Спектр излучения и стабильность двух типов электронно-дырочной жидкости в мелких Si/Si-=SUB=-1-x-=/SUB=-Ge-=SUB=-x-=/SUB=-Si квантовых ямах

The dynamics of the phase transition from an electron-hole plasma to an exciton gas is studied during pulsed excitation of heterostructures with Si1 − x Ge x /Si quantum wells. The scenario of the phase transition is shown to depend radically on the germanium content in the Si1 − x Ge x layer. The electron-hole system decomposes into a rarefied exciton and a dense plasma phases for quantum wells with a germanium content x = 3.5% in the time range 100–500 ns after an excitation pulse. In this case, the electron-hole plasma existing in quantum wells has all signs of an electron-hole liquid. A qualitatively different picture of the phase transition is observed for quantum wells with x = 9.5%, where no separation into phases with different electronic spectra is detected. The carrier recombination in the electron-hole plasma leads a gradual weakening of screening and the appearance of exciton states. For a germanium content of 5–7%, the scenario of the phase transition is complex: 20–250 ns after an excitation pulse, the properties of the electron-hole system are described in terms of a homogeneous electron-hole plasma, whereas its separation into an electron-hole liquid and an exciton gas is detected after 350 ns. It is shown that, for the electron-hole liquid to exist in quantum wells with x = 5–7% Ge, the exciton gas should have a substantially higher density than in quantum wells with x = 3.5% Ge. This finding agrees with a decrease in the depth of the local minimum of the electron-hole plasma energy with increasing germanium concentration in the SiGe layer. An increase in the density of the exciton gas coexisting with the electron-hole liquid is shown to enhance the role of multiparticle states, which are likely to be represented by trions T + and biexcitons, in the exciton gas.

We discuss the mechanism of efficient photoluminescence (PL) from Si quantum dots and wells. Luminescence properties of SiO2-capped Si nanocrystals are different from those of H-terminated Si nanocrystals, but are very similar to those of Si quantum wells sandwiched by SiO2 layers. The size-dependence of PL properties and resonantly excited PL spectra of SiCb-capped Si dots and wells indicate that excitons are localized near the interface between the crystalline Si core and surface oxide layer, and the strong coupling of electronic and vibrational excitations causes the broad PL spectrum. The exciton localization plays an essential role in the efficient luminescence process in nanoscale Si/SiO2 systems.

Several models of oxygenated and hydrogenated surfaces of Si quantum wells and Si nanocrystals of variable shapes have been constructed in order to assess curvature effects on energy gaps due to the three Si-suboxides. Si-suboxides in partially oxydized models of nanocrystals of spherical shapes (or quantum dots) are shown to reduce energy gaps compared to the hydrogenated nanocrystals, consistent with previous results in the literature. This trend is shown to be reversed when planar interfaces are formed in Si-NC inside SiO2 thin films, as in Si quantum wells. At planar interfaces (or surfaces) the electronic charge density is shown to become extended and to distribute among the Si-suboxides, thus generating along these planar interfaces extended, or delocalized, states. This delocalization of the electronic states then increase the energy gap compared to equivalent hydrogenated interfaces (or surfaces). Determination of geometric effects of curvature on the band gaps are based on density functional theory (DFT) using the real-space numerical approach implemented in the ``Pseudopotential Algorithm for Real-Space Electronic Structure'' (PARSEC) program. A method for measuring interface effects due to Si-suboxides in Si-NC in SiO2 is also suggested: by comparing photoluminescence (PL) between as-grown and annealed Si-NC samples. The DFT calculations suggest that blueshift of more than 0.2 eV of the PL should be observed in as-grown samples having Si-suboxides at their planar interfaces, in comparison to annealed samples above 950 degree Celcius that have fewer or no Si-suboxides once annealed, providing that the bulk of Si-NC remains unaltered.

Cyclotron resonance of two-dimensional electrons in a Si quantum well

Magnetron sputter epitaxy of [formula omitted] heterostructures

Many-body carrier interaction effects shift the ground- and excited-state subband energy levels of quantum wells. We derive carrier-carrier Coulomb exchange and direct interaction energies as a function of carrier density, quantum well composition, and well width. For values of these parameters of physical interest, the excited-state exchange energy is small compared to the ground-state exchange energy. Our expressions for the ground- and excited-state direct interactions yield considerably smaller energy values than those reported by previous investigators. This is due to the fact that in our derivations we differentiate between quantum well width, carrier doping width, and ground- and excited-state wave function effective confinement lengths. Direct interaction energies increase linearly with carrier density, and for large carrier densities substantially negate exchange interaction effects. We illustrate carrier-carrier interaction effects for the heavy-hole subband of electrically neutral, p-type, isolated strained-layer ${\mathrm{Ge}}_{\mathrm{x}}$ ${\mathrm{Si}}_{1\mathrm{\ensuremath{-}}\mathrm{x}}$ /Si quantum wells.

Strong electron intersubband infrared absorption is observed for Sb δ-doped Si/Si1−xGex multiple quantum well structures grown on (110) Si substrates. The intersubband absorption is shown to be allowed for both the optical field components perpendicular and parallel to the quantum wells due to the tilted ellipsoidal of constant energy surfaces. About 90% infrared absorption is measured by a Fourier transform infrared spectrometer using a waveguide structure with 10 internal reflections. For various samples used in experiments, absorption peaks ranging from 4.9 to 5.8 μm are observed. The peak energy is shown to be tunable by changing the Ge composition in the Si1−xGex barriers and the doping concentration in the Si quantum wells.

We report the first observation of photoluminescence from electron-hole plasmas in Si/Si0.8Ge0.2/Si quantum wells. While at liquid helium temperature, luminescence due to shallow bound excitons is observed. At 77 K electron-hole plasma (EHP) luminescence dominates the spectra over a wide range of pump powers. Convolution of the occupied electron and hole densities of states gives an excellent fit to the photoluminescence line shape. A band-gap reduction of up to 15 meV at high carrier densities is observed for wide quantum wells, but no such shift is detected for narrow quantum wells.

Non-linear decay of photoluminescence polarization was observed in strained Si/sub 1-x/Ge/sub x//Si quantum wells under longitudinal electric field at low temperature, which is interpreted in terms of delayed field ionization of excitons.

A single-frequency admittance spectroscopy technique is presented to determine the band discontinuity of the Si/Si1−xGex/Si single quantum well. The accuracy of determining the activation energy for thermionic emission of holes over the Si barrier is improved as compared with the multiple-frequency admittance spectroscopy measurement. To derive the band offset from the activation energy, the Fermi energy with respect to the band edge is calculated more exactly by solving the Poisson equation which accounts for the carrier transfer at the heterointerface. The accuracy of this technique is verified by the reconstruction of the temperature dependence of capacitance and conductance by the computer simulation.

Photoluminescence (PL) of Si1-x Gex /Si quantum wells with abrupt interfaces formed by segregant-assisted growth (SAG) is reported. It was found that SAG-grown samples show spectral red-shift compared to those grown by conventional solid-source molecular beam epitaxy due to the suppression of Ge surface segregation during growth. Crystal quality of SAG-grown samples is discussed based on the excitation variation of PL intensity.

Si 1− xGe x/Si valence band offset determination using current–voltage characteristics

The admittance spectroscopy technique has been used to study the hole confinement in Si/Si1-xGex/Si and boron highly doped superthin Si quantum wells. Based on the carrier thermal emission medel of the carriers in the well, the value of activation energy derived from the admittance spectra was considered as the distance from the position of the heavy hole ground state in the well to the top of it . For Si/Si1-xGex/Si quantum wells we found the value of activation energy increased with increasing annealing time at 800℃. The phenomena could be explained the annealing-induced interdiffusion of Si and Ge, the interface broadening, the weakening of quantum confinement, the decrease of the position of the heavy hole ground state, and so the increase of the activation energy. At the annealing temperature of 900℃, we only see the decrease of the value of activation energy with increasing annealing time. For boron highly doped superthin Si quantum wells due to quantum confinement effect, we observed the activation energy are different for samples with the same doping density but different well widths.

The determination of valence band offsets Qv by fitting the optical transmission spectra for AlxGa1−xAs/GaAs and InyGa1−yAs/GaAs multiple quantum wells (MQWs) is discussed. The valence band offset Qv is defined as Qv =ΔEv/(ΔEv+ΔEc), where ΔEv and ΔEc are the valence and conduction band discontinuities at the heterojunction. For AlxGa1−xAs/GaAs (Type I for both heavy holes and light holes, i.e., electrons, heavy holes, and light holes are located in the same layers) the energy separation ΔE between the first light and heavy hole subbands in the valence band is used to determine the valence-band offset Qv. In order to improve the precision, MQWs with narrow well widths and higher Al mole fraction x should be used. Further, the inclusion of coupling between the valence and conduction subbands in Kane’s three-band model is also essential. However, Qv is still not determined uniquely by the above energy difference ΔE. Different combinations of Qv and heavy hole effective masses mhh in GaAs can lead to the same ΔE. The forbidden transition E13h in a narrow well can be used to identify the actual values of mhh and Qv. [The designation Enmh(l) refers to the energy separation of the nth conduction subband and mth heavy (light) hole subband.] Following such a procedure, we find that Qv =0.33 and mhh =0.34 m0 for AlxGa1−xAs/GaAs MQWs (x<0.37). For InyGa1−yAs MQWs (Type I for heavy holes and Type II for light holes, i.e., both electrons and heavy holes are located in the same layers, but the light holes are in different layers), the transition energies between the conduction subbands and the light hole valence subbands are much better parameters to determine the band offsets in comparison with the transitions involving the heavy holes. Both the optical measurements and the calculations including the strain effect show that E11l between the first conduction subband and the first light hole valence subband in the MQWs with wider well widths is a sensitive enough parameter to determine the band offsets. By comparing with the optical measurements we obtained Qvh =0.30 and Qvl =−0.23 (Type II) (for 0.13<y<0.19) for the heavy hole and light hole valence band offsets, respectively.

The density functional theory was used to calculate the energy of a quasi-two-dimensional electron-hole liquid (EHL) in Si/Si1-xGex/Si quantum wells. We have performed a numerical solution of the three one-dimensional nonlinear Schrödinger equations for electrons, heavy and light holes. It is shown that EHL contains heavy and light holes for small x. With an increase in x, a transition to a state with one type of hole occurs, and the density of electron-hole pairs decreases by more than three times. The results of the calculations are in good agreement with the experimental results.