Reverse Monte Carlo Configurations Research Articles

Local Structure Properties of Hydrogenated and Nonhydrogenated Amorphous In–Ga–Zn–O Thin Films Using XAFS and High-Energy XRD

The strong effect of impurity hydrogen (H) on subgap electronic states in amorphous In–Ga–Zn–O thin-film transistors (a-IGZO TFTs) was confirmed using standard (STD) and ultrahigh-vacuum (UHV) sputtering systems with different base pressures of 10–4 and 10–7 Pa, respectively. However, comprehensive studies of the atomic-scale structure have yet to be reported. We investigated the correlations between the atomic-scale structure, the electronic state, and the impurity hydrogen content in a-IGZO by high-energy X-ray diffraction (HEXRD) coupled with reverse Monte Carlo (RMC) modeling and X-ray absorption fine structure (XAFS) spectroscopy. XAFS probed the distribution of unoccupied electronic states above the Fermi level and the local coordination structure around the In, Ga, and Zn atoms. A possible contribution for H from voids in the a-IGZO films was observed by the HEXRD and RMC configuration models. In contrast, the STD 3% film has many voids, which are occupied by impurity H. The proportion of lower coordinated M-O (M = Zn, Ga, In) structures is increased because of the H in voids. We revealed that this electronic and atomic-scale structure of the a-IGZO TFTs, which results in enhanced TFT characteristics, can be stabilized by the hydrogen-passivated defects resulting from STD sputtering with an optimum oxygen flow rate ratio of 3%.

Atomic structure of Ca–Mg biodegradable metallic glass

Very low density eutectic Ca72Mg28 at.% alloy is a precursor of complex biodegradable alloys with potential use as bioresorbable alloy for orthopaedic applications. The structure of the amorphous alloy was investigated by using X-ray, neutron diffraction and reverse Monte Carlo (RMC) modelling. The RMC configuration was decomposed into polyhedral holes whose faces are all triangles consisting of chemical bonds. Free volumes in the respective polyhedral holes were evaluated with reference to the packing efficiency of crystalline CaMg2 HCP phase. The tetrahedral holes, accounting for about 55% of the whole space, are regarded as densely packed units because the average packing efficiency of them is approximately equal to that of the corresponding crystal phase. At the same time, various types of polyhedral holes which have a certain free volume have been observed, and some of them are connected with each other. The densely packed coordination polyhedra consisting only of tetrahedral holes tend to be clustered.

Neutron Pair Distribution Function Study of FePO4 and LiFePO4

Neutron powder diffraction studies of the compounds FePO4 and LiFePO4 are reported. Rietveld refinement of the diffraction data provides averaged structures for both materials that are in good agreement with the published structures. In addition, detailed investigations of the short-range ion-ion correlations within each compound have been performed using the reverse Monte Carlo (RMC) modeling of the total scattering (Bragg plus diffuse) data. Although the short-range structural information for LiFePO4 is consistent with the long-range (averaged) picture, a small, but statistically significant, proportion of the anions is displaced away from their ideal sites within the RMC configurations of FePO4. These anion displacements are discussed in terms of a small concentration of Li+/Fe2+ occupying the empty octahedral sites, probably arising from incomplete delithiation of the LiFePO4 and/or antisite (Li+-Fe2+) defects introduced during the delithiation process.

Free volume distribution and structural inhomogeneity in Ni50V50 amorphous alloy

The structure of Ni50V50 amorphous alloy was investigated by using neutron, X-ray diffraction and reverse Monte Carlo (RMC) modelling. The RMC configuration was decomposed into polyhedral holes whose faces are all triangles consisting of chemical bonds. Free volumes in the respective polyhedral holes were evaluated with reference to the packing efficiency of Ni-V metastable BCC phase. The tetrahedral holes, which account for about 70% of the whole space, are regarded as densely packed units because the average packing efficiency of them is approximately equal to that of the corresponding crystal phase. On the other hand, various types of polyhedral holes, which have a certain free volume, are observed and some of them are connected with each other. The densely packed coordination polyhedra consisting only of tetrahedral holes tend to be clustered.

Inhomogeneity of local packing density and atomic bonding of Ni67Zr33 amorphous alloy

Pulsed neutron diffraction and synchrotron X-ray diffraction measurements were performed on a melt-quenching Ni67Zr33 amorphous alloy sample. The Ni-Ni, Ni-Zr and Zr-Zr nearest neighbour distributions were obtained from differences in the scattering intensities between the two probes in conjunction with the Gaussian fitting. The results demonstrate that the Ni-Zr and Zr-Zr nearest neighbour distributions are respectively represented by single Gaussian peaks but the Ni-Ni one is an asymmetric shape. Three-dimensional structure models were constructed by using reverse Monte Carlo (RMC) modelling method with fitting to the neutron and X-ray diffraction data sets. The Voronoi-Delaunay method was used on the RMC configurations to investigate the local packing properties of each coordination cluster. With reference to the macroscopic packing fraction, inhomogeneous structural feature arising from dense and loose packing domains is observed in the structure of Ni67Zr33 amorphous alloy.

Determination of the structure of liquids containing free radical molecules: Inter-molecular correlations in liquid chlorine dioxide

X-ray diffraction spectra of liquid ClO2 at 204, 223, 248 and 273 K were measured at the BL04B2 beamline in the SPring-8 synchrotron radiation facility. The obtained structure factors are well reproduced by reverse Monte Carlo (RMC) structural modelling. With increasing temperature, the first peak located at around Q = 1.8 Å −1 slightly shifts to higher Q and the tail at the lower-Q side intensifies. In the pair distribution function, g(r), the intensity of the peak at around 3.2 Å that corresponds to the nearest-neighbour distance decreases with increasing temperature. The analysis of the RMC configurations shows that strong directional O–O interaction exists in liquid ClO2 and the instantaneous inter-molecular orientation depends on this interaction.

Static disorder and local structure in zinc(II) isonicotinate, a quartzlike metal–organic framework

Using a combination of Rietveld and reverse Monte Carlo (RMC) refinement of neutron total scattering data, we find that the 10 K structure of zinc(II) isonicotinate shows strong evidence of static disorder. This disorder takes the form of transverse displacements of the isonicotinate ligand and results in elongated atomic displacement parameters and dampening of the experimental G(r). We analyse the RMC configurations using an approach derived from geometric algebra. Complications regarding the inclusion of hydrogenous guest molecules within the pore structure are discussed. This study highlights the way in which structural flexibility can give rise to multiple low-energy ground states in metal-organic framework materials.

Reverse Monte Carlo study of Cu—O bond distortions in YBa2Cu3O6.9

A combination of neutron total scattering measäurement and reverse Monte Carlo (RMC) refinement is applied to the study of apical Cu—O bond distortions in the high-Tc superconductor YBa2Cu3O6.93. We show that the average structure is not consistent with a split-site model for the corresponding Cu and O positions, but that the local structure nevertheless reveals the existence of two separate apical Cu—O bond lengths. Using G(r) data obtained from a variety of Qmax values we show that this result is independent of the data treatment methodology. We also find that the resulting ‘short’ and ‘long’ Cu—O bond lengths agree well with the results of previous äEXAFS studies. The existence of bimodal apical Cu—O bond distributions in the context of a single-site average structure model is interpreted in terms of correlated displacements of the Cu and O atoms. We investigate the possibility of clustering of short apical Cu—O bonds within our RMC configurations.

Structure of Mg50Ni50 amorphous alloy studied by using X-ray diffraction and reverse Monte Carlo modelling

X-ray diffraction measurement was performed on an Mg50Ni50 amorphous alloy prepared by mechanical alloying under an Ar atmosphere. Reverse Monte Carlo (RMC) simulation based on the X-ray diffraction data was used to construct a three-dimensional atomic arrangement of the amorphous alloy. In addition, the local environments around the Mg and Ni atoms were investigated by using Voronoi polyhedral analysis on the RMC configuration. The results show that isosahedral structures are predominant around the Mg and Ni atoms in Mg50Ni50, although prismatic local structures are present around some of the Ni atoms.

Structural study on Zr 0.39Ni 0.61 and (Zr 0.39Ni 0.61)D 0.59 amorphous alloys by neutron and X-ray diffraction

Neutron and X-ray diffraction and reverse Monte Carlo (RMC) simulation were performed on Zr 0.39Ni 0.61 and (Zr 0.39Ni 0.61)D 0.59 amorphous alloys to investigate the rearrangement of metal atoms due to the deuterium absorption. The RMC models showed that the Zr–Zr distances slightly increase but the Zr–Ni and Ni–Ni distances remain virtually unchanged after the deuterium absorption. Additionally, the Voronoi polyhedral analysis of the RMC configurations showed that there is not much difference in the local structure around Zr and Ni between Zr 0.39Ni 0.61 and (Zr 0.39Ni 0.61)D 0.59 amorphous alloys.

Local structure in Ag3[Co(CN)6]: colossal thermal expansion, rigid unit modes and argentophilic interactions

Local structure in the colossal thermal expansion materialAg3[Co(CN)6] isstudied here using a combination of neutron total scattering and reverse Monte Carlo (RMC) analysis.We show that the large thermal variations in cell dimensions occur with minimal distortion of the[Co(CN)6 ] coordination polyhedra, but involve significant flexing of the Co–CN–Ag–NC–Co linkages.We find real-space evidence in our RMC configurations for the importance oflow-energy rigid unit modes (RUMs), particularly at temperatures below 150 K.Using a reciprocal-space analysis we present the phonon density of statesat 300 K and show that the lowest-frequency region is dominated by RUMsand related modes. We also show that thermal variation in the energies ofAg...Ag interactions is evident in both the Ag partial pair distribution function and in the Agpartial phonon density of states. These findings are discussed in relation to thethermodynamic properties of the material.

Ferroelectric nanoscale domains and the905Kphase transition inSrSnO3: A neutron total-scattering study

The $905\phantom{\rule{0.3em}{0ex}}\mathrm{K}$ $Pnma--Imma$ phase transition in $\mathrm{Sr}\mathrm{Sn}{\mathrm{O}}_{3}$ is studied here using a combination of variable-temperature neutron total scattering together with the reverse Monte Carlo (RMC) refinement method. The real-space RMC configurations obtained are analyzed in terms of bond distance and bond-angle distributions, and a geometric algebra approach is used to quantify the associated octahedral-tilting distributions. What emerges from this analysis is that the transition is displacive in nature, in contrast to the results of a recent average-structure investigation in which an order-disorder model was proposed [E. H. Mountstevens et al., Phys. Rev. B 71, 220102(R) (2005)]. Three-dimensional diffuse scattering patterns calculated from the same RMC configurations reveal the existence of an additional disorder mechanism which persists across the $Pnma\text{\ensuremath{-}}Imma$ transition. The ``reflection conditions'' of this diffuse scattering, together with displacement correlation calculations, point to the existence of ferroelectric nanoscale domains within the configurations, which are found to extend across planar regions of approximately $10--15\phantom{\rule{0.3em}{0ex}}\mathrm{\AA{}}$ in diameter.

Effective potentials from complex simulations: a potential-matching algorithm and remarkson coarse-grained potentials

The projection of complex interactions onto simple distance-dependent or angle-dependentclassical mechanical functions is a long-standing theoretical challenge in the field ofcomputational sciences concerning biomolecules, colloids, aggregates and simple systemsas well. The construction of an effective potential may be based on theoreticalassumptions, on the application of fitting procedures on experimental data and on thesimplification of complex molecular simulations. Recently, a force-matching method waselaborated to project the data of Car–Parrinello ab initio molecular dynamicssimulations onto two-particle classical interactions (Izvekov et al 2004 J. Chem.Phys. 120 10896). We have developed a potential-matching algorithm as a practicalanalogue of this force-matching method. The algorithm requires a large number ofconfigurations (particle positions) and a single value of the potential energy for eachconfiguration. We show the details of the algorithm and the test calculations onsimple systems. The test calculation on water showed an example in which asimilar structure was obtained for qualitatively different pair interactions. Theapplication of the algorithm on reverse Monte Carlo configurations was tried as well.We detected inconsistencies in a part of our calculations. We found that the coarsegraining of potentials cannot be performed perfectly both for the structural and thethermodynamic data. For example, if one applies an inverse method with an input ofthe pair-correlation function, it provides energetics data for the configurationsuniquely. These energetics data can be different from the desired ones obtainedby all atom simulations, as occurred in the testing of our potential-matchingmethod.

Structure of amorphous LaNi 5D 3.3 studied by neutron and X-ray diffraction

In situ neutron and X-ray diffraction measurements in a D 2 gas atmosphere at a constant pressure of 1 MPa were performed on amorphous LaNi 5D 3.3, which was prepared by mechanical alloying in a D 2 gas atmosphere. Reverse Monte Carlo (RMC) simulation based on the neutron and X-ray diffraction data was applied to construct a three-dimensional atomic arrangement of this amorphous alloy. The RMC model shows that more than 90% of the deuterium atoms occupy tetrahedral sites consisting of La and/or Ni atoms. Furthermore, the local environments around the Ni and La atoms were investigated by Voronoi polyhedral analysis of the RMC configuration of the metal atoms. The results show the presence of a number of prismatic-like polyhedra around a Ni atom.

Atomic configuration of metal atoms in (Ti 0.676Zr 0.324)D 0.31 metallic glass studied by X-ray, neutron diffraction and reverse Monte Carlo modeling

We have obtained a three-dimensional atomic configuration in (Ti 0.676Zr 0.324)D 0.31 glass using the reverse Monte Carlo (RMC) modeling method with neutron and X-ray diffraction data. Voronoi polyhedral analysis for the RMC configuration of metal atoms revealed the presence of a lot of icosahedral-like clusters.

Structural analysis for crystalline and amorphous RFe 2D x (R: Ho, Tb) by X-ray/neutron diffraction and reverse Monte Carlo modeling

Atomic structures of crystalline (c-)HoFe 2D 3.6 and amorphous (a-)HoFe 2D 3.0 prepared by deuterium absorption of the C15 Laves phase compound HoFe 2 were studied by X-ray and neutron diffraction techniques and these were compared with previous results for c- and a-TbFe 2D x . The structural parameters such as an interatomic distance and a coordination number are qualitatively in agreement with those for c- and a-TbFe 2D x . Namely, a marked difference was observed in the arrangement of metal atoms in a-HoFe 2D 3.0 compared to that in c-HoFe 2D 3.6 from interatomic distances and coordination numbers and there are clusters of Fe and Ho atoms in a-HoFe 2D x . RDF( r)s observed by the neutron diffraction indicated that the D atoms occupy tetrahedral sites consisting of 2Ho+2Fe in c-HoFe 2D 3.6, while the D atoms occupy selectively tetrahedral sites consisting mainly of Ho atoms in a-HoFe 2D 3.0 because of strong affinity between D and Ho atoms. The reverse Monte Carlo (RMC) modeling method was employed to visualize a three-dimensional atomic arrangement by using combined X-ray and neuron diffraction data of RFe 2D x (R: Ho, Tb). Chemical fluctuation was visually observed in the RMC configurations. Moreover, it was found that about 40% of the D atoms occupy sites consisting of 4R.

Investigation of the uniqueness of the reverse Monte Carlo method: Studies on liquid water

Reverse Monte Carlo (RMC) simulation of liquid water has been performed on the basis of experimental partial pair correlation functions. The resulted configurations were analyzed in various aspects; the hydrogen bond angle distribution, three body correlation and orientational correlation were calculated. The question of the uniqueness of the RMC method was also examined. In order to do this, two conventional computer simulations of liquid water with different potential models were performed, and the resulted pair correlation function sets were fitted by RMC simulations. The resulted configurations were then compared to the original configurations to study how the RMC method can reproduce the original structure. We showed that the configurations produced by the RMC method are not uniquely related to the pair correlation functions even if the interactions in the original system were pairwise additive. Therefore the difference between the original simulated and the RMC configurations can be a measure of the uncertainty of the RMC results on real water. We found that RMC produces less ordered structure than the original one from various aspects. However, the orientational correlations were reproduced rather successfully. The RMC method exaggerates the amount of the close packed patches in the structure, although these patches certainly exist in liquid water.

Molecular dynamics applied to a reverse Monte Carlo configuration for amorphous Ni 81B 19

The reverse Monte Carlo (RMC) method supplies atomic configurations that are in almost perfect agreement with experimental diffraction data. A serious drawback of the method, however, is the unphysical process by which the atoms are positioned. In this paper an RMC configuration for amorphous Ni 81B 19 is investigated by molecular dynamics (MD) simulations, a study that at the same time consitutes a test for the potentials used. It is found that one has to rescale the energy scales of the potentials in order to obtain satisfactory radial distribution functions for the MD configurations, both when the RMC configuration is used as a starting point and when the system is first melted and quenched. Most importantly, the atomic-level information for these MD configurations is very similar to that supplied by RMC calculations. The present results may seen as a cross-validation of both computer simulation techniques.