2D THz and GHz signature for identification of explosive on reflected THz signal

Abstract

The method of THz spectrum dynamics analysis (SDA - Spectral dynamics analysis - method) is applied for the detection and identification of substances by using the signal reflected from sample. It allows to obtain the spectrogram - composite Fourier spectrum dynamics - of the signal and to analyze the dynamics of many spectral lines simultaneously, even if the measurements are made on short time interval (less than 50 ps). The efficiency of the SDA method used for longer time intervals (more than 100 ps) is discussed also. The Fourier-Gabor sliding window method is used for obtaining the spectrogram. We consider the examples of finding the pure RDX and harmless materials (L-Tartaric Acid, Sucrose, PTFE) or their mixture in pellets by using a THz pulse reflected from them. A THz pulse with a few cycles falls on the sample and reflects from it. The receiver makes the discrete measurements of electric field strength of signal reflected from the sample. To restore the signal to the required accuracy the SVD - Single Value Decomposition - technique is used. Our investigations show that the spectrograms and dynamics of several spectral lines of the THz pulse reflected differ from the corresponding spectrograms and dynamics of spectral lines for the reference pulse under certain conditions and hence it is possible to detect the presence of the material in the sample of interest. The comparison of the Fourier spectrum of the substance with the corresponding spectrum calculated on the base of using an autocorrelation function for obtaining the spectrum shows that the AC-spectrum gives us essential less information about substance. From our consideration follows that in some cases the analysis of reflected signal on the short time interval (less than 50 ps) is insufficient for the reliable identification. It is necessary to analyze the response on the long time interval (about 300 - 400 ps). The analysis of spectrogram and spectral lines dynamics on the long time intervals gives the additional information about registering sub-pulses, which are appeared due to multi-reflection of THz wave from the boundaries of sample and from the reflecting elements of set-up, and relaxation times of excited energy levels of molecules. Hence, the reliability of identification increases.