S C

IE N

T IF

IC H

IG H

LI G

H T

S E

L E

C T

R O

N IC

S T

R U

C T

U R

E ,

M A

G N

E T

IS M

A N

D D

Y N

A M

IC S

1 1 3 I H I G H L I G H T S 2 0 2 2

PRINCIPAL PUBLICATION AND AUTHORS

Non-Fermi liquid behavior below the Néel temperature in the frustrated heavy fermion magnet UAu2, C.D. O Neill (a), J.L. Schmehr (a) H.D.J. Keen (a), L. Pritchard Cairns (a), D.A. Sokolov (a), A. Hermann (a), D. Wermeille (b), P. Manuel (c), F. Krüger (c,d), A.D. Huxley (a), Proc. Natl. Acad. Sci. USA 118, 49 (2021); https:/doi.org/10.1073/pnas.2102687118 (a) School of Physics and Astronomy and Centre for Science at Extreme Conditions, University of Edinburgh (UK) (b) ESRF (c) ISIS Pulsed Neutron and Muon Source, Science Technology Facilities Council Rutherford Appleton Laboratory, Oxon (UK) (d) London Centre for Nanotechnology, University College London (UK)

The length scale of the composite state formed is xK = ℏvf/kB TK (vf, the Fermi velocity, is the speed of the conduction electrons). Since TK is typically small (10s of Kelvin), and vf > 104 ms-1, the Kondo length is large compared to the atomic spacing, making the increase in charge density very difficult to see.

This work investigated UAu2, which has a triangular lattice of uranium atoms with a high-temperature magnetic moment. The geometry of the triangular lattice frustrates antiferromagnetic order as shown in Figure 106. UAu2 does however partially order below TN = 42 K with a sinusoidally modulated ordered moment. Importantly, geometric frustration prevents the moments from taking their full value, meaning that Kondo screening is still required but is now modulated by the underlying magnetic structure. Since we are dealing with a periodic modulation, interference amplifies the intensity with which X-rays are diffracted from any charge accumulation compared with simply adding intensities from the charge density around single ions. Combined with the high X-ray flux and low background of the X-ray source at BM28, this gives an exquisite sensitivity for detecting any charge accumulation. Since the magnetism is incommensurate with the lattice, the diffracted intensity from the charge accumulation is not hidden by diffraction from the crystalline lattice.

The observed diffraction intensity from the magnetic structure and from the charge modulation are shown in Figure 107. The magnetic signal was measured with neutrons and confirmed with X-rays carefully tuned to an absorption edge (uranium M4). The charge modulation was measured with a much higher X-ray intensity away from the edge and occurs with half the period (twice the wavevector). At low temperature, the magnetic intensity saturates, which is normal behaviour for magnetic order. No harmonics in the magnetic signal were seen, showing that the modulation is sinusoidal and does not evolve towards a square wave at low temperature. Unlike the magnetic intensity, the charge intensity continues to grow at low temperature, indicating that it is determined by a lower temperature scale than TN. This charge modulation can therefore be identified with charge accumulation brought about by Kondo screening below TK.

In detail, the Kondo effect in UAu2 is more complicated than the simple Kondo effect described above and there is a distribution of temperature scales (equivalent to TK) extending down to very low temperature. This distribution may also explain another unusual property of this material, an unusual heat capacity C/T ∝log(T*/T). The heat capacity signature is of a marginal Fermi-liquid, distinct from the constant C/T found in almost all metals at low temperature.

Fig. 107: The upper (blue) line shows the magnetic diffraction intensity. This saturates

at low temperature. The lower (green) line shows the charge diffraction intensity, which

is at twice the modulation wavevector (half the period). This has a different temperature

dependence. The square of the upper curve is also plotted, which would be the

form for a lattice modulation arising from magnetostriction.