New material development is ongoing as more demanding applications are requiring more robust metal alloys. One of the challenges in designing systems is to find materials that will shrink when heated and expand upon cooling.

Dr. Daniel Mazzone, postdoctoral Fellow at Brookhaven National Laboratory in Upton, N.Y., says, “Traditionally, most materials will expand when heated and shrink when cooled. The source for this effect can be found at the atomic level. As the temperature expands, the vibrations of atoms get bigger, leading the material to expand.”

The opposite phenomenon is known as negative thermal expansion and has only been seen in a few materials. One classic case is that of water, which will expand and become less dense upon freezing to form ice. 

In searching for other materials, Mazzone has been working with lanthanide series elements that contain 4f atomic orbitals. He says, “The elements in this series exhibit very interesting properties, including negative thermal expansion, magnetism and superconductivity.”

The latter property was discussed in a previous TLT article,¹ which covered the synthesis of a new lanthanide superhydride with potential to function as a superconductor. Under extreme conditions, lanthanide superhydrides are formed from lanthanide and ammonia borane. Superconducting properties were detected at temperatures as high as 20 K below room temperature.

Negative thermal expansion was discovered in the compound samarium sulfide, which is derived from the lanthanum series element, samarium. This compound has been found to exist in two forms known as black-colored and golden samarium sulfide. The former is a semi-conductor that can be converted into the latter under hydrostatic pressure or through the introduction of the element yttrium.

Mazzone says, “Yttrium is a smaller atom in size than samarium and acts as an internal pressure device forcing samarium sulfide into a gold-colored material that is present in a metallic state. The resulting golden samarium sulfide exhibits negative thermal expansion, which is tunable based on the concentration of yttrium present.”

An important factor in this transition is the change in the valence state of samarium from mainly a 2+ in the black-colored sulfide to a mixed state between 2 and 3+ in the gold-colored sulfide. There is a direct relationship between the concentration of yttrium in golden samarium sulfide and the extent of the negative thermal expansion effect. As the percentage of yttrium increases, the negative thermal effect of the golden samarium sulfide decreases. 

Mazzone says, “Samarium sulfide was discovered approximately 50 years ago.” But no one has determined the microscopic mechanism behind its unusual thermal properties until now.

X-Ray spectroscopy and diffraction
Mazzone and his colleagues synthesized golden samarium sulfide, conducted X-ray spectroscopy and diffraction analyzes and used the Kondo-volume-collapse model to determine the mechanism for how this material undergoes negative thermal expansion. 

Golden samarium sulfide was prepared through the Bridgeman method. Mazzone says, “Samarium, sulfur and yttrium powders were added in an elongated quartz crucible and heated to an elevated temperature in an oven. Upon completion of the process, the crucible was then pulled slowly out of the melt, leading to the formation of the crystalline product.”

The researchers prepared four versions of golden samarium sulfide with yttrium present at molar equivalents of 0, 0.03, 0.14, 0.23 and 0.33. X-ray analysis was used to determine how the 4f electronic properties change as the temperature decreases. 

Mazzone says, “X-ray absorption in the partial fluorescence mode was conducted at the samarium L3 edge, providing estimates of the samarium 4f valence with a higher accuracy than regular X-ray absorption spectroscopy. Here the incident photon energy is tuned across the samarium L3 edge, exciting electrons from the samarium 2p orbital core-level into the empty 5d orbital conduction states (E₁), followed by a radiative decay of the samarium 3d orbital electrons into the 2p core hole (E₂). The energy of the 2p->5d transition is different for samarium 2+ and 3+, which enables us to estimate the value of the samarium valence between these two transitions.”

Figure 2 shows a schematic of the E₁ and E₂ transitions.


Figure 2. An estimate of the value of the samarium valence was obtained from an analysis of the E₁ and E₂ transitions occurring in a samarium atom undergoing X-ray absorption in the partial fluorescence mode. Figure courtesy of Brookhaven National Laboratory.

The researchers evaluated the unit cell volume of each samarium sulfide at temperatures ranging from 10 K to 300 K. Upon cooling, black-colored samarium sulfide showed a conventional contraction while each of the golden samarium sulfide showed some degree of negative thermal expansion. 

As the temperature decreased, the valence state of the samarium in golden samarium sulfide decreased to more of a 2+ state. Mazzone says, “This result suggested to us that the origin of negative thermal expansion is related to a change in the valence state.”

The researchers then used the Kondo-volume-collapse model to determine the mechanism for the negative thermal expansion effect for samarium sulfide. Mazzone says, “The reference point we used is that the element cerium undergoes a somewhat similar process, though the effect is reversed, leading to a strong positive thermal expansion.”

The negative thermal expansion is produced due to the triggering of a strong magnetic moment in the 4f orbital of the samarium atom. Mazzone says, “This orbital is almost half-full with electrons, and the strong magnetic moment attracts a cloud of electrons that circle around it. As the temperature decreases, more electrons move into the 4f orbital increasing the unit volume of the samarium atom, which directly leads to the negative thermal expansion and also decreasing the valence state.” 

Future work will involve fine tuning the percentage of yttrium in samarium sulfide to determine how this will influence negative thermal expansion. The researchers also are interested in evaluating other lanthanide series elements (thulium and ytterbium, in particular) that could exhibit similar expansion upon cooling.

Additional information can be found in a recent paper² or by contacting Mazzone at daniel.mazzone@psi.ch.

REFERENCES
1. Canter, N. (2019), “Lanthanide superhydride: Potential for superconductivity near room temperature,” TLT, 75 (5), pp. 18-19.
2. Mazzone, D., Dzero, M., Abeykoon, AM., Yamaoka, H., Ishii, H., Hiraoka, N., Rueff, J., Ablett, J., Imura, K., Suzuki, H., Hancock, J. and Jarrige, I. (2020), “Kondo-Induced Giant Isotropic Negative Thermal Expansion,” Physical Review Letters, 124 (12), 125701.