Spring 2018, Number 12 Department of Physics Newsletter PLASMA WAVES How exploiting the constructive interference of plasma waves could contribute to the next generation of particle accelerators THIN FILM DEVICES Tracking defects in molecular semiconductors SPIN LIQUIDS Finding new materials FINDING NEARBY EXOPLANETS Space-based transit surveys www.physics.ox.ac.uk PROUD WINNERS OF: SCIENCE NEWS SCIENCE NEWS www.physics.ox.ac.uk/research www.physics.ox.ac.uk/research Designer materials a) b) What is a good strategy for finding SPIN LIQUIDS spin liquid materials? We should avoid systems in which interactions favour FROM THEORY TO EXPERIMENT: FINDING NEW MATERIALS a limited number of regular, solid-like Prof John Chalker spin configurations. Instead we require A central aim in condensed matter physics is matter physics for almost half a century. As occurs in the so-called fractional quantum Hall frustration. Then there may be many to understand and classify possible phases of hinted at by the analogies, quantum fluctuations effect. In that instance, electrons (which are classical configurations that provide matter. Everyday examples of such phases are can be large enough in some magnetic systems indivisible particles from the viewpoint of high- equally good compromise solutions to gases, liquids and solids, and we know that nearly to maintain a liquid-like state of spins even at energy physics) break into pieces, each with a the problem of minimising the energy everything becomes solid at sufficiently low zero temperature, and we are now beginning to fraction (one third in the simplest case) of the of the system, and a quantum ground temperature. The sole exceptions to this rule are understand the properties of these spin liquid elementary charge. A second example arises in state that is a superposition of these the two isotopes of helium, in which quantum phases. Progress has been made by theorists using magnetic materials that have one-dimensional different configurations. fluctuations are so large that a crystalline lattice pencil, paper and computers to study simplified arrangements of spins. The geometry makes this Fig 2: Kitaev model on the honeycomb lattice as described in the text. cannot form at ordinary pressures, even at the models, and by experimentalists who synthesise the easiest case to visualise, as explained in the Prof Radu Coldea A fruitful route to frustration is shown absolute zero of temperature. and characterise new materials. box below. We hope that spin liquids will give us in fig 2b. Consider first the interaction many more examples of fractionalisation. The between a pair of spins, and suppose For Ir4+ ions inside edge-sharing cubic additional interactions no longer The features that distinguish a solid from a liquid Spin liquids turn out to be even more interesting search for spin liquids is very much curiosity- that energy is minimised only when octahedra of oxygen ions, as shown in interfere, we expect a spin liquid or gas arise microscopically from the fact that than the remarkable superfluid phases of liquid driven basic physics. It sets out to employ the both are aligned along a specified fig 2a, the strong spin-orbit coupling with fractionalised quasiparticles as its atoms have a fixed arrangement in space. helium (at least in the doting eyes of condensed enormous scope offered by the periodic table axis, as shown in fig 2b (top). Then at the Ir site ensures that the magnetic theoretically predicted. Realising such Alternative phases of matter can be characterised matter physicists!). In particular, we expect of elements and materials science, as a way of either spoiling the co-alignment (fig 2b exchange interaction is firmly linked a phase experimentally would open up by degrees of freedom other than the positions some of them (we now know that there are realising some of the astonishing consequences middle) or aligning along a different to the direction of the electronic completely new avenues to explore and of atoms. In magnetic materials, the relevant many different types) to have what are known as of quantum mechanics. Oxford physicists have What happens axis to the specified one (fig 2b bottom) orbitals making up the Ir-O-Ir bond. manipulate fractionalised phases of degrees of freedom are the orientations of the fractionalised excitations. By this we mean that been at the centre of these international efforts. in a magnetic will cost energy. Frustration arises if Edge-sharing octahedra forming a quantum matter. The current research atomic magnetic moments, or spins. At sufficiently the microscopic constituents of the material – Prof Radu Coldea's group has been concerned material when we take a lattice of such spins, with honeycomb arrangement represent frontier in the field is the exploration of high temperatures, spins in these systems are spins or electrons – correlate their motion with with finding and studying new materials that interactions favouring different axes a very stable structural framework, the phase diagram of candidate Kitaev thermally disordered and the materials are in the one another so strongly that one can no longer quantum are candidate spin liquids [1, 2], while Prof John for different pairs. An example is the found in several layered honeycomb materials as a function of pressure- paramagnetic phase – the magnetic equivalent of a describe the collective behaviour in terms of fluctuations are honeycomb lattice shown in fig 2a, Chalker, Dr Dmitry Kovrizhin and collaborators materials, including Na2IrO3, α-Li2IrO3 tuning, bond-angle deformation via gas. At low temperature, many magnetic materials individual electrons or spins. In effect, the original large? During where the red, green and blue links have done calculations to predict what should be and α-RuCl3, where in the latter the doping or inter-layer intercalations, have ordered arrangements of spin orientations, particles dissolve into a quantum soup and are seen in experiments on these phases [3]. the last few years indicate three competing preferred magnetic ion is Ru3+. or magnetic field tuning, hunting for analogous to the regular positions of atoms in a lost. In their place, when we probe the material some answers directions. signatures of such a spin liquid. crystal. For example, the macroscopic magnetism (for example, using magnetic neutron scattering have taken The materials studied so far develop of a fridge magnet is due to a net alignment of its to make excitations) we find new particles with Models of this type were suggested solid-like magnetic order at low [1] R Coldea, D A Tennant and Z Tylczynski, Phys Rev B 68, shape to this by the Caltech physicist Alexei Kitaev atomic magnetic moments. quantum numbers corresponding to fractions of 134424 (2003). temperatures, thought to be stabilised the original constituents. question, which and they turned out to be exactly [2] A Biffin, R D Johnson, I Kimchi, R Morris, A Bombardi, by additional interactions that co- Fig 3 (below): a) Counter-rotating What happens in a magnetic material when J G Analytis, A Vishwanath and R Coldea, Phys Rev Lett 113, has loomed solvable. At first, they seemed simply exist with a dominant frustrated spiral magnetic order in the 3D quantum fluctuations are large? During the last At present, physicists have only a few fully 197201 (2014). over condensed to be theorists’ toys, amenable to a exchange. The ordering patterns are honeycomb lattice in -Li IrO few years some answers have taken shape to this certified examples of phases of matter that show [3] J Knolle, D L Kovrizhin, J T Chalker and R Moessner, beautiful mathematical analysis, but γ 2 3 matter physics rather unusual: see fig 3a. They give stabilised by Kitaev interactions question, which has loomed over condensed fractionalisation. Perhaps the most spectacular Phys Rev Lett 112, 207203 (2014). far from reality. Remarkably, however, for almost half a direct evidence that Kitaev interactions [2]. b) X-ray diffraction data on two it turns out that transition metal ions century. indeed exist in nature and stabilise different magnetic Bragg peaks with strong spin orbit interaction may unconventional magnetism. If we used to deduce the magnetic Fractionalisation: splitting the electron have the key ingredients to realise such can modify the materials or the structure in a) (solid lines are fits to physics. In some types of quantum matter, new excitations or quasiparticles experimental conditions so that the model). can emerge, which are quite different from the original building blocks. This is most easily pictured for a one-dimensional quantum antiferromagnet. Magnetic materials host atomic magnetic moments, a) b) which arise from orbital currents of electrons and from the intrinsic magnetic moments of electrons. In a one-dimensional version of such a material these moments – known as spins – are arranged in well- separated chains. Interactions between neighbouring spins within a chain favour opposite alignment, and interactions between spins in different chains are negligible. The ground state for a system of this type is shown (schematically!) as the top row of fig 1a. Adding energy to the system, we can reverse one of the spins, as depicted in the Fig 1: a) Ground and excited states of a spin-1/2 antiferromagnetic middle row of fig 1a. Contrary to the obvious expectation, this process chain. b) Inelastic neutron scattering from pairs of spin-1/2 spinons generates not one excitation, but two independent ones. These in Cs2CuCl4 [1]. excitations, known as spinons, are not the reversed spin (coloured red in fig 1a), but rather the domain walls between parallel spins row of fig 1a. These excitations have been observed in the material Fig 3 data were collected using instrumentation at (pink planes in fig 1a). Two of these are created when a single spin is Cs2CuCl4 using inelastic neutron scattering. Some of the resulting reversed; by inserting a segment of reversed spins, the two spinons data appear in fig 1b, which shows the response of the material as a the Diamond synchrotron can be separated without energy cost, as illustrated in the bottom function of the energy and momentum transferred to it.