Spin and Valley Dynamics in 2D van der Waals Materials – 2D-vdW-Spin

For the majority of samples investigated in publications P1-P14 our approach was as follows. : An exfoliated monolayer MoS2 with large lateral size of 120 µm lies on top of a 150 nm thick hBN, while different exfoliated WSe2 monolayers (red-dashed lines) are deliberately transferred on top of MoS2 in different twist angles after alignment of the long edges of the flakes. As a result, three different heterostructures are formed, namely HS1, HS2 and HS3. In all cases, there is optical access to the bare monolayers to confirm the validity of the results collected from the heterostructures. The same, thin 10 nm top-hBN covers the whole structure. The uniformity of the top and bottom hBN thickness is important because thin-film interference effects can modify the SHG intensity and the reflectivity shape/amplitude when comparing different samples. We measure differential reflectivity, see P13 for details. The overall shape of the differential reflectivity spectrum depends on cavity effects (thin layer interference) given by the top and bottom hBN and SiO2 thicknesses. As a result, the exciton transition lineshape varies in amplitude and sign in the presented spectra.
For SHG measurements, we use pulses with a pulse duration of 1ps. The pulses are generated by a tunable optical parametric oscillator (OPO) synchronously pumped by a mode-locked Ti:sapphire laser. The SHG signal is collected in reflection geometry while before any spectral acquisition the pulse duration and spectral shape of the fundamental are monitored for every different wavelength by an autocorrelator and a wavemeter. In addition, the average power of the fundamental is adjusted according to a power meter, normalized to each wavelength. This allows us to directly compare the measurements from different samples. For the polarization-resolved SHG experiments a superachromatic half-lambda waveplate was used, the error in retardance within the wavelength range studied is typically smaller than 0.1 %.

The Project main results include :
This project on investigation of polarization dynamics in atomically thin layers of transition metal dichaclogenides relies on innovative spectroscopy techniques put in place and performed by the partners on high quality sample structures, designed, fabricated and tested by the partners. We have fabricated high quality heterostructures of 4 different active materials, namely MoSe2, MoTe2, WSe2 and MoS2 monolayers encapsulated in hBN. We have shown that all 4 materials show narrow optical transition linewidth and for MoSe2 and MoTe2 we have carried out detailed linear and non-linear optical spectroscopy experiments [P1, P6, P7, P8, P9, P10, P12, P 13]. Initially, high quality samples for this project have only been obtained for samples with active regions encapsulated in exfoliated hBN flakes. This approach works for samples of several micrometer lateral size, but ideally larger area samples are desirable. We have take a first step towards up-scaling by using not an exfoliated flake but epitaxially grown hBN as the bottom, base layer of our structure. The hBN covers the whole substrate of about 1 cm x 1 cm in size and we transferred the active MoSe2 and MoS2 layer on top. The resulting optical spectra show narrow linewidth in these hybrid samples [P2, P4, P9, P11] . We also performed work on multilayer samples for quantum state control [P3, P5, P14,]and reported a new fabrication route for upscaling monolayer materials [P15].

The versatile optical spectroscopy techniques and sample fabrication routes investigated during the ANR project 2D vdW Spin are important tools for uncovering optical properties of new materials from the large catalogue of layered compounds that await investigation. Interesting technical developments in optical spectroscopy are underway with the main targets of improving spatialresolution, accessible wavelength range and compatibility with other microscopy techniques. One of the future challenges is to image moiré potentials and perform spectroscopy on a particular local atomic registry that lies on the nanometre scale. This will be important for understanding the origin of localized emitters and how optical properties depend on moiré periodicity. Nanometre- scale moiré superlattices can be imaged by using near- field spectroscopy techniques, SEM, atomic- resolution transmission electron microscopy, atomic force microscopy or scanning tunnelling spectroscopy. One possible way to increase spatial resolution in optical spectroscopy below the diffraction limit is to perform tip- enhanced spectroscopy, as recently demonstrated. In this technique, a nanometre- sized silver tip is scanned across a nanobubble in WSe2 to locally enhance emission. PL spectra are collected by illuminating the tip by laser in a in a near-field scanning optical microscope.

P1. «Exciton states in monolayer MoSe2 and MoTe2 probed by upconversion spectroscopy«
Physical Review X 8, 031073 (2018)

P2. «Spectrally narrow exciton luminescence from monolayer MoS2 exfoliated onto epitaxially grown hexagonal BN«
Applied Physics Letters 113, 032106 (2018)

P3. «Interlayer excitons in bilayer MoS2 with strong oscillator strength up to room temperature«
Physical Review B 99, 035443 (2019)

P4 “Controlling interlayer excitons in MoS2 layers grown by chemical vapor deposition”
Nature Communications 11, 2391 (2020)

P5 «Giant Stark splitting of an exciton in bilayer MoS2«
Nadine Leisgang, Shivangi Shree, Ioannis Paradisanos, Lukas Sponfeldner, Cedric Robert, Delphine Lagarde, Andrea Balocchi, Kenji Watanabe, Takashi Taniguchi,Xavier Marie, Richard J. Warburton, Iann C. Gerber, and Bernhard Urbaszek
Nature Nanotechnology 15, pages 901–907 (2020)

P6 «Observation of exciton-phonon coupling in MoSe2 monolayers«
S. Shree, M. Semina, C. Robert, B. Han, T. Amand, A. Balocchi, M. Manca, E. Courtade, X. Marie, T. Taniguchi, K. Watanabe, M. M. Glazov, and B. Urbaszek
Physical Review B 98, 035302 (2018) Editors' suggestion and arXiv:1804.06340

P7 “Control of the Exciton Radiative Lifetime in van der Waals Heterostructures”
Physical Review Letters 123, 067401 (2019)

P8 “Revealing exciton masses and dielectric properties of monolayer semiconductors with high magnetic fields”
Nature Communications 10, 4172 (2019)

P9 “High optical quality of MoS2 monolayers grown by chemical vapor deposition”
Shivangi Shree, Antony George, Tibor Lehnert, Christof Neumann, Meryem Benelajla, Cedric Robert, Xavier Marie, Kenji Watanabe, Takashi Taniguchi, Ute Kaiser, Bernhard Urbaszek*, Andrey Turchanin
2D Materials 7, 015011(2020)

P10 «Measurement of the Spin-Forbidden Dark Excitons in MoS2 and MoSe2 monolayers«
Nature Communications 11, 4037 (2020)

P11 «Unveiling the optical emission channels of monolayer semiconductors coupled to silicon nanoantennas«
Jean-Marie Poumirol, et al
ACS Photonics 2020 dx.doi.org/10.1021/acsphotonics.0c01175

P12 «Efficient phonon cascades in hot photoluminescence of WSe2 monolayers«
Ioannis Paradisanos, Gang Wang, Evgeny M Alexeev, Alisson R Cadore, Xavier Marie, Andrea C Ferrari, Mikhail M Glazov, Bernhard Urbaszek
Nature Communications 12, Article number: 538 (2021)

P13 “Guide to optical spectroscopy of layered semiconductors”
Shivangi Shree, Ioannis Paradisanos, Xavier Marie, Cedric Robert and Bernhard Urbaszek
Nature Reviews Physics 3, 39–54 (2021)

P14 “Interlayer exciton mediated second harmonic generation in bilayer MoS2”
Shivangi Shree et al,
NATURE COMMUNICATIONS | (2021) 12:6894 |

P15 “Room Temperature Micro-Photoluminescence Studies of Colloidal WS2 Nanosheets”
André Philipp Frauendorf, André Niebur, Lena Harms, Shivangi Shree, Bernhard Urbaszek, Michael Oestreich, Jens Hübner, and Jannika Lauth«
J. Phys. Chem. C 2021, 125, 34, 18841–18848 ;

Following the extremely successful research on graphene, work on layered van der Waals materials in general has opened up many new lines of research and potential applications. Here monolayer (ML) MoS2, MoSe2, WS2, and WSe2 stand out due to their versatility : these semiconductors have a direct-bandgap in the visible region of the optical spectrum located at the K-points of the Brillouin zone. Besides their promise for 2D opto-electronics, ML transition metal dichalcogenides (TMDCs) are interesting for exploiting both the spin and valley pseudospin of electrons and holes for potential applications in (quantum) information processing. The information may be encoded not only by whether an electron (or hole) has spin up or down, but also by whether it resides in the K+ or K- valley in momentum space. Strong spin-orbit coupling separates the spin states in energy in the valence and conduction band and crystalline asymmetry leads to valley specific optical selection rules for interband transitions. Therefore K+ or K- valleys can be selectively populated and probed using polarized light, in contrast with most conventional III-V, II-VI, and group-IV semiconductors.
The intrinsic timescales of spin and valley dynamics in ML TMDCs are therefore of considerable interest and the consortium LPCNO Toulouse – Leibniz University Hannover will investigate the mechanisms governing the stability of spin and valley polarization and tune the polarization lifetime. Both partners have already successfully conducted common research projects in the field of all optical spin noise spectroscopy.
Optical properties of TMDC ML are governed by excitons, which decay very rapidly on ps timescales and are therefore of limited use for valley and spin manipulation schemes.
Here we concentrate on longer-lived excitations with more stable spin and valley states. Our approach is to optically initialize spin and valley polarization for hole and electron doped samples. The doping is achieved chemically or in gated structures already operational in Toulouse. We will investigate and tune the spin dynamics with 3 different techniques.
First, time resolved photoluminescence measurements will monitor the polarization dynamics and transfer during the radiative lifetime of the charged exciton (trion). Here also the role of spin-forbidden (‘dark’) trions will be studied.
Second, to probe the spin and valley polarization of resident electrons and holes after recombination we will employ pump-probe measurements developed in Hannover.
Third, as a world premier in this system, we aim to perform spin noise measurements that give access to the long-term, intrinsic polarization lifetimes of carriers. The key advantage of noise spectroscopy is to access the complete temporal dynamics at thermal equilibrium without driving the system out of equilibrium and therefore avoiding a possible alteration of the intrinsic spin dynamics. The Hannover group is a world leader in this field and will study the underlying depolarization mechanisms as we go from n- to p-type samples and vary the carrier concentration and external parameters (temperature, E-and B-fields).
High quality samples are already available to the consortium, with optical transition linewidth approaching the homogenous limit in encapsulated samples fabricated in Toulouse (below 2 meV at FWHM). We will also study different alloys and materials with spin splittings that vary in sign and magnitude. The role of localization will be investigated as the polarization dynamics will change when going from 2D confinement in MLs to 0D confinement on defects in the matrix, that are interesting as emitters for quantum optics. We will measure the polarization and the charge transfer dynamics in Van der Waals heterostructures. Here we aim to switch between optical carrier recombination that is either direct or indirect in real- or momentum-space, respectively, by exploiting the band alignment of different TMDC materials.