The Berry phase not only has close connections to the optical selection rules that allow optical generation and detection of the valley-polarized carriers by circularly polarized photons, but also plays a central role in novel electron dynamics and transport phenomena in TMD and graphene layers, such as the valley Hall effect. In principle the Berry phase, together with other effects from inversion symmetry breaking, can have profound consequences for the wave function and energy spectrum of the excited states in two-dimensional materials. TMD monolayers are known to host strongly bound excitons with a remarkably large exciton binding energy due to enhanced Coulomb interactions in 2D. It was recently predicted that the Berry curvature of Bloch states can add an anomalous term to the group velocity of electrons and holes and creates an energy splitting between exciton states with opposite angular momentum Fig. 1a shows a simplified exciton energy spectrum illustrating the exciton fine structure based on our ab initio GW-Bethe-Salpeter equation calculations. The 2p+ and 2p− exciton states are split in energy with opposite order for the K and K’ valleys due to the opposite chirality in the two valleys. Such novel exciton fine structure, which embodies important wave function properties arising from the Bloch band geometry, can strongly modify the intraexcitonic light-matter interactions. Experimental observation of this predicted exciton spectrum, however, has been challenging, blueberry grow pot because it requires new spectroscopic probe that can distinguish both the momentum valley and the exciton angular momentum. Here, we report the first observation of the Berry-phase effect in the exciton spectrum of MoSe2 monolayer using intraexciton optical Stark spectroscopy.
We demonstrate that the degeneracy between the 2p±-exciton states is lifted by the Berry phase effect, and enabling a valley-dependent Autler-Townes doublet from strong intraexciton light-matter coupling. We coherently drive the intraexciton transitions using circularly-polarized infrared radiation, which couples the 1s exciton to the 2p+ or 2p− states selectively through the pump photon polarization . The pump-induced changes in the 1s exciton transition are detected by circularly polarized probes, which selectively measure the K or K’-valley excitons. Independent control of pump and probe photon polarization enables us to distinguish the exciton fine structures in the K and K’-valleys. We determine an energy splitting of 14 meV between the 2p+ and 2p− exciton states within a single valley, and this energy splitting changes sign between K and K’-valleys. We determine the 1s-2p transition dipole moment to be 55±6 Debye. This leads to an optical Stark shift that is almost 40 times larger than the interband counterpart under the same pump detuning and driving optical field strength. Such strong and valley-dependent intraexciton transitions open-up new pathways for the coherent manipulation of quantum states in 2D semiconducting materials using infrared radiation. To investigate the fine structure of the excitonic p-manifold, we fabricated a high quality MoSe2 monolayer that is encapsulated in hexagonal boron nitride layers using mechanical exfoliation and stacking following Ref. 21. The sandwiched hBN-MoSe2-hBN heterostructure was then transferred to an alumina-coated silver surface . The device was kept in vacuum at 77K for all optical measurements. This Aexciton peak arises from the optical transition between the ground state and the lowest energy 1s exciton state in MoSe2 monolayer, which is well-separated from the higher-lying exciton states due to strong Coulomb interactions in TMD monolayers.
It shows clearly that the 1s exciton transition exhibits avoided-crossing behavior in both valleys, which evolves gradually from energy blue shift to splitting and then to redshift as the pump photon energy is decreased. Due to the time-reversal symmetry between K and K’-valleys in MoSe2 monolayer, this observation also indicates that the 2p+ and 2p− exciton states are non-degenerate and has an energy difference of 14 meV in a single valley. We further plot the blue- and red-shifted 1s resonance as a function of the infrared pump photon energy in Fig. 3b. We find that the energy shifts induced by the intraexciton optical Stark effect are almost 40 times larger than its interband counterpart at the same pump intensity and resonance detuning21–23. To better understand the experimental results, we performed ab initio GW-BSE calculations using the BerkeleyGW26–28 package to determine the exciton energy levels and optical selection rules of exciton and intraexciton transitions in monolayer MoSe2. In these calculations, environmental screening effects from the hexagonal boron nitride encapsulation layers are included18 from first-principles . The simulation confirms the energy level diagram of the 1s, 2p+, and 2p− excitons and the optical selection rules in K and K’- valleys in Fig. 1a. Our calculations find that the energies of the 1s and 2p− exciton states are separated by 117 meV, with 2p+ exciton states further separated by 7 meV in K-valley. The energetic order of 2p+ , and 2p− excitons states is opposite in the K’-valley, as a result of time-reversal symmetry. Although the 2p± excitons are dark in linear optics, they are optically active when coupled to the 1s exciton with circularly-polarized light . For example, our calculations show that the 1s-2p+ intraexciton transition couple exclusively to the left-handed circularly polarized light with a transition dipole moment of 42 Debye. The 1s-2p− intraexciton transition, on the other hand, coupled exclusively to the right-handed circularly polarized light.
The experimentally observed intraexciton dipole moment and valley-dependent exciton fine structure match reasonably well with the ab initio GW-BSE calculations. The combination of 2p±-exciton splitting and extremely strong intraexcitonic light-matter interaction allow us to observe valley-dependent Autler-Townes doublets at higher pump intensity in MoSe2 monolayer. Towards this goal, we fabricated a hBN-encapsulated MoSe2 heterostructure on a zinc-sulphide substrate, where the local field factor on the sample for the infrared pump light is more favorable than that for MoSe2 on alumina coated silver substrate . Fig. 4c,d show the splitting energy in the Autler-Townes doublet at resonant excitation scales linearly with the excitation field strength, as expected from Eq. 129,30. At an effective driving intensity of 50±10 MW/cm2, which corresponds to a local optical field strength of 200±20 kV/cm, the Autler-Townes splitting can reach ~24 meV in both valleys. This Autler-Townes doublet leads to a valley dependent electromagnetically induced transparency in the 1s exciton transition, where the absorption at the 1s exciton resonance is reduced by more than 10-fold compared to the undriven exciton . Our findings offer a new and effective pathway to coherently manipulate the quantum states and excitonic excitations using infrared radiation coupled to the 1s-2p+ intraexciton transition.The dashed-lines indicate the peak position of unperturbed A-exciton. The dotted lines are guides to the eyes for the peak position at different driving energies. The spectra are offset for clarity and labelled according to the excitation energy . The spectra evolve from energy redshift to splitting and then to blue shift, as the driving energy is increased. The calculation is based on the Hamiltonian shown in Eq. 1. Exciton-photon coupling leads to avoided-crossing and the observed peak splitting at resonant coupling. This resonant coupling occurs at driving photon energy of 142 meV and 128 meV in the K and K’ valleys, respectively. The MoSe2 monolayer encapsulated in hBN flakes were prepared with a polyethylene terephthalate stamp by a dry transfer method21. Monolayer MoSe2 and hBN flakes were first exfoliated onto silicon substrate with a 90 nm oxide layer. We used PET stamp to pick-up the top hBN flake, monolayer MoSe2, and bottom hBN flake in sequence with accurate alignment based on an optical microscope. The hBN/MoSe2/hBN heterostructure was then stamped on a silver substrate coated with a 85 nm alumina layer or on a zinc sulphide substrate. Polymer and samples were heated to 60oC for the pickup and 130oC for the stamping process. Finally, hydroponic bucket the PET was dissolved in dichloromethane for 12 hours at room temperature. The sample temperature was kept at 77 K in a liquidnitrogen cooled cryostat equipped with BaF2 window during optical measurements. Pump-probe spectroscopy study is based on a regenerative amplifier seed by a mode-locked oscillator . The regenerative amplifier delivers femotosecond pulses at a repetition rate of 150 kHz and a pulse duration of 250 fs, which were split into two beams. One beam was used to pump an optical parametric amplifier and the other beam was focused onto a sapphire crystal to generate supercontinuum light for probe pulses. Femtosecond mid-infrared pump pulses with tunable photon energies were generated via difference frequency mixing of the idler pulses from the optical parametric amplifier and residual of fundamental output from regenerative amplifier in a 1 mm thick silver gallium sulphide crystal.
The mid-infrared pulse duration is ~350 fs. The pump-probe time delay was controlled by a motorized delay stage. The probe light was detected by high sensitivity CCD line camera operated at 1000 Hz. The helicity of pump and probe pulses was independently controlled using Fresnel rhomb and broadband quarter-wave plates, respectively. The experiment followed a reflection configuration with a normal incidence and collinear pump-probe geometry, where the absorption spectra are extracted from the reflectance contrast as described in the supporting information.This work was primarily supported by the Center for Computational Study of Excited State Phenomena in Energy Materials, which is funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division under Contract No. DE-AC02-05CH11231, as part of the Computational Materials Sciences Program which provided the experimental measurements and GW-BSE calculations. The sample fabrication and linear optical spectroscopy was supported by the US Army Research Office under MURI award W911NF-17-1-0312. The pump-probe setup was supported by the ARO MURI award W911NF- 15-1-0447. This research used resources of the National Energy Research Scientific Computing Center , a DOE Office of Science User Facility supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231, and the Extreme Science and Engineering Discovery Environment , which is supported by National Science Foundation grant number ACI-1548562. S.T. acknowledges support from NSF DMR-1552220. K.W. and T.T. acknowledge support from the Elemental Strategy Initiative conducted by the MEXT, Japan and the CREST , JST. E.C.R acknowledges support from the Department of Defense through the National Defense Science & Engineering Graduate Fellowship Program. C.-K.Y. and C.S.O. acknowledges useful discussion with Prof. Ajit Srivastava. Viruses are obligate intracellular pathogens that require living host cells to replicate and spread in the infected plant. During compatible interactions, viruses overcome the plant immune system and hijack host cellular processes to establish active infections . Viruses disrupt the plant cell cycle, inhibit cell death pathways, restrict macromolecular trafficking, alter cell signaling, protein turnover, and transcriptional regulation, and suppress defense mechanisms. The interference with these processes in the host leads to a wide range of plant developmental and physiological defects . Cultivated grapevines are highly susceptible to a variety of viruses and viroids, which cause significant crop losses and shorten the productive life of vineyards. More than 65 different viral species classified in at least 15 families have been reported to infect grapevines, which represents the highest number of viruses so far detected in a single cultivated plant species . Although these viruses are generally transmitted by plant-feeding insects or soilborne nematodes, they can also be spread through infected propagation material . Grapevine red blotch is a viral disease discovered in northern California in 2008 that has become a major economic problem for the wine industry in the USA . This disease is caused by the Grapevine red blotch-associated virus , a circular ssDNA virus with resemblance to geminiviruses, which infects wine grape cultivars with significant detrimental effects on productivity . The incidence and severity of the red blotch symptoms vary depending on the grape cultivar, environmental conditions, and cultural practices . In red-skinned varieties, GRBaV infections result in the appearance of red patches on the leaf blades, veins, and petioles; in white-skinned varieties, they manifest as irregular chlorotic regions on the leaf blades. GRBaV also affects berry physiology, causing uneven ripening, higher titratable acidity, and lower sugar and anthocyanin content, among others . Consequently, must and wine produced from infected berries present altered flavor and aroma. To date, there is limited information on how GRBaV infections affect grape metabolism. Comprehensive analyses to study specific cellular processes that GRBaV exploits to promote infections in berries are still needed, in particular those that relate to changes in berry chemical composition during fruit development. Grape berry development exhibits a double sigmoid growth pattern with three distinct phases: early fruit development, lag phase, and berry ripening.