Deep Inelastic Neutron Scattering Experiments Research Articles

Force constant disorder in the Ni44Nb56 bulk metallic glass as observed by deep inelastic neutron scattering augmented by isotopic substitution

In this work, the force-constant disorder in nickel-niobium metallic glass, Ni44Nb56, was studied using the deep inelastic neutron scattering (DINS) technique augmented by isotopic substitution. The distributions of DINS observables (the nuclear kinetic energies, the width of the nuclear momentum distributions, and the effective force constants) were measured in Ni44Nb56 and compared with their counterparts obtained from ab initio harmonic lattice (HLD) simulations for the crystalline forms of nickel, niobium, and the NiNb crystal and from the reverse Monte Carlo (RMC) simulations augmented by effective force fields performed for Ni44Nb56. The force-constant distribution of nickel, obtained from the analysis of the results of the DINS experiments, was found to be two times broader than its counterparts estimated based on the HLD and RMC simulations. In the case of niobium, the force-constant distribution inferred from the DINS experiments is estimated to be an order of magnitude broader than the ab initio HLD prediction in the NiNb crystal. Moreover, no disorder-induced softening (with respect to its crystalline counterparts) of the effective force constants of Ni and Nb in Ni44Nb56 was observed. The lack of disorder-induced softening in Ni44Nb56 is consistent with the correlation between the short-range order, defined by the average coordination number and the interatomic distances, and the magnitudes of the effective force constants. The obtained results are consistent with a picture, whereby disorder induces symmetrical broadening of phonon dispersion curves, and phonon softening is limited to low-energy modes carrying negligible amounts of nuclear kinetic energy. The obtained results have important ramifications for engineering the properties of bulk metallic glasses.

The Harmonic Picture of Nuclear Mean Kinetic Energies in Heavy Water

This paper presents a study of the total mean kinetic energy, ⟨EK⟩, and of individual projections along a given molecular axis, ⟨EK⟩α, for D and O nuclei in D2O, derived using a harmonic model. Our theoretical approach assumes decoupling amongst translational, rotational and vibrational modes. Resulting values of these dynamical quantities are discussed in terms of the anisotropy of the quantum kinetic energy tensor, its relation to the local potential, and deviations from the hypothesis of harmonicity and mode decoupling. Results are compared with corresponding quantities obtained from Deep Inelastic Neutron Scattering experiments performed on liquid and solid D2O, where the short-time dynamics and local environment of D and O atoms were probed. The present study confirms an overall picture where even small changes in the short-range environment of D and O nuclei have a strong influence on the quantum behaviour of heavy water.

Ground state proton dynamics in stable phases of water

The proton momentum distribution, n(p), and the mean kinetic energy, 〈EK〉, of water are intimately connected to the nuclear quantum effects associated to the equilibrium ground state of water. The physical quantities have been studied via Deep Inelastic Neutron Scattering experiments performed at the ISIS neutron source, in the temperature range 300K⩽T⩽673K. The Compton lineshape analysis provides values of the proton mean kinetic energy 〈EK〉 which are constantly in excess of the classical values as T increases. From the analysis of n(p) the effective proton potentials are derived. The equilibrium dynamics of H-bonded protons is discussed.

Kinetic energy of protons in ice Ih and water: A path integral study

The kinetic energy of H and O nuclei has been studied by path integral molecular-dynamics simulations of ice Ih and water at ambient pressure. The simulations were performed by using the q-TIP4P/F model, a point-charge empirical potential that includes molecular flexibility and anharmonicity in the OH stretch of the water molecule. Ice Ih was studied in a temperature range between 210 and 290 K, and water between 230 and 320 K. Simulations of an isolated water molecule were performed in the range 210--320 K to estimate the contribution of the intramolecular vibrational modes to the kinetic energy. Our results for the proton kinetic energy ${K}_{H}$ in water and ice Ih show both agreement and discrepancies with different published data based on deep inelastic neutron-scattering experiments. Agreement is found for water at the experimental melting point and in the range 290--300 K. Discrepancies arise because data derived from the scattering experiments predict in water two maxima of ${K}_{H}$ around 270 and 277 K, and that ${K}_{H}$ is lower in ice than in water at 269 K. As a check of the validity of the employed water potential, we show that our simulations are consistent with other experimental thermodynamic properties related to ${K}_{H}$, such as the temperature dependence of the liquid density, the heat capacity of water and ice at constant pressure, and the isotopic shift in the melting temperature of ice upon isotopic substitution of either H or O atoms. Moreover, the temperature dependence of ${K}_{H}$ predicted by the q-TIP4P/F model for ice Ih is found to be in good agreement with results of path integral simulations using ab initiodensity-functional theory.

Multiple scattering and attenuation corrections in Deep Inelastic Neutron Scattering experiments

Multiple scattering and attenuation corrections in Deep Inelastic Neutron Scattering experiments are analyzed. The theoretical basis of the method is stated, and a Monte Carlo procedure to perform the calculation is presented. The results are compared with experimental data. The importance of the accuracy in the description of the experimental parameters is tested, and the implications of the present results on the data analysis procedures is examined.

Recent developments of the e.VERDI project at ISIS

Epithermal neutrons are unique probes for deep inelastic neutron scattering (DINS) at high-energy and high-momentum transfer and inelastic neutron scattering (INS) at high-energy and low-wave vector transfers. In the latter case the kinematic range has recently been revisited given that the high values of initial and final wave vectors achievable in a scattering process coupled to small scattering angles allow to access a so far unexplored region in q, ω, i.e ω>1 eV and q<10 Å −1. In order to exploit DINS and INS in the high-energy and low-momentum range, a very low angle detector (VLAD) bank will be delivered within the e.VERDI Project together with a novel bank at intermediate scattering angles. The former will provide a new tool for investigating high-energy excitations in magnetic systems and semiconductors. The VLAD bank will be constructed following the concept of a resonance detector (RD) spectrometer, recently revised for DINS experiments in the range 1–100 eV.This is a more promising approach in terms of efficiency and signal-to-noise ratio compared to standard neutron counting techniques at eV energies. Results from two devices, cadmium zinc telluride (CZT) solid-state detectors and YAlO 3 perovskite (YAP) scintillators used in RD configurations, will be presented.

Experimental test of a theoretical analysis of deep inelastic neutron scattering experiments for H and D nuclei

In a recent paper by Blostein et al. (Physica B 304 (2001) 357) was presented a critical theoretical analysis of different procedures, with which data of deep inelastic neutron scattering (DINS) experiments are usually processed. One part of that criticism—which is experimentally tested in the present paper—claimed the existence of serious errors in the determined areas of overlapping DINS-peaks, in particular in the case of the considerable overlapping of the strong proton peak with the weak deuteron peak. Our experimental test presented here consists in the comparison of the D-peak areas determined from DINS-data of H/D mixed systems, taken in both the “forward” and the “backward” scattering directions. Since there is no H-peak in the “backward” scattering direction, the claimed error cannot occur in this regime. New experimental results from (A) pure D 2O, and (B) the equimolar H 2O/D 2O mixture, are presented. The liquids were contained in a Nb can. For both systems, it is shown that the D-peak areas in the “forward” and the “backward” scattering directions are essentially equal. This proves that the aforementioned criticism, although very stimulating, is of no relevance in the specific context of our (previous and present) DINS experiments. Moreover, the new experiments show that the cross-section of D is fully consistent with conventional expectations.

Theory of deep inelastic neutron scattering: Hard-core perturbation theory.

Details are presented of a new many-body theory for deep inelastic neutron scattering (DINS) experiments to measure momentum distributions in quantum fluids and solids. The high-momentum and energy-transfer scattering law in helium is shown to be a convolution of the impulse approximation with a final-state broadening function which depends on the scattering phase shifts and the radial distribution function. The predicted broadening satisfies approximate Y scaling, is neither Lorentzian nor Gaussian, and obeys the f, ${\ensuremath{\omega}}^{2}$, and ${\ensuremath{\omega}}^{3}$ sum rules. The derivation uses a combination of Liouville perturbation theory, projection superoperators, and semiclassical methods which I term ``hard-core perturbation theory.'' A review is presented of the predictions of prior theories for DINS experiments in relation to the present work. A subsequent paper will present massive numerical predictions and a discussion of DINS experiments on superfluid $^{4}\mathrm{He}$.

Atomic momentum distributions in condensed matter

The theory of the momentum distribution function for atoms in condensed matter is reviewed and compared with results of deep-inelastic neutron-scattering experiments. We discuss, in particular, the case of classical and almost-classical liquids, harmonic and anharmonic crystals, and solid and liquid 4He. Except for liquid 4He in the superfluid phase, the momentum distribution is always Gaussian to a good approximation. In some cases this Gaussian behavior is of dynamical origin while, in others, it is a consequence of the central-limit theorem. The observed momentum distribution in superfluid 4He provides direct experimental evidence for the macroscopic occupation of the zero-momentum state and the value of the condensate fraction can be obtained.