Searcharxiv⌕ Search

arXiv subjects

C. L. Henley

Publications and source records attributed to C. L. Henley.

At least 37 records · Page 2Linked to original sources
arXiv2005

Co-rich decagonal Al-Co-Ni: predicting structure, orientational order, and puckering

We apply systematic methods previously used by Mihalkovic et al. to predict the structure of the `basic' Co-rich modification of the decagonal Al70 Co20 Ni10 layered quasicrystal, based on known lattice constants and previously calculated pair potentials. The modelling is based on Penrose tile decoration and uses Monte Carlo annealing to discover the dominant motifs, which are converted into rules for another level of description. The result is a network of edge-sharing large decagons on a binary tiling of edge 10.5 A. A detailed analysis is given of the instability of a four-layer structure towards $c$-doubling and puckering of the atoms out of the layers, which is applied to explain the (pentagonal) orientational order.

cond-mat.mtrl-sci↗
arXiv2005

Discussion on clusters, phasons, and quasicrystal stabilisation

This paper summarises a two-hour discussion at the Ninth International Conference on Quasicrystals, including nearly 20 written comments sent afterwards, concerning (i) the meaning [if any] of clusters in quasicrystals; (ii) phason elasticity, and (iii) thermodynamic stabilisation of quasicrystals.

cond-mat.mtrl-sci↗
arXiv2005

Semiclassical ordering in the large-N pyrochlore antiferromagnet

We study the semiclassical limit of the $Sp(N)$ generalization of the pyrochlore lattice Heisenberg antiferromagnet by expanding about the $N \to \infty$ saddlepoint in powers of a generalized inverse spin. To leading order, we write down an effective Hamiltonian as a series in loops on the lattice. Using this as a formula for calculating the energy of any classical ground state, we perform Monte-Carlo simulations and find a unique collinear ground state. This state is not a ground state of linear spin-wave theory, and can therefore not be a physical (N=1) semiclassical ground state.

cond-mat.str-el↗
arXiv2004

Power-law spin correlations in pyrochlore antiferromagnets

The ground state ensemble of the highly frustrated pyrochlore-lattice antiferromagnet can be mapped to a coarse-grained ``polarization'' field satisfying a zero-divergence condition From this it follows that the correlations of this field, as well as the actual spin correlations, decay with separation like a dipole-dipole interaction ($1/|R|^3$). Furthermore, a lattice version of the derivation gives an approximate formula for spin correlations, with several features that agree well with simulations and neutron-diffraction measurements of diffuse scattering, in particular the pinch-point (pseudo-dipolar) singularities at reciprocal lattice vectors. This system is compared to others in which constraints also imply diffraction singularities, and other possible applications of the coarse-grained polarization are discussed.

cond-mat.stat-mech↗
arXiv2003

Combined energy -- diffraction data refinement of decagonal AlNiCo

We incorporate realistic pair potential energies directly into a non-linear least-square fit of diffraction data to quantitatively compare structure models with experiment for the Ni-rich $d$(AlNiCo) quasicrystal. The initial structure models are derived from a few {\it a priori} assumptions (gross features of the Patterson function) and the pair potentials. In place of the common hyperspace approach to the structure refinement of quasicrystals, we use a real-space tile decoration scheme, which does not rely on strict quasiperiodicity, and makes it easy to enforce sensible local arrangements of the atoms. Inclusion of the energies provides information complementary to the diffraction data and protects the fit procedure from converging on spurious solutions. The method pinpoints sites which are likely to break the symmetry of their local environment.

cond-mat.mtrl-sci↗
arXiv2003

Hole on a stripe in a spinless fermion model

In the spinless fermion model on a square lattice with infinite nearest-neighbor repulsion, holes doped into the half-filled ordered state form stripes which, at low doping, are stable against phase separation into an ordered state and a hole-rich metal. Here we consider transport of additional holes along these stripes. The motion of a single hole on a stripe is mapped to a one-dimensional problem, a variational wavefunction is constructed and the energy spectrum is calculated and compared to energies obtained by exact diagonalization.

cond-mat.str-el↗
arXiv2002

Dilute limit of a strongly-interacting model of spinless fermions and hardcore bosons on the square lattice

In our model, spinless fermions (or hardcore bosons) on a square lattice hop to nearest neighbor sites, and also experience a hard-core repulsion at the nearest neighbor separation. This is the simplest model of correlated electrons and is more tractable for exact diagonalization than the Hubbard model. We study systematically the dilute limit of this model by a combination of analytical and several numerical approaches: the two-particle problem using lattice Green functions and the t-matrix, the few-fermion problem using a modified t-matrix (demonstrating that the interaction energy is well captured by pairwise terms), and for bosons the fitting of the energy as a function of density to Schick's analytical result for dilute hard disks. We present the first systematic study for a strongly-interacting lattice model of the t-matrix, which appears as the central object in older theories of the existence of a two-dimensional Fermi liquid for dilute fermions with strong interactions. For our model, we can (Lanczos) diagonalize the 7 by 7 system at all fillings and the 20 by 20 system with four particles, thus going far beyond previous diagonalization works on the Hubbard model.

cond-mat.str-el↗
arXiv2002

Stripes and holes in a two-dimensional model of spinless fermions and hardcore bosons

We consider a Hubbard-like model of strongly-interacting spinless fermions and hardcore bosons on a square lattice, such that nearest neighbor occupation is forbidden. Stripes (lines of holes across the lattice forming antiphase walls between ordered domains) are a favorable way to dope this system below half-filling. The problem of a single stripe can be mapped to a spin-1/2 chain, which allows understanding of its elementary excitations and calculation of the stripe's effective mass for transverse vibrations. Using Lanczos exact diagonalization, we investigate the excitation gap and dispersion of a hole on a stripe, and the interaction of two holes. We also study the interaction of two, three, and four stripes, finding that they repel, and the interaction energy decays with stripe separation as if they are hardcore particles moving in one (transverse) direction. To determine the stability of an array of stripes against phase separation into particle-rich phase and hole-rich liquid, we evaluate the liquid's equation of state, finding the stripe-array is not stable for bosons but is possibly stable for fermions.

cond-mat.str-el↗
arXiv2002

Temperature-dependent "phason" elasticity in a random tiling quasicrystal

Both ``phason'' elastic constants have been measured from Monte Carlo simulations of a random-tiling icosahedral quasicrystal model with a Hamiltonian. The low-temperature limit approximates the ``canonical-cell'' tiling used to describe several real quasicrystals. The elastic constant K2 changes sign from positive to negative with decreasing temperature; in the ``canonical-cell'' limit, K2/K1 appears to approach -0.7, about the critical value for a phason-mode modulation instability. We compare to the experiments on i-AlPdMn and i-AlCuFe.

cond-mat.mtrl-sci↗
arXiv2002

Icosahedral Ti-Zr-Ni: A groundstate quasicrystal?

The first complete ab initio zero-temperature ternary phase diagram is constructed from the calculated energies of the elemental, binary and ternary Ti-Zr-Ni phases; for this, the icosahedral i-TiZrNi quasicrystal phase is approximated by periodic structures of up to 123 atoms/unit cell, based on a decorated-tiling model. The approximant structures containing the 45-atom Bergman cluster were nearly degenerate, and stable against the competing binary phases. It is speculated that i-TiZrNi may be a ground state quasicrystal, as it is experimentally the low-temperature phase for its composition.

cond-mat.mtrl-sci↗
arXiv2002

Structure of the icosahedral Ti-Zr-Ni quasicrystal

The atomic structure of the icosahedral Ti-Zr-Ni quasicrystal is determined by invoking similarities to periodic crystalline phases, diffraction data and the results from ab initio calculations. The structure is modeled by decorations of the canonical cell tiling geometry. The initial decoration model is based on the structure of the Frank-Kasper phase W-TiZrNi, the 1/1 approximant structure of the quasicrystal. The decoration model is optimized using a new method of structural analysis combining a least-squares refinement of diffraction data with results from ab initio calculations. The resulting structural model of icosahedral Ti-Zr-Ni is interpreted as a simple decoration rule and structural details are discussed.

cond-mat.mtrl-sci↗
arXiv2001

Effective Hamiltonian and low-lying energy clustering patterns of four-sublattice antiferromagnets

We study the low-lying energy clustering patterns of quantum antiferromagnets with p sublattices (in particular p=4). We treat each sublattice as a large spin, and using second-order degenerate perturbation theory, we derive the effective (biquadratic) Hamiltonian coupling the p large spins. In order to compare with exact diagonalizations, the Hamiltonian is explicitly written for a finite-size lattice, and it contains information on energies of excited states as well as the ground state. The result is applied to the face-centered-cubic Type I antiferromagnet of spin 1/2, including second-neighbor interactions. A 32-site system is exactly diagonalized, and the energy spectrum of the low-lying singlets follows the analytically predicted clustering pattern.

cond-mat.str-el↗
arXiv2001

Total-energy-based prediction of a quasicrystal structure

Quasicrystals are metal alloys whose noncrystallographic symmetry and lack of structural periodicity challenge methods of experimental structure determination. Here we employ quantum-based total-energy calculations to predict the structure of a decagonal quasicrystal from first principles considerations. We employ Monte Carlo simulations, taking as input the knowledge that a decagonal phase occurs in Al-Ni-Co near a given composition, and using a few features of the experimental Patterson function. The resulting structure obeys a nearly deterministic decoration of tiles on a hierarchy of length scales related by powers of $τ$, the golden mean.

cond-mat↗
arXiv2000

Spinless fermions and charged stripes at the strong-coupling limit

Spinless fermions on a lattice with nearest-neighbor repulsion serve as a toy version Hubbard model, and have a symmetry-broken even/odd superlattice at half-filling. At infinite repulsion, doped holes form charged stripes which are antiphase walls (as noted by Mila in 1994). Exact-diagonalization data for systems up to 36 sites around 1/4 filling, and also for one or two holes added to a stripe of length up to 12, indicate stability of the stripe-array state against phase separation. In the {\it boson} version of the model, the same behavior can be stabilized by addition of a four-fermi term.

cond-mat↗
arXiv2000

Structure determinations for random-tiling quasicrystals

How, in principle, could one solve the atomic structure of a quasicrystal, modeled as a random tiling decorated by atoms, and what techniques are available to do it? One path is to solve the phase problem first, obtaining the density in a higher dimensional space which yields the_averaged_ scattering density in 3-dimensional space by the usual construction of an incommensurate cut. A novel direct method for this is summarized and applied to an i(AlPdMn) data set. This averaged density falls short of a true structure determination (which would reveal the typical_unaveraged_ atomic patterns.) We discuss the problematic validity of inferring an ideal structure by simply factoring out a ``perp-space'' Debye-Waller factor, and we test this using simulations of rhombohedral tilings. A second, ``unified'' path is to relate the measured and modeled intensities directly, by adjusting parameters in a simulation to optimize the fit. This approach is well suited for unifying structural information from diffraction and from minimizing total energies derived ultimately from ab-initio calculations. Finally, we discuss the special pitfalls of fitting random-tiling decagonal phases.

cond-mat.mtrl-sci↗
arXiv2000

Effective potentials for 6-coordinated Boron: structural geometry approach

We have built a database of ab-initio total energies for elemental Boron in over 60 hypothetical crystal structures of varying coordination $Z$, such that every atom is equivalent. Fitting to each subset with a particular $Z$, we extract a classical effective potential, written as a sum over coordination shells and dominated by three-atom (bond angle dependent) terms. In the case Z=6 (lowest in energy and most relevant), the classical potential has a typical error of 0.1 eV/atom, and favours the ``inverted-umbrells'' environment seen in real Boron.

cond-mat.mtrl-sci↗
arXiv1999

Semiclassical mechanics of a non-integrable spin cluster

We study detailed classical-quantum correspondence for a cluster system of three spins with single-axis anisotropic exchange coupling. With autoregressive spectral estimation, we find oscillating terms in the quantum density of states caused by classical periodic orbits: in the slowly varying part of the density of states we see signs of nontrivial topology changes happening to the energy surface as the energy is varied. Also, we can explain the hierarchy of quantum energy levels near the ferromagnetic and antiferromagnetic states with EKB quantization to explain large structures and tunneling to explain small structures.

cond-mat↗