Photoconductive receivers at 1030 nm for high average power pulsed THz detection

This dataset is accompanying the paper "Photoconductive receivers at 1030 nm for high average power pulsed THz detection"

General data acquisition:

THz is generated with the tilted pulse front approach in lithium niobate at room temperature. The pump power is controlled by a motorized lambda/2-waveplate (PI DT-80) in connection with a thin-film polarizer and calibrated to a THz power meter (Ophir 3A-P-THz). The computer-controlled rotation stage allows to sweep the THz power in a reliable and reproduceable way. The probe beam is guided over an oscillating delay line, having a delay range of approximately 15 ps and a shaking frequency of 20 Hz, leading to 4800 THz traces in over 2 min measurement time. In the path of the probe beam is also a motorized rotation stage and a polarizer positioned, to control the laser probe power in the same way as the THz pump power.

The THz is received by a ErAs:InAlGaAs photoconductive antenna (PCA) developed at TU Darmstadt, which is optimized for 1030 nm. The THz receiving side has a silicon lens, concentrating the THz radiation on an H-dipole antenna with a center dipole length of 25 µm and a photoconductive gap of 5 µm. On the backside, the probe beam is coupled with a microscope objective for the near infrared range (Mitutoyo M Plan Apo NIR 20X). The small current of the PCA receiver is converted to a useable voltage range with a transimpedance amplifier (Femto DLPCA-200) with a gain of 10^7 V/A.

For each combination of THz pump power and laser probe power (sampling beam), a measurement file with 2 min recording length and a sampling rate of 200 kSa/s is recorded with the DAQ (Dewesoft Sirius Mini). The exported file-format HDF-5.

HDF-5 is an efficient (binary), cross-platform data format and can be read easily by i.e. Python or Matlab. The graphical user interface “HDFView” can be downloaded for free (after registration) from https://www.hdfgroup.org/downloads/hdfview/. It allows to explore the folder structure of an HDF-5 file but is not necessary when using Python or Matlab.

The structure of a single HDF-5 from the uploaded raw data is

A folder called "AI", standing for analog input

The dataset "AI 1" stands for the first analog channel, containing the position of the scanning delay line. The voltage can be converted to “THz time” with a conversion factor, where 20 V correspond to 15 ps delay.

The dataset “AI 2”, which is voltage signal from the transimpedance amplifier and proportional to the THz electric field.

There are additional attributes in the root-folder of the HDF-5 file.

The processed data (used in the figures) can be found in the files labeled Fig.X.h5, except for Fig.7, which only contains a single curve and is a .txt file.