BNL Home

research: experimental validation

illustration

Optics

experiments: Christopher Homes
material growth: Cedomir Petrovic
theory: Sangkook Choi, Walber de Brito

In our laboratory, we reveal the complex optical properties of solids. The temperature-dependence of the reflectance is measured at a near-normal angle of incidence over a wide frequency range (terahertz to the ultraviolet region) on a Bruker Vertex 80v. The optical properties can then be calculated using a Kramers-Kronig analysis. In anisotropic single crystals, polarizers may be used to study the optical response along the principle optical axes. The measurements are performed by attaching the samples and reference mirrors to optically-black cones and then fitting them to the bottom of a compact open-flow cryostat that has a rotational degree of freedom - at each temperature, the reflectance of the sample is compared to that of a reference mirror. A thin layer of gold (silver, or aluminium) is evaporated onto the sample after that, and the measurements are repeated. When the ratio of these measurements is taken, the reference mirror is eliminated, leaving just the ratio of the original sample with that of the (gold-)coated sample. As the optical properties of gold (silver, or aluminium) are well known, the absolute reflectance of the sample may then be determined. We typically use an overfilling technique that compensates for irregular surfaces, misalignments, etc. It is generally considered to be the most accurate method for determining the absolute reflectance of a material [Applied Optics 32, 2976 (1993)].

photo of spectrometer

Spectrometer and reflectance unit.  The figure on the left shows the compact open flow cryostat attached to an external sample chamber on the Bruker Vertex 80v. This arrangement has a rotational degree of freedom that allows the sample to be interchanged with a reference mirror.  The high vacuum in the cryostat is separated from the rough vacuum in the spectrometer by window holder that is translatable under vacuum, allowing different optical windows to be used for different spectral regions. The optical arrangement for the reflectance also enables the use of internal liquid-nitrogen cooled detectors, as well as external liquid-helium cooled detectors. The figure to the right shows the tail of the cryostat with the radiation shield removed. The sample and reference mirror are mounted on optically-black cones that are in turn mounted onto the copper block at the bottom of the cryostat, allowing an over-filling technique to be used.

For Comscope we provide experimental validation of the Comsuite codes by measuring the real and imaginary parts of the optical conductivity and comparing them to the theoretical calculations. This work is done in close collaboration with sample growers and theorists.

A specific target was the complex optical properties of the colossal thermopower material iron antimonide, FeSb2. The surprising aspect of this work is that we initially thought that the optical properties of FeSb2 were a "solved problem", and that our optical measurements would only add more details to published work. However, they have fundamentally changed the way we view this material. Prior to our measurements, we followed the prevailing view that the extraordinarily high thermoelectric power factor observed in this material at low temperature was possibly either due to electronic correlations or the so-called phonon drag effect. The optical properties of a material provide information on both the (electronic) optical conductivity, as well as the infrared-active lattice vibrations at the center of the Brillouin zone and are therefore well-suited to investigate this controversy. But surprisingly, we found a completely new physical aspect of the material: we observed, for the first time, one-dimensional semiconducting behavior along the b axis at low temperature - a property that is supported by theoretical first-principle calculations. [C.C. Homes et al., submitted for publication]. In addition, we revealed an anisotropic optical conductivity at room temperature, which develops an optical gap with decreasing temperature and several prominent phonon anomalies along the a axis below about 100 K, the temperature at which this material undergoes a metal-to-insulator transition. These anomalies suggest a weak structural distortion or a phase transition. When taken together, the reduced dimensionality and the phonon anomalies lead to a new mechanism that is likely responsible for the extremely large thermoelectric power in FeSb2: a strong coupling of the electronic and lattice degrees of freedom.

graph

The reflectance of FeSb2 at 75 K for light polarized along the b axis. The red curve shows the reflectance of the sample with respect to the reference mirror and the orange curve the reflectance of the gold-coated sample with respect to the reference mirror.  In addition to the features associated with the sample, there is a considerable amount of structure from the optical window in the cryostat, as well as the windows and filters in the liquid-helium cooled bolometer used in this experiment, that arise due to the different optical paths for the sample and the reference. However, when the ratio of these two curves is taken the reference mirror ratios out and the extrinsic features are removed, yielding the intrinsic reflectance of the sample (with respect to gold), shown by the the indigo line.

Related Publications

Unusual electronic and vibrational properties in the colossal thermopower material FeSb2.
C. C. Homes, Q. Du, C. Petrovic, W. H. Brito, S. Choi, and G. Kotliar,
Scientific Reports 8: 11692 (2018)

Optical properties of the perfectly compensated semimetal WTe2.
C. C. Homes,  M. N. Ali, and R. J. Cava,
Phys. Rev. B 92, 161109(R) (2015)

Silicon beam splitter for far-infrared and terahertz spectroscopy.
C.C. Homes, G. L. Carr, R. P. S. M. Lobo, J. D. LaVeigne, and D. B. Tanner,
Applied Optics 46, 7884 (2007)

Technique for measuring the reflectance of irregular, submillimeter-sized samples.
Christopher C. Homes, M. Reedyk, D. A. Cradles, and T. Timusk,
Applied Optics 32, 2976 (1993)