Far-infrared magneto-spectroscopy of novel semiconductor materials in megagauss magnetic fields using quantum-cascade lasers – TeraMegaSpec

We are using quite unique installation capable do deliver magnetic fields well above 200 Tesla in a semi-destructive manner. The semi-destructive manner means that the coil explodes every time, while the sample space remains intact. On the other hand, the time-scale of the experiment is comparable with explosion of the coil and fixed then within several micro-seconds. This is very important limitation in terms of THz spectroscopy, especially for the detection side. The THz detectors are usually very slow. Another challenge is THz waveguiding since no metal can be put inside the megagauss magnetic coil.

We have constructed the prototype of the cyclotron resonance (CR) spectrometer probe. It incorporates the sample holder, installed in the center of the single-turn coil, and also the QCL just 10 cm above the sample. The THz radiation emitted by the QCL is guided with our custom-made waveguide to the sample space. The transmitted THz radiation through the sample is then guided further with the similar flexible waveguide to the externally placed Ge:Ga detector, mounted in the separate liquid-He cryostat. In December 2020, we have successfully detected the first radiation signal without any sample. However, the signal level remains relatively low. We are currently optimizing the setup in the following ways: the CR probe is redesigned to ensure better QCL cooling, the THz waveguide alignment is optimized in the sample space, and the Ge:Ga detector preamplifier is redesigned to reduce pick-up noise. The performance under non-destructive magnetic field pulses will be soon tested.

Due to Covid pandemic we are late with respect to the initial schedule, but still hope to deliver working spectrometer by the end of the project.

For the moment, we are in the technical development phase

Magneto-spectroscopy at terahertz (THz) frequencies in megagauss (MG) magnetic fields is an innovative approach for the investigation of novel semiconducting materials. Recently, this approach has become possible through both, the development of THz quantum-cascade lasers (QCLs) and the realization of pulsed magnetic fields with field strengths above 1 MG (100 T). The principal target of the project is to bring together leading scientists in these two fields. The implementation of this technique would open up cyclotron resonance spectroscopy to determine the carrier effective masses in materials with lower mobilities.

In this project, we will demonstrate a THz magneto-spectrometer operating under MG fields using compact lasers such as THz QCLs as the source of the far-infrared radiation, which can be installed in proximity to magnets reaching MG fields. Optimized QCLs for this application will be developed. As a proof of principle to demonstrate the applicability of magneto-spectroscopy, in particular of cyclotron resonance spectroscopy, this technique will be applied to study the effective electron mass in a set of MnSi samples with both, stoichiometric and nonstoichiometric composition.

We will directly probe the effective electron mass of MnSi using cyclotron resonance absorption measurements, where the absorption occurs when the frequency of the probing radiation matches the cyclotron frequency. The observability condition, i.e., the requirement that the electron completes one cyclotron orbit before a scattering event, in combination with the expected large effective electron mass and the low electron mobility in MnSi leads to an excitation wavelength in the THz region, while the magnetic field strengths enter the MG range. Cyclotron resonance spectroscopy in the THz range using MG fields has not been demonstrated so far.

For the experiments in MG magnetic fields, single-shot THz QCLs with a constant output power and emission frequency over about 20 microseconds are required. At the same time, the peak powers should exceed 10 mW. Already existing QCLs are a starting point to adapt them for transmission experiments in the THz spectral range in pulsed magnetic fields. The emission properties of these lasers have to be tested to determine how they can be operated under the unusual conditions during the generation of magnetic fields above 1 MG. The output power of the devices has to be optimized to generate a sufficiently large signal-to-noise ratio during the length of the magnetic field pulse. The required pulse length of 20 microseconds lies between the currently used typical pulse length of less than 1 microsecond for pulsed operation and infinitely long pulses for continuous-wave operation of THz QCLs. While we have already shown that such a pulse length is feasible, the required frequency and power stability have still to be demonstrated for these rather long pulses.