We are constructing a confocal microscope to study electron spin configurations in semiconductor nanostructures. For this, we are using an incident laser beam linearly polarized (630 nm) and the objective is to obtain a circularly polarized signal with the help of a quarter-wave plate, after interacting with the sample. However we have some troubles because the beamsplitters that we tested modify the signal polarization to an elliptical one (cubic beamsplitter) or make alignment very difficult (plate and wedged plate beamsplitters). Someone there has some suggestions on type of mechanical components, beamsplitters and/or microscope configurations to to get better results in this type of measure? I will be very grateful with the help.