Terahertz Time-domain Spectroscopy Data for Submission to Review of Scientific Instruments (2026)

Fig. 3(a) was created by getting the raw data for 'Time (ps)' and 'Lockin X' followed by baselining the X values. (Values taken from the non fft file) The Time values were adjusted by subtracting the time at which the maximum 'Lockin X' value occurred so this value would correspond to $\tau$ = 0. This offset $\tau$ value was used to offset all the Times in Fig. 4(a).Fig. 3(c) was created by plotting abs(Y) vs X from the fft file and a semilog scale was added to the vertical axis. This fft file is half the length of the 'Time (ps)' array length due to windowing out the second pulse by adding [:500] to the end of the data set in python.Fig. 4(a) was created by getting the raw data for 'Time (ps)' and 'Lockin X'. Only 5/6 the data set was used (adding [:833] to the end of the data set in python to create this window), followed by baselining the X values. (Values taken from the non fft file). An offset was added to the 'Lockin X' values as T increased to separate the data sets. A unique symbol and color was assigned to the end of each data set to identify the set in Fig. 4(d)Fig. 4(b) was created by plotting abs(Y) vs X from the fft file for each magnetic field. These fft files are half the length of the 'Time (ps)' array length due to windowing out the second pulse by adding [:500] to the end of the data sets in python. The colors and symbols here are the same as those in Fig. 4(a). The location of the symbols on each absorption region correspond to the minimum location determined by the fitting model described below.Fig. 4(c) was created by windowing a portion of the frequency plot for 4 T, where the absorption occurred. This region was fit using a combination of Gaussian model and a linear model from LMFit. Fig. 4(d) was created by finding the minimum location of each absorption region's fitted model, as done for Fig. 4(c). The minimum values were plotted and fit with a linear model from LMFit.