Detection Of Defects On The Surface Of A Semiconductor By Terahertz Surface Plasmon Polaritons

Surface plasmon polaritons(SPPs), also known as surface wave, is now recognized as the collective charge oscillations at an interface formed by a dielectric medium and a conductive medium. SPPs have been researched intensively for its potential application value in electronics and photonics. THz has been regarded as one of the most important science and technology issues in this century, but little research is focus on THz SPPs. In this thesis we propose a new method for detecting small defects on the surface of a semiconductor by analyzing the transmission spectrum of terahertz surface plasmon polaritons.First, the dispersion relation of THz SPPs and the propagation properties of THz SPPs are introduced. After that, the characteristic length of THz SPPs is calculated. The results show that SPPs at a frequency approaching the plasma frequency of the semiconductor have a relatively short propagation length and a long decay length in semiconductor. And the influence of the propagation characteristics of the SPPs after passing through the defects is discussed. The larger defect on semiconductor surface, the SPPs’ propagation length is shorter. When the defect is beneath the surface of the semiconductor, the SPPs’ propagation length is getting increased as the size of the defect increased.Based on the theory research, the field distributions caused by detecting different sized defects is simulated with finite element method. The computational results show that the system can be used to detect a defect semiconductor surface with a minimum size of the 11 μm, while it also could be capable to detecing a defect beneath the semiconductor surface with a size of 16 μm.Experimentally, using a terahertz time domain spectrometer(THz-TDS), we measure the transmission spectrum of terahertz surface plasmon polaritons passing through particles on the surface of an intrinsic In Sb wafer. Our results show that the measured temporal waveform and frequency spectra are distinctly changed due to the presence of the particles, thereby confirming the effectiveness of this method for detecting defects. For increased detection efficiency, the frequency of the surface plasmon polaritons has to be slightly lower than the plasma frequency of the semiconductor. In comparison with traditional methods, our approach offers the merits of detecting both on-surface as well as sub-surface defects, which is critical in monitoring the quality of semiconductor wafers.