Raman microspectroscopy is a frequently used method for material characterization.
It allows the determination of chemical, magnetic, thermal and electrical properties and provides information on the structure of gratings in solids. More and more, Raman measurements are used at various temperatures to gain data like for example the anharmonicity of photons which has a great impact on the thermal properties of a material.
In regular setups it is difficult and sometimes downright impossible to measure samples with a weak signal at low temperatures. To date, there are two standard setups: A cryo objective with short working distance is installed in the specimen chamber. This provides a high numeric aperture, but may cause problems both due to the objective’s temperature-dependency and regarding alignment. Installing the objective outside the specimen chamber increases the working distance, which leads to decreased collection efficiency.
The work group around Prof. Kenneth Burch at the Boston College together with Montana Instruments have now developed a brand-new setup approach for low-temperature Raman measurements. This setup combines high collection efficiency with second-to-none temperature accuracy and outstanding thermal stability .
The setup uses the actively temperature-controlled objective mount of the Cryostation (Montana Instruments) to install a 0.9 NA objective near the sample (fig. 1).
The objective is held at room temperature with a PID control. The sample mount has a local heating unit that allows quick and precise temperature changes; for example from 4 K to 350 K in only 5 minutes. The excellent temperature accuracy of a few millikelvin reduces the sample mount drift to a minimum. In combination with the high mechanical stability, this setup facilitates measurements with long integration times at a spectral resolution of 30 cm-1.
As the first measurement, the topological insulator Bi2Se3 and a V2O3 sample were analyzed. The Bi2Se3 single crystal has a very low thermal conductivity and thus very good thermoelectric properties, which makes it very interesting for a variety of different applications. To avoid heating, the laser intensity must be kept low for Raman measurements. However, this also means the Raman signal is very weak. In this experiment, a 532 nm laser with only 40 µm was used and focused on a spot size of 1 µm. The measurement demonstrates the temperature-dependency of Raman lines with an outstanding S/N ratio (fig. 2).
The V2O3 thin film is of interest because of its phase transition from metal to insulator. This transition has already been examined with a number of measurement methods, however, its origin is still not completely explained.
Raman spectroscopy at variable temperatures provides information on phonon-phonon interactions. Due to the weak signal, domains of nanometer scale and the small temperature range in which the transition takes place, Raman measurements on this material are exceptionally difficult. V2O3 shows hysteresis effects and lays out extra difficult requirements for temperature control. Figure 3 shows the measurement of a 200 nm thick V2O3 film. Each spectrum consists of 2 - 4 measurements of 15 minutes. The mechanical and thermal drift of the sample have to be very low because of the long duration of the measurement. The temperature-dependent Raman spectra can clearly be seen and show the well-known first-order structural phase transition .
The metallic phase shows a wide, asymmetric peak at 236 cm-1.
The insulating phase is depicted by four distinct modes in the range 324.9 cm-1 - 340.4 cm-1.
Both measurements demonstrate the enormous power of this setup in regards to s/n ratio, temperature control and drift.
 Low vibration high numerical aperture automated variable temperature Raman microscope
Yao Tian et al., Review of Scientific Instruments 87, 043105 (2016)
 Effect of disorder on the metal-insulator transition of vanadium oxides: Local versus global effects
Juan Gabriel Ramirez et al., Phys. Rev. B 91, 205123 (2015)