K.-J. Tielrooij, Competing ultrafast energy relaxation pathways in photoexcited graphene. 84, 045429 (2011). Nano Lett. 14, 5839–5845 (2014). 60. J. Horng, C. F. Chen, B. Geng, C. Girit, Y. Zhang, Z. Hao, H. A. Bechtel, M. Martin, A. Zettl, 37. A. J. Frenzel, C. H. Lui, Y. C. Shin, J. Kong, N. Gedik, Semiconducting-to-metallic M. F. Crommie, Y. R. Shen, F. Wang, Drude conductivity of Dirac fermions in graphene. photoconductivity crossover and temperature-dependent Drude weight in graphene. Phys. Rev. B. 83, 165113 (2011). Phys. Rev. Lett. 113, 56602 (2014). 61. E. H. Hwang, S. Adam, S. Das Sarma, Carrier transport in two-dimensional graphene 38. A. Tomadin, S. M. Hornett, H. I. Wang, E. M. Alexeev, A. Candini, C. Coletti, D. Turchinovich, layers. Phys. Rev. Lett. 98, 186806 (2007). M. Kläui, M. Bonn, F. H. L. Koppens, E. Hendry, M. Polini, K.-J. Tielrooij, The ultrafast 62. H. Henck, Z. B. Aziza, D. Pierucci, F. Laourine, F. Reale, P. Palczynski, J. Chaste, M. G. Silly, dynamics and conductivity of photoexcited graphene at different Fermi energies. Sci. F. Bertran, P. L. Fèvre, E. Lhuillier, T. Wakamura, C. Mattevi, J. E. Rault, M. Calandra, Adv. 4, eaar5313 (2018). A. Ouerghi, Electronic band structure of two-dimensional WS2/graphene van der Waals 39. H. I. Wang, M.-L. Braatz, N. Richter, K.-J. Tielrooij, Z. Mics, H. Lu, N.-E. Weber, K. Müllen, heterostructures. Phys. Rev. B. 97, 155421 (2018). D. Turchinovich, M. Kläui, M. Bonn, Reversible photochemical control of doping levels 63. A. Ramasubramaniam, Large excitonic effects in monolayers of molybdenum in supported graphene. J. Phys. Chem. C 121, 4083–4091 (2017). and tungsten dichalcogenides. Phys. Rev. B. 86, 115409 (2012). 40. G. Jnawali, Y. Rao, H. Yan, T. F. Heinz, Observation of a transient decrease in terahertz 64. S. Das, H.-Y. Chen, A. V. Penumatcha, J. Appenzeller, High performance multilayer MoS2 conductivity of single-layer graphene induced by ultrafast optical excitation. Nano Lett. transistors with scandium contacts. Nano Lett. 13, 100–105 (2013). 13, 524–530 (2013). 65. Y. Liu, P. Stradins, S.-H. Wei, Van der Waals metal-semiconductor junction: Weak 41. S. L. Palfrey, T. F. Heinz, Coherent interactions in pump–probe absorption measurements: Fermi level pinning enables effective tuning of Schottky barrier. Sci. Adv. 2, e1600069 The effect of phase gratings. JOSA B. 2, 674–679 (1985). (2016). 42. G. Grimaldi, R. W. Crisp, S. Ten Brinck, F. Zapata, M. Van Ouwendorp, N. Renaud, 66. T. Le Quang, V. Cherkez, K. Nogajewski, M. Potemski, M. T. Dau, M. Jamet, P. Mallet, N. Kirkwood, W. H. Evers, S. Kinge, I. Infante, L. D. A. Siebbeles, A. J. Houtepen, J.-Y. Veuillen, Scanning tunneling spectroscopy of van der Waals graphene/ Hot-electron transfer in quantum-dot heterojunction films. Nat. Commun. 9, 2310 semiconductor interfaces: Absence of Fermi level pinning. 2D Mater. 4, 35019 (2017). (2018). 67. K. I. Bolotin, K. J. Sikes, Z. Jiang, M. Klima, G. Fudenberg, J. Hone, P. Kim, H. L. Stormer, 43. H. Wang, C. Zhang, F. Rana, Ultrafast dynamics of defect-assisted electron–hole Ultrahigh electron mobility in suspended graphene. Solid State Commun. 146, 351–355 recombination in monolayer MoS2. Nano Lett. 15, 339–345 (2015). (2008). 44. P. D. Cunningham, K. M. McCreary, A. T. Hanbicki, M. Currie, B. T. Jonker, L. M. Hayden, 68. W. Norimatsu, M. Kusunoki, Epitaxial graphene on SiC {0001}: Advances and perspectives. Charge trapping and exciton dynamics in large-area CVD grown MoS2. J. Phys. Chem. C Phys. Chem. Chem. Phys. 16, 3501–3511 (2014). 120, 5819–5826 (2016). 69. L. Banszerus, M. Schmitz, S. Engels, J. Dauber, M. Oellers, F. Haupt, K. Watanabe, 45. C. J. Docherty, P. Parkinson, H. J. Joyce, M.-H. Chiu, C.-H. Chen, M.-Y. Lee, L.-J. Li, L. M. Herz, T. Taniguchi, B. Beschoten, C. Stampfer, Ultrahigh-mobility graphene devices M. B. Johnston, Ultrafast transient terahertz conductivity of monolayer MoS2 and WSe2 from chemical vapor deposition on reusable copper. Sci. Adv. 1, e1500222 (2015). grown by chemical vapor deposition. ACS Nano 8, 11147–11153 (2014). 46. H. Wang, C. Zhang, F. Rana, Surface recombination limited lifetimes of photoexcited Acknowledgments: We thank P. Soltani, A. S. Hassan, Z. Liu, M. Ballabio, A. Tries, K. Krewer, carriers in few-layer transition metal dichalcogenide MoS2. Nano Lett. 15, 8204–8210 B. Kutus, and M. Kläui for the fruitful discussions. Funding: S.F. acknowledges fellowship (2015). support from Chinese Scholarship Council (CSC). X.J. acknowledges financial support by DFG 47. B. Schuler, D. Y. Qiu, S. Refaely-Abramson, C. Kastl, C. T. Chen, S. Barja, R. J. Koch, through the Excellence Initiative by the Graduate School of Excellence Materials Science in D. F. Ogletree, S. Aloni, A. M. Schwartzberg, Large spin-orbit splitting of deep in-gap Mainz (MAINZ) (GSC 266) and support from the Max Planck Graduate Center mit der Johannes defect states of engineered sulfur vacancies in monolayer WS2. Phys. Rev. Lett. 123, 76801 Gutenberg-Universität Mainz (MPGC). A.J.H. acknowledges support from the European (2019). Research Council Horizon 2020 ERC grant no. 678004 (Doping on Demand). ICN2 was 48. M. R. Rosenberger, H.-J. Chuang, K. M. McCreary, C. H. Li, B. T. Jonker, Electrical supported by the Severo Ochoa program from Spanish MINECO (grant no. SEV-2017-0706). characterization of discrete defects and impact of defect density on photoluminescence K.-J.T. acknowledges funding from the European Union’s Horizon 2020 Research and in monolayer WS2. ACS Nano 12, 1793–1800 (2018). Innovation Programme under grant agreement no. 804349 (ERC StG CUHL) and financial Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 9 of 10 SCIENCE ADVANCES | RESEARCH ARTICLE support through the MAINZ Visiting Professorship. Author contributions: H.I.W. conceived Submitted 20 July 2020 and supervised the project. S.F. conducted THz studies, with help from X.J., J.X., X.Y., H.Z., W.Z., Accepted 12 January 2021 and S.K. I.d.F. conducted TA measurements and analyzed the data under the supervision of Published 26 February 2021 A.J.H. X.J. and K.-J.T. modeled the electron temperature in graphene based on THz 10.1126/sciadv.abd9061 conductivity. All authors contributed to data interpretation. S.F. and H.I.W. wrote the paper with input from all authors. Competing interests: The authors declare that they have no Citation: S. Fu, I. du Fossé, X. Jia, J. Xu, X. Yu, H. Zhang, W. Zheng, S. Krasel, Z. Chen, Z. M. Wang, competing interests. Data and materials availability: All data needed to evaluate the K.-J. Tielrooij, M. Bonn, A. J. Houtepen, H. I. Wang, Long-lived charge separation following conclusions in the paper are present in the paper and/or the Supplementary Materials. pump-wavelength–dependent ultrafast charge transfer in graphene/WS2 heterostructures. Additional data related to this paper may be requested from the authors. Sci. Adv. 7, eabd9061 (2021). Fu et al., Sci. Adv. 2021; 7 : eabd9061 26 February 2021 10 of 10