The manipulation of spin degree of freedom have been demonstrated in spin polarized electron plasma in a heterostructure by using exchange-interaction induced dynamic spin splitting rather than the Rashba and Dresselhaus types, as revealed by time resolved Kerr rotation. The measured spin splitting increases from 0.256meV to 0.559meV as the bias varies from -0.3V to -0.6V. Both the sign switch of Kerr signal and the phase reversal of Larmor precessions have been observed with biases, which all fit into the framework of exchange-interaction-induced spin splitting. The electrical control of it may provide a new effective scheme for manipulating spin-selected transport in spin FET-like devices.

By biasing a single barrier heterostructure with a 500nm-thick GaAs layer as the absorption layer, the spin dynamics for both of the first and second subband near the AlAs barrier are examined. We find that when simultaneously scanning the photon energy of both the probe and pump beams, a sign reversal of the Kerr rotation (KR) takes place as long as the probe photons break away the first subband, and start to probe the second subband. This novel feature, while stemming from the exchange interaction, has been used to unambiguously distinguish the different spin dynamics ( g E ) all display a sudden change, due to the "resonant" spin exchange coupling between two spin opposite bands.* To whom the correspondence should be addressed. E-mail: hzzheng@red.semi.ac.cnThe behaviors of spin coherence in bulk semiconductors and their low-dimensional quantum structures have been extensively studied [1] in order to make it feasible that the spin degree of freedom can be employed as an alternative carrier of information in the next generation of electronics. Most experimental investigations have focused on the dynamics of spin decoherence in ground states [2]. However, few have studied the spin coherence in excited states. It was theoretically predicted that due to strong inter-subband scattering, the spin decoherence rate of electrons in ground and excited subbands were almost identical, despite the large difference in the

The characteristics of a resonant cavity-enhanced InGaAs/GaAs quantum-dot n-i-n photodiode with only a bottom distributed Bragg reflector used as the cavity mirror, are reported. To suppress the dark current, an AlAs layer is inserted into the device structure as the blocking layer. It turns out that the structure still possesses the resonant coupling nature, and makes Rabi splitting discernible in the photoluminescence spectra. The measured responsivity spectrum of the photocurrent shows a peak at l ¼ 1030 nm, and increases rapidly as the bias voltage increases. A peak responsivity of 0.75 A/W, or equivalently an external quantum efficiency of 90.3%, is obtained atIntroduction: 1.06 mm is a specific wavelength used in Nd:YAG laser systems. These systems have been widely deployed in industrial manufacturing, medicine, remote sensing, space communication, etc. Unfortunately, 1.06 mm falls into the 'sensitivity valley' between the long-wavelength limit of the spectral response of Si-photodiodes (PDs) and the short-wavelength limit of the spectral response of InGaAs/InP-PDs. To achieve high sensitive detection at this wavelength, resonant-cavity-enhanced (RCE) PDs with quantum dots (QDs) as absorption layers were explored to reach a peak quantum efficiency of 65 -75% [1, 2]. A RCE-PD is usually constructed by placing multiple QD layers at the peak positions of electromagnetic fields in a resonant cavity, sandwiched between two distributed Bragg reflectors (DBRs) [3]. Since DBR mirrors consist of multiple pairs of AlAs/ GaAs quarter-wave stacks, they give the cavity a high finesse over several thousands. At the same time, DBR mirrors also bring difficulties in extracting photogenerated carriers from them [4].