Abstract In this thesis, We have probed the semiconductor microstructure by photoreflectance. A laser beam was used to modulate the dielectric constant and the parameter £GR and R were probed with a white light source. First, we perform measurement and simulation for a single quantum well structure grown by molecular beam epitaxy(MBE).Three subband transition can be found in our simulation, and a 2D third derivative model was used to fit experiment data. For the absorption coefficient and refractive index, we perform a simulation with a ideal model ( i.e. FWHM=0). In the simulation, we can understand that the photoreflectance spectrum is composed with symmetric and anti-symmetric line shapes in our fitting model. In addition, we can observe FKO effect in this sample and obtain electric field from FKO data. For asymmetric InGaAs/InGaAlAs multiple-quantum wells, we perform measurement, simulation and curve fitting for both modulation doping and non-modulation doping samples. We can observe many high order sub-band transitions. Besides, some quantum well signals for the modulation doped samples were not observed due to the cancellation of build-in electric field.

Photoreflectance (PR) is a sensitive technique to measure semiconductor properties at room temperature due to its derivative nature. The normalized

Semiconductor band-gap energy can be measured by using photoreflectance (PR) spectroscopy. It used a pump-beam and a probe-beam and measure modulated reflectance (DR) of the probe-beam by chopping on and off the pump-beam. The photon-energy of the pump-beam has to be greater than the band-gap energy of the sample so that electron-hole pair can be produced in the sample. The electron-hole pairs are separated by built-in electric field (Fbi) of the sample and thus reduce strength of Fbi. In this work the Hg lamp was used for the pump beam to observe E_1 and E_1+∆E_1 transitions of GaAs, and its photon energy is greater than E_1 and E_1+∆E_1 bandgaps. We also used the purple, green and red laser to observe GaAs, and their photon energy are greater than E_0 but smaller than E_1gaps. Finally, we will compare the amplitude of the PR spectra with the strength of the Fbi reduced by the pump-beam.

InN is a semiconductor material of vary high electron mobility, so InN have potential for high speed electronic device. But the bandgap is not sure. We use photoreflectance spectroscopy to investigate bandgap of InN at different temperature. We use third derivative reflectance formula of low field to fit experimental data and appraisal the type of electron transition.

The excitonic transitions of a c-oriented wurtzite InN thin film, grown on sapphire substrate by plasma-assisted molecular beam epitaxy, were studied by photoreflectance (PR) measurement from 10 to 120 K. The energies of the observed features have a tendency to decrease with increasing temperature. They are assigned to excitonic transition rather than band-to-band transitions because the features can be observed only below 110 K. The justifications of such assignments are discussed in the context of binding energies of the excitons. The PR spectra of various power of pumping beam (Ppu) were also measured. The energies of the observed features become red-shifted with decreasing Ppu. This is consistent with assignment of the excitonic transition.

We have studies the II-VI ternary compound semiconductor ZnSe grown by molecular beam epitaxy ¡]MBE¡^Method. The modulation spectroscopy was used to study ZnSe. ZnSe epilayer was grown on GaAs substrate. The lattice mismatch¡]0.27 ¢H¡^between GaAs and ZnSe create a strain at the GaAs/ZnSe interface. The strain will remove the degeneracy of heavy and light holes to conduction band transition energies. We use the photoreflectance to measure the energy of different thickness ZnSe epilayers at low temperature. It was found that as the epilayer thickness becomes larger, the£GE will become smaller. We have also analyzed the energy of different temperatures in terms of Varshni relation, and the temperature dependence of the broadening parameters.

Photoreflectance (PR) spectra of a GaN thin film and an AlGaN/GaN heterostructure were measured by using a HeCd laser or a mercury lamp as a pump beam. The wavelengths (£f) of the HeCd laser and the mercury lamp are 325 nm and 253.7 nm, respectively. The energy of the HeCd laser is lower than band-gap energy of AlxGa1-xN (x > 0.2) so that electron-hole pairs cannot be generated in the AlGaN layer. Hence, the PR of the AlGaN was measured by using Argon ion laser (£f=300 nm) or quadrupled Nd:YAG (£f=266 nm) rather than HeCd laser in the previous works. In this work, the mercury lamp (£f=254 nm)was used as the pump beam. The problem with using the mercury lamp as the pump beam is because it is a diffused source so that it cannot be focused to a small spot. Nevertheless, defocused pump and probe beams were used in the PR measurement to improve signal to noise ratio. Hence, the diffused property of the mercury lamp is not a hindrance to the PR measurements.

Neste trabalho fazemos um estudo dos campos elétricos existentes em super-redes δ-Si:GaAs As técnicas de espectroscopia Raman e fotorefletancia foram utilizadas para fins de comparação do campo elétrico medido indiretamente (Raman) e diretamente (fotorefletancia). Com o objetivo de eliminar os campos elétricos através do efeito fotovoltaico, foram feitas medidas de fotorefletancia para altas intensidades do feixe de prova, porem apenas reduções de ate 30% no valor destes campos foram obtidas. Os tempos de resposta dos portadores responsáveis pela modulação dos campos elétricos foram inferidos através de medidas de fotorefletancia sensível a fase e estão em bom acordo com os tempos medidos através de uma técnica de fotorefletancia desenvolvida em nosso laboratório / In the present work we report a study of surface and interface electric fields in δ-Si:GaAs superlattices. Techniques such as photoreflectance (PR) and Raman spectroscopy were used to determine surface electric fields in a direct and indirect way, respectively. The results agree with those expected by Fermi level pinning at the surface. In order to eliminate electric fields via photovoltaic effect, photoreflectance spectra were obtained at high probe intensities, but reductions of just 30% were achieved. In time domain, we show that carriers response time responsible for modulation of electric field can be obtained by means of phase sensitive PR, and are in good agreement with those measured in a time resolved PR technique, developed in our laboratory

Neste trabalho fazemos um estudo dos campos elétricos existentes em super-redes δ-Si:GaAs As técnicas de espectroscopia Raman e fotorefletancia foram utilizadas para fins de comparação do campo elétrico medido indiretamente (Raman) e diretamente (fotorefletancia). Com o objetivo de eliminar os campos elétricos através do efeito fotovoltaico, foram feitas medidas de fotorefletancia para altas intensidades do feixe de prova, porem apenas reduções de ate 30% no valor destes campos foram obtidas. Os tempos de resposta dos portadores responsáveis pela modulação dos campos elétricos foram inferidos através de medidas de fotorefletancia sensível a fase e estão em bom acordo com os tempos medidos através de uma técnica de fotorefletancia desenvolvida em nosso laboratório / In the present work we report a study of surface and interface electric fields in δ-Si:GaAs superlattices. Techniques such as photoreflectance (PR) and Raman spectroscopy were used to determine surface electric fields in a direct and indirect way, respectively. The results agree with those expected by Fermi level pinning at the surface. In order to eliminate electric fields via photovoltaic effect, photoreflectance spectra were obtained at high probe intensities, but reductions of just 30% were achieved. In time domain, we show that carriers response time responsible for modulation of electric field can be obtained by means of phase sensitive PR, and are in good agreement with those measured in a time resolved PR technique, developed in our laboratory

The photoreflectance(PR) and Electroreflectance(ER) of surface-intrinsic-n+ type doped GaAs exhibit many Franz-Keldysh oscillations (FKOs), which enable the electric field (F) to be determined from the technique of the fast Fourier transform (FFT). It is known that F's determined from PR are subjected to photovoltaic effect, but it is difficult to estimate the strength of modulating field (dF) of the pump beam in the PR measurements . Alperovich et. al. have used imaginary part of FFT to determine the strengths of dF's in the ER measurements [V. L. Alperovich, et. al. Appl. Phys . Lett. 71, 2788 (1997)]. Here, we will apply this method to the PR measurements. The dF's thus obtained will be compared with those deduced from photo-voltage measurements. The result shows that the method of Alperovich's can be used to determine the strength of dF in the PR measurements.