Page 18 - ILL Annual Report 2019

MAGNETISM
Tom Fennell. British
Laboratory for Neutron Scattering, Paul Scherrer Institute, Switzerland
‘I study frustrated magnets, especially pyrochlores, using diffuse and inelastic neutron scattering. I am interested in the description of such systems by emergent field
theories and the exploration of new material properties, such as unconventional correlations and dynamics that are highlighted by this approach.’
Multiple Coulomb phase in the
fluoride pyrochlore CsNiCrF 6
Three-axis spectrometer ThALES
In this study we investigated CsNiCrF6,
a pyrochlore in which both structural and magnetic correlations can be identified
as Coulomb phases: a charge ice and
an antiferromagnetic Coulomb phase respectively. As far as we can determine, neither has been observed independently before. However, the combination is particularly interesting as it affords the possibility of studying not only structural or magnetic frustration but also their interplay.
AUTHORS
T. Fennell (Paul Scherrer Institute, PSI, Switzerland) M. Boehm and P. Steffens (ILL)
ARTICLE FROM
Nat. Phys. (2019)—doi: https://doi.org/10.1038/s41567-018-0309-3
REFERENCES
[1] C.L. Henley, Ann. Rev. Cond. Matt. Phys. 1 (2010) 179
[2] S.T. Banks and S.T. Bramwell, EPL 97 (2012) 27005
[3] P.H. Conlon and J.T. Chalker, Phys. Rev. Lett. 102 (2009) 237206 [4] T. Fennell et al., Nat. Phys. 15 (2019) 60
A Coulomb phase [1] is a state of matter in which the correlations of local degrees of freedom can be described by a non-divergent field (figure 1). Many systems can support Coulomb phases, including dimer models, ice models, correlated structural disorder and spin systems. Rather than long-range order, characterised by an order parameter and broken symmetry, or exponentially decaying short-range order, a Coulomb phase is critical with correlations that decay as a power-law (1/r3 for three-dimensional examples). Coulomb phases are therefore of interest in the study of emergent, many- body physics and unconventional phase transitions: the former aspect derives from identification of the low energy states of the Coulomb phase with a free field in a vacuum, whose co- operative fluctuations are one type of dynamics of the system and in which fractional quasiparticle excitations that are the charges of the relevant field may be created/annihilated (and also give rise to dynamics); the latter aspect stems from the form of the correlations, which means that phase transitions out of a Coulomb phase will often fall outside the Landau– Ginzburg–Wilson paradigm.
Some of the most studied examples are spin systems, including spin ice and the pyrochlore Heisenberg antiferromagnet. Spin ice can be associated with the question of distributing two types of cation (i.e. with different charges) on the pyrochlore lattice. To minimise the energy of the structure, the cations should obey the condition that there are two of each type
on every tetrahedron of the pyrochlore lattice—an example of an ice rule. This structure, a so-called charge ice, is interesting in the sense that it is a crystalline solid that hosts a fascinating type of disorder. The cations are disordered but not random—a form of correlated disorder whose simple underlying motif gives rise to non-trivial correlations.
CsNiCrF6 is a candidate charge ice, since Ni2+ and Cr3+ jointly form the pyrochlore lattice. However, it has a further interesting feature in that both cations are magnetic and no magnetic order is observed, even far below the Curie–Weiss temperature. Despite the destruction of the high symmetry
of exchange interactions in the pyrochlore lattice by the correlated disorder, a frustrated spin system is still obtained. The average structure of CsNiCrF6 is a pyrochlore with Ni2+ and Cr3+ jointly distributed on the pyrochlore lattice (Vivaldi, ILL and TRiCS (now Zebra), PSI). However, separation of structural and magnetic correlations using polarised diffuse neutron scattering (D7, ILL) shows that both structural and
ANNUAL REPORT 2019