research

RESEARCH ACTIVITIES

The primary thrust of the research in my group is to study the influence of high hydrostatic pressures on the magnetic, superconducting and structural properties of exotic condensed matter systems. In this regard the application of pressure is used to test our understanding of interesting physical phenomena as a function of lattice parameter, to search for new phenomena in solids, as well as, in combination with high temperatures, to synthesize new materials unavailable under ambient pressure conditions. Some of the samples we study are synthesized in my laboratory. In the following I give a brief summary of the principal research projects of current interest to my group. For a listing of our recent publications on these topics in downloadable PDF form, see "Recent Publications" on our WWW-homepage. Click on Publications for a compilation of all our publications since 1990.

PRESSURE-INDUCED SUPERCONDUCTIVITY IN EUROPIUM METAL

Of the 92 naturally occurring elements in the periodic table, 52 of them were known to be superconducting: 30 at ambient pressure and 22 more only under high pressure. The high-pressure elemental superconductors include such unlikely elements as oxygen, iron, and silicon!

Early in 2009 MATHEW DEBESSAI in our group discovered that europium metal also becomes superconducting near 2 K for pressures greater than 80 GPa (800,000 atm), thus becoming the 53rd elemental superconductor in the periodic table. What makes this discovery particularly interesting is the fact that europium, like almost all the rare-earth elements, possesses a strong local magnetic moment which totally suppresses superconductivity. The fact that europium becomes superconducting under extreme pressures implies that the magnetism has been either destroyed or at least weakened. In fact, if divalent europium becomes trivalent under pressure, as all other rare-earth metals except ytterbium are, then its ground state would indeed be expected to be non-magnetic. But the first excited state, which is magnetic, lies very close to the ground state, making trivalent europium a Van Vleck paramagnetic, like the actinide element americium. Interesting is that the value of europium's superconducting transition temperature (2 K) is very low compared to that of other trivalent s,p,d-electron elements like Sc, Y, La and Lu (10 - 20 K). Perhaps superconducting europium, at least to pressures as high as 142 GPa (1.4 Megabar), is not really trivalent but rather "mixed valent". Future experiments by graduate student WENLI BI at the Advanced Photon Source (APS) at the Argonne National Labs are designed to clarify what exactly europium valence state is at extreme pressures. WENLI has also carried out extensive x-ray diffraction studies at the APS and found four structural phase transitions to 92 GPa in europium metal.

SUPERCONDUCTING PHASE DIAGRAM OF LI METAL TO 67 GPa

In 2002 two groups reported that Li becomes superconducting under pressures greater than 20 GPa, confirming an earlier (1986) indication of superconductivity in Li by Lin and Dunn. Since none of these three experiments used any pressure medium, the hard diamond anvils and stiff gasket walls pressed directly onto the Li sample, generating shear stresses and plastic deformation. The question is whether the superconducting state is intrinsic or perhaps only arises because of the shear stresses. We decided to use the softest solid known, dense helium, as pressure medium; surrounding the very reactive Li sample with helium might also have a further bonus -- the reduction of reactions between Li and diamond. On her very first try, student SHANTI DEEMYAD loaded Li into a rhenium-gasketed diamond-anvil cell along with tiny ruby spheres and helium pressure medium and reached a pressure of 67 GPa, a new record for our group at that time! She indeed confirmed superconductivity in Li above 20 GPa, but the detailed dependence of T_c on pressure, reaching values as high as 14 K, differed considerably from that published earlier. Structural phase transitions were indicated at 20, 30, and 62 GPa. The large increase in T_c for pressures between 20 and 30 GPa is highly unusual and likely due to the fact that the very large compression of Li brings the ion cores close together which forces the conduction electrons into interstitial sites, thus leading to strong anomalies in the electronic properties. The strong enhancement in the pseudopotential not only pushes T_chigher, but also can lead to symmetry-breaking phase transitions which likely occur at 30 and 62 GPa. This ground breaking work points the way to many further experiments on the alkali metals under high pressure conditions. For example, Takahiro Matsuoka in Katsuya Shimizu's group has very recently (2008) shown that above 70 GPa Li actually turns into a semiconductor. This counterintuitive behavior was predicted in 1999 by Neaton and Ashcroft.

Following theoretical work by Feng, Ashcroft, and Hoffmann at Cornell University, we recently studied CaLi₂ under extremely high pressures and confirmed that this compound also exhibits the anomalous properties found for elemental Li and Ca. This implies that the predictions of Ashcroft's group, that all properties (electronic, structural, magnetic) become highly anomalous under pressures sufficient to bring the ion cores in close proximity, is not restricted to elemental solids but has very general validity for all forms of matter such as multielement compounds and alloys, crystalline or amorphous. This is now the primary thrust of our group in the field of superconductivity. Many exciting experiments are waiting to be carried out.

DEPENDENCE OF T_c FOR Sc, Y, La, Lu ON MEGABAR PRESSURES

Students JAMES HAMLIN and MATHIEW DEBESSAI have recently studied the d-electron superconductors Sc, Y, and Lu to pressures of almost 2 Megabars and found that T_c reaches values as high as 20 K (Y and Sc), the second highest value for any elemental superconductor! All three of these elements do not superconduct at ambient pressure. As with the alkali and alkaline-earth systems above, this anomalous behavior of T_c originates from the sharp reduction in the space available to the conduction electrons as extreme pressure pushes the ion cores together. A very interesting systematics reveals itself for Sc, Y, Lu, and La if their T_c's are plotted versus the relative amount of free volume available to the conduction electrons outside the ion cores.

More generally, these interesting phenomena are but a precursor of what happens to all properties of solids if astronomic pressures are applied which are sufficient to actually break up the shell structure of the constituent atoms, leaving only a Thomas-Fermi gas behind.

DEPENDENCE OF T_c OF SUPERCONDUCTING MgB₂ ON HYDROSTATIC PRESSURE

Like the isotope effect, the dependence of T_c on pressure can give information on the nature of the superconducting pairing interaction and test theoretical models. Shear stresses from the pressure medium or from grain boundaries in polycrystalline samples, however, may falsify the measured T_c(P) dependence, particularly in elastically anisotropic substances like MgB₂ a superconductor discovered in January 2001. For this reason it is advantageous to determine T_c(P) on single crystals using the most hydrostatic pressure medium known, helium.
Students TAKAHIRO TOMITA and JAMES HAMLIN have determined T_c(P) for both boron isotopes of MgB₂ to 0.7 GPa in a helium gas apparatus; student SHANTI DEEMYAD extended these studies to much higher pressures in a diamond-anvil cell loaded with dense helium to 28 GPa for single-crystalline and 32 GPa for polycrystalline samples. T_c decreases from 39 K at ambient pressure to 15 K at 32 GPa with an initial slope dT_c/dP = -1.11(2) K/GPa. Evidence is found that the differing values of dT_c/dP reported in the literature result primarily from shear-stress effects in non-hydrostatic pressure media and not differences in the samples. Although comparison of these results with theory supports phonon-mediated superconductivity, a critical test of theory must await volume-dependent calculations based on the solution of the anisotropic Eliashberg equations.

OXYGEN ORDERING IN HIGH-TEMPERATURE SUPERCONDUCTORS

Although many thousands of scientific papers have been written on the oxide superconductors since their discovery more than 15 years ago, our understanding is far from complete. Indeed, one does not yet really understand why they are superconducting at all. One of the barriers to understanding is the fact that these materials exhibit a myriad of different types of structural defects. Perhaps the most important of these, and likely the most subtle, is the possibility of local oxygen ordering processes. In the oxide superconductors, oxygen defects possess are quite mobile, even at room temperature and below, and may order on a short length scale in the tetragonal crystalline lattice. When the temperature or pressure is varied, the degree of oxygen ordering changes; this in turn leads to charge transfer to or from the CuO₂ planes and thus to marked shifts in the value of the superconducting transition temperature T_c. My group has found that oxygen still has considerable mobility even at 15 K, a temperature well below the value of T_c for many systems. At the very least, one has to be aware that these structural defect can strongly influence the superconducting state; it is, however, possible that these mobile oxygen defects themselves play an important role in the superconducting state itself. Future experiments are designed to explore the nature and extent of oxygen-ordering phenomena. Three of my students, SASCHA SADEWASSER, ANNE-KATRIN KLEHE and CRAIG LOONEY have carried out extensive experiments in this project. This work involved a collaboration with John Wagner (University of North Dakota), Allen Hermann (University of Colorado, Boulder) and Boyd Veal (Argonne National Labs).

In collaboration with Boyd Veal (Argonne National Labs), student TAKAHIRO TOMITA studied pressure-induced oxygen-ordering effects in the critical current density in bicrystalline rings of YBCO, where the mismatch angle between the single crystalline lattices varies over a wide range, as does the oxygen content. Takahiro found clear oxygen ordering effects and found that J_c increases substantially with pressure. In fact, in the PRL that he first authored, Takahiro showed that the presence or absence of oxygen ordering effects in the grain boundaries, which limit the value of J_c, can be used to monitor the level of occupation of the oxygen sites in grain boundaries, thus revealing whether further oxygenation is necessary or not in HTSC applications materials.

INTRINSIC DEPENDENCE OF T_c ON PRESSURE IN HIGH-TEMPERATURE SUPERCONDUCTORS

A large number of studies in my group, principally by students SASCHA SADEWASSER, ANNE-KATRIN KLEHE and STEFAN KLOTZ, of the dependence of T_c on pure hydrostatic pressure in high-temperature superconductors have uncovered a clear systematics which has advanced our understanding of these complex systems. We conclude that the value of T_c depends little on the separation between the planes, but primarily on the in-plane lattice parameter a, Tc increasing as the inverse 4th to 5th power of a. Strategies to further enhance T_c over the current record of 134 K for Hg-2223 include searching for compounds where a is compressed chemically.