Single layer MoS2 is an ideal material for the emerging field of "valleytronics" in which charge carrier momentum can be finely controlled by optical excitation. This system is also known to exhibit strong many-body interactions as observed by tightly bound excitons and trions. Here we report direct measurements of valley relaxation dynamics in single layer MoS2, by using ultrafast transient absorption spectroscopy. Our results show that strong Coulomb interactions significantly impact valley population dynamics. Initial excitation by circularly polarized light creates electron-hole pairs within the K-valley. These excitons coherently couple to dark intervalley excitonic states, which facilitate fast electron valley depolarization. Hole valley relaxation is delayed up to about 10 ps due to nondegeneracy of the valence band spin states. Intervalley biexciton formation reveals the hole valley relaxation dynamics. We observe that biexcitons form with more than an order of magnitude larger binding energy compared to conventional semiconductors. These measurements provide significant insight into valley specific processes in 2D semiconductors. Hence they could be used to suggest routes to design semiconducting materials that enable control of valley polarization.

The electron-phonon interaction in monolayer graphene is investigated using density-functional perturbation theory. The results indicate that the electron-phonon interaction strength is of comparable magnitude for all four in-plane phonon branches and must be considered simultaneously. Moreover, the calculated scattering rates suggest an acoustic-phonon contribution that is much weaker than previously thought, revealing an important role of optical phonons even at low energies. Accordingly it is predicted, in good agreement with a recent measurement, that the intrinsic mobility of graphene may be more than an order of magnitude larger than the already high values reported in suspended samples. DOI: 10.1103/PhysRevB.81.121412 PACS number͑s͒: 72.10.Di, 71.15.Mb, 72.80.Vp Graphene, a two-dimensional ͑2D͒ sheet of carbon atoms in a honeycomb lattice, continues to attract much attention due to its unique physical properties. Aside from a substantial academic interest resulting from the relativisticlike behavior of charge carriers, this material is considered very promising in device applications as it has an extremely high intrinsic mobility, even at room temperature. Although in realistic conditions ͑i.e., placed on a substrate͒ the mobility tends to decrease significantly due to the presence of additional scattering mechanisms at the interfaces, 1-3 much effort is currently being devoted to eliminate, or at least minimize, these effects which are detrimental to graphene transport characteristics. Therefore, it is crucial to develop an accurate knowledge of the electron-phonon scattering as it determines the ultimate limit of any electronic device performance. The strength of electron-phonon coupling is typically estimated using the deformation potential approximation ͑DPA͒; it has been applied for graphene by a number of authors. [4][5][6] When the corresponding deformation potential constant was estimated from the transport measurements, however, the results revealed a discrepancy that is too large to be ignored. 1,2,7 Moreover, a very recent observation of mobilities in excess of 10 7 cm 2 / V s at T Շ 50 K in the decoupled graphene 8 drastically departs from the conventionally accepted values, raising serious questions about the current understanding of the intrinsic transport characteristics of graphene. A detailed theoretical analysis of electron-phonon interaction beyond the DPA is clearly called for. In this work, we apply a first-principles approach based on density-functional theory ͑DFT͒ to calculate the electronphonon coupling strength in graphene. The obtained electron-scattering rates associated with all phonon modes are analyzed and the intrinsic resistivity and mobility of monolayer graphene are estimated as functions of temperature. The results clearly elucidate the role of different branches ͑particularly, the significance of optical phonons and intervalley scattering via acoustic phonons͒ as well as limitations of DPA. The obtained effective deformation potential constants suggest the possibil...

We measured the lifetime of optically created valley polarization in single layer WS 2 using transient absorption spectroscopy. The electron valley relaxation is very short (< 1ps). However the hole valley lifetime is at least two orders of magnitude longer and exhibits a temperature dependence that cannot be explained by single carrier spin/valley relaxation mechanisms. Our theoretical analysis suggests that a collective contribution of two potential processes may explain the valley relaxation in single layer WS 2 . One process involves direct scattering of excitons from K to K valleys with a spin flip-flop interaction. The other mechanism involves scattering through spin degenerate Γ valley. This second process is thermally activated with an Arrhenius behavior due to the energy barrier between Γ and K valleys.

A spin field effect transistor (FET) is proposed by utilizing a graphene layer as the channel. Similar to the conventional spin FETs, the device involves spin injection and spin detection by ferromagnetic source and drain. Due to the negligible spin-orbit coupling in the carbon based materials, spin manipulation in the channel is achieved via electrical control of the electron exchange interaction with a ferromagnetic gate dielectric. Numerical estimates indicate the feasibility of the concept if the bias can induce a change in the exchange interaction energy of the order of meV. When nanoribbons are used for a finite channel width, those with armchair-type edges can maintain the device stability against the thermal dispersion.

We suggest an approach to the problem of free electron spin evolution in a semiconductor with arbitrary anisotropy or quantum structure in a magnetic field. The developed approach utilizes quantum kinetic equations for average spin components. These equations represent the relaxation in terms of correlation functions for fluctuating effective fields responsible for spin relaxation. In a particular case when autocorrelation functions are dominant, the kinetic equations reduce to the Bloch equations. The developed formalism is applied to the problem of electron spin relaxation due to exchange scattering in a semimagnetic quantum well (QW) as well as to the spin relaxation in a QW due to Dyakonov-Perel mechanism. The results permit to separate the longitudinal T1 and transversal T2 relaxations times in a strong enough magnetic field and to trace the cases of undistinguished parameters T1 and T2 in zero and small magnetic fields. Some new predictions of the developed theory are discussed.