Pulsed Terahertz Signal Reconstruction

• Published in: Journal of applied physics 2007-12, Vol.102 (11), p.113105-113105-8
• Main Authors: Fletcher, J. R., Swift, G. P., Dai, DeChang, Chamberlain, J. M., Upadhya, P. C.
• Format: Article
• Language: English
• ISSN: 0021-8979; 1089-7550
• DOI: 10.1063/1.2818361

A procedure is outlined which can be used to determine the response of an experimental sample to a single, simple broadband frequency pulse in terahertz frequency time domain spectroscopy (TDS). The advantage that accrues from this approach is that oscillations and spurious signals (arising from a variety of sources in the TDS system or from ambient water vapor) can be suppressed. In consequence, small signals (arising from the interaction of the radiation with the sample) can be more readily observed in the presence of noise. Procedures for choosing key parameters and methods for eliminating further artifacts are described. In particular, the use of input functions which are based on the binomial distribution is described. These binomial functions are used to unscramble the sample response to a simple pulse: they have sufficient flexibility to allow for variations in the spectra of different terahertz sources, some of which have low frequency as well as high frequency cutoffs. The signal processing procedure is validated by simple reflection and transmission experiments using a gap between polytetrafluoroethylene (PTFE) plates to mimic a void within a larger material. It is shown that a resolution of 100 μ m is easily achievable in reflection geometry after signal processing.