Effects of surface roughness on the extraction of spectral parameters in transmissive terahertz time-domain spectroscopy (THz-TDS)

Yang Yu,Zhao-hui Zhang,Xiao-yan Zhao,Tian-yao Zhang
DOI: https://doi.org/10.1117/12.2604346
2021-09-09
Abstract:Extracting spectral parameters of materials is an essential application of terahertz spectroscopy technology, based on the characteristics of coherent detection. However, the commonly used algorithm for extracting spectral parameters requires a parallel and smooth surface as a prerequisite, and the surface roughness will affect the extraction result. Nevertheless, the effect degree and the mechanism are not precise before. Firstly, in this research, optical parameters of samples with different roughness extracted by the algorithm are displayed. Therefore, the effect degree of different roughness on the optical extraction algorithm is clarified. After that, the mechanism of the influence is analyzed through the method of microelement modeling. As a result, it shows that when the sample surface is slightly rough (roughness<60m), it will not significantly impact the extraction results. The shape of the optical curve will not be significantly distorted. As the roughness increases, the change of the statistical distribution of the Fresnel coefficient and the phase change are the reason for the attenuation of the terahertz wave amplitude and the decrease of the extraction accuracy.
What problem does this paper attempt to address?