Search

Apparent superlattice reflections obtained in electron diffraction patterns of modulated calcite and dolomite have been ascribed to local domains of various different metastable Ca-Mg ordering schemes that are not known as macroscopic phases. We show that the type "c" reflections in diffraction patterns of supposed superstructures can in fact be produced by superposition of diffraction from the host crystal and that from domains that are in an orientation related to the host by twinning on (104). From details of the additional reflections present, we deduce that the carbonate anions are orientationally disordered in the twin nanodomains, which have the R3m space group of high-temperature disordered calcite. This twinning can explain the diffraction ascribed to type "γ"/"μ" /"ν" superstructures, and resolves controversies over the occurrence of these purportedly different superstructures. The relationships between composition, orientational order of the carbonates, molar volume and known macroscopic structures, and the possibility of interfacial strain reduction by static disorder in the twin domain, are discussed. We stress the importance of checking for presence of twinned nanodomains using microdiffraction before attributing an apparent superstructure modulation to local cation ordering. High-resolution imaging may not be diagnostic, since overlap of small domains and host matrix can give Moire patterns that resemble superlattice fringes.

The commensurate superstructures of a NiAs/Ni2In type parent structure, Ni3.32InTe2 and Ni3.12In0.86Te2.14 (q = gamma[0 0 1]*, gamma = 2/3) as well as one dimensionally incommensurate structure of Ni31nTe2 (gamma = 0.71) were refined from neutron powder diffraction data (R-wp = 4.77%, 4.53% and 4.91 % for the three structures, respectively, at 298 K). The commensurate structures were refined in the P6(3)/mmc space group (c = 3c(NiAs)). The stacking sequence at the hcp array is -In/Te/Te/- and the trigonal bipyrainidal site within the In layer, Ni(2), is partially occupied while it is empty in the Te layers. The octahedral position in between the In and Te layers, Ni(1a), is fully occupied while the octahedral position in between two adjacent Te layers, Ni(1b), is partially occupied. With decreasing In and Ni content, the modulation wave vector,, was found to increase continuously until gamma = 1. From this, crenel functions to describe the whole homogeneity range of the solid solution were constructed with the length of the atomic domains Delta(Te) = gamma (and hence Delta(In) = Delta(Ni) = 1-gamma) and Delta(Ni(1b)) = gamma/2 (and hence Delta(Ni(1a)) = 1-gamma/2) which were then used for the refinement of the incommensurate structure of Ni3InTe2. The corresponding effect in real space is that the single In layers separating double layers of Te, occur less frequent when y in increasing until at gamma = 1 the CdI2 type structure of Ni1+xTe2 is reached.

We explore the use of tetraethoxysilane (TEOS) as a silica source for the formation of carbonate-silica composite materials known as ‘biomorphs’. The basic hydrolysis of TEOS furnishes silica in a controllable fashion, allowing a significantly higher reproducibility of the obtained silica–barium and silica–strontium carbonate co-precipitates compared to commercial water glass silica used so far. We further discuss the influence of ethanol used as a co-solvent on the morphologies of biomorphs, which are examined by optical microscopy, field emission scanning electron microscopy (FESEM) and energy dispersive X-ray analysis (EDX).

The ( 3 + 1 ) -d incommensurately modulated structures of four members of the NiGe 1 - x P x solid solution field have been successfully refined from X-ray powder diffraction data ( R w ( all ) / R wp = 2.13 / 3.92 ; 1.52/4.25; 1.27/3.44 and 2.00/4.03 for x = 0.4 , 0.5, 0.6 and 0.7, respectively). The 4-d superspace group symmetry is Amam ( 0 0 γ ) s 0 0 ( Z = 4 ; a = 5.0468 ( 2 ) , 5.0188(2), 4.9796(2) and 4.9651 ( 1 ) A ; b = 6.0636 ( 3 ) , 6.0576(2), 6.0183(2) and 6.0031 ( 1 ) A ; c = 3.4877 ( 2 ) , 3.4812(2), 3.4593(1) and 3.45442 ( 7 ) A ; γ = 0.7769 ( 2 ) , 0.7467(1), 0.7241(1) and 0.7046(1) for x = 0.4 , 0.5, 0.6 and 0.7, respectively). The underlying average structure is of NiAs type while the (in general) incommensurate primary modulation wave-vector, γ c * , varies continuously and smoothly with composition. The two largest amplitude displacive atomic modulation functions (AMFs), for all samples, were the Ni displacement along b AMF and the Ge/P displacement along a AMF. The refined amplitude of the former was found to systematically increase with P content from 0.215 A for NiGe 0.6 P 0.4 to 0.294 A for NiGe 0.3 P 0.7 while the magnitude of the latter was found to increase with P content from 0.177 A for NiGe 0.6 P 0.4 to 0.253 A for NiGe 0.3 P 0.7 . These displacive shifts significantly modulate the local crystal chemistry i.e. the local interatomic distances and co-ordination polyhedra. This continuously variable, incommensurately modulated, intermediate structure type is shown to provide a natural link or bridge between the two extreme end-member structures i.e. NiGe (of MnP structure type) and NiP by simply choosing the commensurate options with γ = 1 and 1 2 respectively.

The crystal structures of two low temperature B82 related superstructure phases in the nickel indium system, Ni13In9 and Ni7−δIn3, have been refined and described in relation to the underlying parent Ni2In structure-type. They are isostructural to In13Pt9 and Cu7In3, respectively and both accommodate the change in stoichiometry from Ni2In by forming “same atom” structural motifs, In-centred indium octahedral units in the case of Ni13In9 and trigonal bipyramidal nickel units in the case of Ni7−δIn3. A sample with the nominal composition Ni5In3, reported to form part of a solid solution in the binary phase diagram, is not stable at 400 °C and forms a two-phase mix of Ni7−δIn3 and Ni13In9.

The wide-range of non-stoichiometric NiAs-type solid solution Ni{sub 1+}/-{sub y}Ge{sub 1-x}P{sub x} has been studied by means of X-ray powder and electron diffraction. The incommensurately modulated structure of Ni(Ge, P) has been found to exist over a wide compositional range which is limited by the end points {approx}NiGe{sub 0.8}P{sub 0.2} and NiGe{sub 0.3}P{sub 0.7} so that the general stoichiometry might be referred to as NiGe{sub 1-x}P{sub x} with 0.2=

We describe self-assembled silica-carbonate aggregates that show a diverse range of morphologies, all of which display complex internal structure, orientational ordering of components, and well-organised, curved global morphologies that bear a strong resemblance to biogenic forms. The internal order is described as a liquid-crystal-like organisation of colloidal particles. We discuss possible causes for the striking morphologies of these inorganic materials, including local nanocrystal packing constraints and global silica membrane templating.

The crystal structures of Co3Se4 and Co6.8Se8 have been refined from twinned single crystal and powder X-ray diffraction data. The refined structures are isostructural to the Cr3±xSe4 and Cr7Se8 types, respectively. The structures are described as superstructures of the Cd(OH)2-type structure, where the empty layers of unoccupied octahedra are replaced with alternating Co filled and empty [100] rods and Co filled Kagome nets for Co3Se4 and Co6.8Se8, respectively. The structure of Co3Se4 is not a substructure of Co7Se8 and the path from Co7Se8 to Co3Se4 by removing 1/4 of the Co atoms in the Kagome nets involves both gain and loss of symmetry. Yet the refinements of Co6.8Se8 suggest an extended homogeneity field.

Clear evidence (in the form of structured diffuse scattering) is found for short range ordering of metal ions and associated induced structural relaxation in two members of the cubic BZN pyrochlore (Bi 1.5− α Zn 0.5− β )(Zn 0.5− γ Nb 1.5− δ )O (7−1.5 α − β − γ −2.5 δ ) solid solution. An average neutron powder diffraction structure refinement is carried out for one of these. Electron probe micro-analysis suggests that the primary mechanism for non-stoichiometry in cubic BZN is the removal of ZnO from the nominally fully occupied (Bi 1.5 Zn 0.5 )(Zn 0.5 Nb 1.5 )O 7 end-member. A detailed bond valence sum analysis of a recently reported average crystal structure is used to suggest possible local cation ordering schemes along with the induced displacive relaxation that is likely to accompany such local cation ordering. The observed diffuse distribution is qualitatively interpreted via Monte Carlo modelling.

Selected-area electron diffraction revealed that Ni10Sn5P3 forms a superstructure of the NiAs-type structure. All observed reflections could be fully indexed as H=G+mq, where G represents the reflections of the underlying hexagonal B8-type average structure, m is an integer and q= 1 8 [ 3 122]*. The resulting space group is P 1 (Z=2). The crystal structure was refined using single-crystal X-ray diffraction methods on a twinned specimen (1651 observed unique reflection, Rw obs=0.0391). The superstructure reflections are caused by ordering of Sn and P atoms on the hcp array coupled to Ni ordering in 1 4 of the trigonal bipyramidal sites. The unit-cell parameters refined from powder X-ray diffraction data are a=6.419(2) A, b=8.322(3) A, c=10.278(3) A, α=73.32(4)°, β=84.41(3)°, γ=82.71(2)°, V=520.60(5) A3. High-resolution electron microscopy images revealed a well-ordered material although macroscopic crystals are twinned.

The title compound was synthesized according to Andersson Acta Chem. Scand. 18 (1964) 2233. The structure is an incommensurately modulated variation of the UO3 structure, in a (3+1)-dimensional super space description isostructural to the composite phases (1−x)Ta2O5•xWO3 Acta Crystallogr., Sect. B 52 (1996) 223 and Acta Crystallogr., Sect. B 51 (1995) 709. The compound crystallizes in the orthorhombic system with cell parameters 3.936(3), 6.153(2), 3.656(1) A. The superspace group is X:Pmmm(00γ):1s 1 , where X refers to a non-standard superspace centering with the lattice vector (0 1 2 1 2 1 2 ) , and γ=1.61 is the modulation wave vector.

The reported high temperature, wide range non-stoichiometric Co2–xSe2, 0⩽x⩽0.50, solid solution field has been carefully re-investigated using electron and X-ray powder diffraction. Close investigation shows that the ‘solid solution’ should be separated into four distinct phase regions (i)–(iv), even though the underlying average structure unit cell appears to distort smoothly and continuously as a function of composition. For values of x between 0 and 0.06 (structure type (i)), the structure is of ‘defect’ NiAs-type as only NiAs type parent reflections are observed. For 0.06

The crystal structures of La2O2CO3 II and Nd2O2CO3 II have been shown by means of high-resolution powder neutron (PND) and synchrotron X-ray diffraction (SXRD) combined with selected area electron diffraction (SAED) studies to be far more complex than earlier anticipated, owing to ordering of carbonate groups between (Ln2O2+2)n layers. In contrast to earlier descriptions, the carbonate groups appear to be rather regular. Relative to an average model, the SAED patterns show additional scattering in the form of closely distributed, but essentially discrete, spots along 〈 1 3 , 1 3 , l〉*. Most of the observed scattering, H, can be described as H=G±mq1+nq2, where G is the Bragg reflections of the underlying average P63/mmc lattice, q1=[ 1 3 , 1 3 , ± 1 2 ]*, q2=[ 1 3 , 1 3 , ± 2 3 ]*, and m and n are integers. The additional scattering reflects ordering of the carbonate groups into trigonal layers between the (Ln2O2+2)n layers, but it remains open whether q1 and q2 represent two separate structures with different stacking sequences of such layers or whether they correspond to an even more complex stacking sequence. In any case, some disorder and rotational domain twinning are present. Two structure models, one for each modulation wave vector, were constructed. Rietveld-type refinements of PND data of La2O2CO3 II were performed, approximating the complex, and at least partly disordered, stacking sequence as a two-phase mixture of the two modulated phases. Satisfactory convergence was achieved with Rp=6.4%, Rwp=8.3%, and χ2=3.32. The isothermal expansivities, αp, for La2O2CO3 II and Nd2O2CO3 II between 298 and 893 K were determined as 2.92×10−5 and 2.70×10−5 K−1, respectively.

The crystal structures of two new ternary, incommensurately modulated phases in the Cr/Mo/Ge system, Cr1-xMoxGe∼1.75withx= 0.65 and 0.84, have been refined using the same composite approach as for the earlier reported compounds, (Mo/Rh)11Ge18and (Mo/Rh)13Ge22, from the Mo/Rh/Ge system. The composite model consists of a rather stable diamond-like transition metal substructure and a helical main group element arrangement. The same super space group pair,W:P4/nnc: q1̅q1P:I41/amd: 11̅ss, the same positions and modulation waves were used. The refinements of the new phases in the Cr/Mo/Ge system converged rapidly toR-values below 5% with GOF of 2.53 and 3.71, respectively. Each structure is built by two sub lattices with the parameters -a,cxand the modulation vectorqx- for Ge anda,cTand qTfor the transition metals. The refined data forx= 0.65 are found to bea= 5.935(2),cx= 2.770(1),cT= 4.839(1),qx= 0.428 [001]*,qT= 0.253 [001]* and forx= 0.84,a= 5.963(1),cx= 2.786(1),cT= 4.875(1),qx= 0.430 [001]*,qT= 0.264 [001]*. It is clearly shown the composite approach is applicable to different chimney-ladder compounds, even though they have quite different compositions and henceq-vectors.

The growth of TaC, TiC and TaxTi1-xC whiskers has been studied in some detail. The whiskers were synthesised via a vapour-liquid-solid (VLS) growth mechanism in the temperature range 1220-1400 degr ...

Tegengrenite, forming octahedra up to 150 μm in size, occurs associated with hausmannite, calcite, brucite, dolomite, clinohumite, kinoshitalite, native copper, barytocalcite, bindheimite, and cerussite in Mn ore from Jakobsberg, Filipstad district, Varmland, Sweden. The mineral is deep red and translucent, with a sub-adamantine luster. In reflected light it is gray and nearly isotropic, with measured reflectance values ( R %) 10.4 (λ= 470 nm), 10.0 (546 nm), 9.9 (589 nm), and 9.8 (650 nm). The refractive index at λ = 589 nm is 1.92(2). There is no cleavage; fracture is conchoidal. D calc is 4.58(1) g/cm 3 . Electron-microprobe analyses gave (average of 35 points; in wt%) MgO 21.83, Al 2 O 3 0.76, SiO 2 1.70, Sb 2 O 5 36.13, TiO 2 1.40, Fe 2 O 3 0.78, MnO 25.74, Mn 2 O 3 (calculated from stoichiometry) 8.14, ZnO 2.66, sum 99.14, yielding the empirical formula \(Mg_{1.22}^{2+}Mn_{0.82}^{2+}Zn_{0.07}(Sb_{0.50}Mn_{0.23}^{3+}Si_{0.06}Ti_{0.04}Al_{0.03}Fe_{0.02})O_{4}\) . Combined X-ray and electron diffraction studies show that tegengrenite is rhombohedral, space group \(\mathit{R}\overline{3}\) or R 3, and pseudocubic. The individual tegengrenite octahedra consist of eight twin domains that give rise to complex diffraction patterns and make them unsuitable for a conventional structure determination. Tegengrenite, like the chemically related mineral filipstadite, bears a close structural relationship to spinel. The eight strongest lines in the X-ray diffraction pattern are [ d (in angstroms) ( I / I 0 )( hkl )]: 4.98(20)(211, 003), 4.32(19)(122), 4.24(18)(113), 3.052(33)(140, 214), 2.608(100)(241, 143, 125), 2.162(28)(244), 1.665(30)(363, 075), 1.531(26)(820) and 1.527(29)(428). The refined unit-cell parameters (hexagonal setting) are a = 16.196(1) and c = 14.948(2) A with Z = 42; for the cubic spinel-type subcell \(\mathit{A}\ =\ \mathit{c}/\sqrt{3}\ =\ 8.63\) A. The new species is named for Felix Tegengren (1884–1980).

Two fluorite-related modulated phases (α′1 and α″1) occurring in the (1−x)12Bi2O3·xCaO, 0.20

The oxide Mn{sub 0.6}Ta{sub 0.4}O{sub 1.65} has been synthesized in air by rapidly cooling a melt from 1,400 C, and it has been studied by X-ray powder diffraction (XRPD), selected-area electron diffraction (SAED), high-resolution electron microscopy (HREM), thermal analysis, and measurements of magnetic susceptibility and electrical conductivity. The average structure is that of a cubic fluorite with a = 4.9826(3) {angstrom} and 18% vacancies on the O atom sites. Prominent diffuse scattering is present in its electron diffraction patterns (EDPs) and a structural model for this scattering is proposed. It is based on ca. 10 {angstrom} large microdomains with a pyrochlore type structure which are separated by antiphase boundaries. The magnetic susceptibility shows an antiferromagnetic interaction between the Mn{sup 2+} ions and an effective number of Bohr magnetons that increases with temperature. When heated in air at 500 C, the compound is oxidized to the composition Mn{sub 0.6}Ta{sub 0.4}O{sub 1.75}, while retaining a cubic fluorite structure with a = 4.8380(4) {angstrom}. This latter phase is a semiconductor with an activation energy of 0.64 eV. 36 refs., 10 figs.

Epitaxial intergrowths of Fe 6 Ge 5 and B 8-type Fe 1+ x Ge was investigated by means of electron diffraction. A crystal structure description of Fe 6 Ge 5 in terms of B 8- and B 1-type structures was established; 4/5 of the Ge atoms in Fe 6 Ge 5 can be described with one-octahedron wide slabs of hexagonal close-packed Ge atoms with all octahedral sites and 1/2 of the trigonal bipyramidal sites therein filled with Fe atoms ( B 8). Additionally, all Ge atoms can be described by a two-octahedra wide slab of cubic close-packed Ge atoms with 4/6 of the octahedral sites filled with Fe atoms and 1/6 of the octahedral sites filled with two Fe atoms, one in each square pyramid ( B 1). At the surface interconnecting these slabs, octahedra from the B 8 and B 1 slabs share faces and the trigonal bipyramid from the hcp array and two edge-sharing tetrahedra from the ccp array are merged to form a pentagonal bipyramid.

The alloys (Co, Ni)1+x(Ge, Sn) form a range of superstructures in which one [1 1 0] repeat of the hexagonalB8 substructure is preserved. Less well-ordered phases also occur in which continuous sinusoidal loci of diffuse scattering are observed trending parallel to c*. The curves can be modeled as cosine waves with maximumkatl=even and minimumkatl=odd. The shape of the curves vary with composition and annealing temperature. Computer simulation was used to generate two-dimensional projections of real-space occupancy patterns that produced similar diffraction patterns. The synthesized real-space arrays were characterized by sets of correlation coefficients. A Monte Carlo algorithm was then used to find sets of two-body interaction energies for which these structures lay at an energy minimum. Good fits between calculated and experimental diffraction patterns were found in all cases. The fitted interaction energies were mainly positive, implying that most two-body interactions were repulsive between sites of like occupancy. Magnitudes were significant out to third-nearest neighboring interstitial sites. The magnitudes tended to be largest nearx=0.5. Additional variations of both interaction energies and resulting correlations with composition and annealing temperature are discussed. It is shown that the double-locus diffraction pattern observed for the Ni–Ge system is not necessarily produced by a mechanical mixture of two structures, but can correspond to a single phase. InteractionsEijout tonth nearest neighbors include a larger number of symmetrically distinct 〈i,j〉 terms than the corresponding 〈u,v,w〉 terms in three dimensions, implying that only approximate three-dimensional energies can be obtained by fitting from the energies of this study. Mutual frustration of repulsive interactions on the interstitial sublattice, which has a large number of triangularly connected neighbors, is responsible for both breaking the hexagonal symmetry of the sublattice and the failure to form structures giving conventional “spot” diffraction patterns.

A detailed TEM and XRD study has been made of the basic crystallography (unit cells, space group symmetries, and twin relationships) of each of the three oxidized phases which occur in the CeTaO 4+ δ system, their structural relationship to stoichiometric Ce +III TaO 4 , and their temperature-dependent redox reactions. Such crystallographic knowledge is essential to understand the structural relationships between the various phases and to gain insight into the oxidation/reduction mechanisms allowing the formation of the oxidized phases. Twinning is found to be endemic in stoichiometric Ce +III TaO 4 as well as in each of the oxidized Series 2, 3, and 4 phases; the twin plane relating the twin variants is derived in each case.

The compounds (Co, Ni)(Ge, Sn) and Cu1+xhave hexagonal average structures intermediate between those of NiAs and Ni2In. Ordering of atoms and vacancies in the partially occupiedA′ sites produces a variety of superstructures of lower symmetry. In the structures of this study, the modulation vector is always in the (hhl) plane. Diffuse scattering observed in some cases implies the occurrence of short-range order only. The effect of two-site interactions on determining the preferred long-range and short-range ordering patterns was investigated by using a Monte Carlo simulation technique to generate two-dimensional arrays of "atoms" and "vacancies" and calculating the corresponding scattering patterns. It is shown that many of the 18 symmetrically distinct first- and second-nearest-neighbor interaction energies are nonzero in these systems but that the preferred ordering pattern is in most cases determined by variation in only three second-nearest-neighbor terms. These three parameters vary with both composition and temperature. The undulating diffuse streaks observed in electron diffraction in the Sn-bearing and Co-Ge systems are shown to arise from a combination of negligible interactions along z, strong interactions in the xy plane, and weak diagonal interactions.

The body centered cubic (a∼4.26 A)γ-type solid solution (ssγ) phase (1−x)Bi2O3·xCaO (∼0.50

Electron diffraction is used to investigate Sn/P and interstitial Ni atom ordering in the apparent wide range, non-stoichiometric Ni{sub 1+m}Sn{sub 1{minus}x}P{sub x} ternary B8-type solid solution. Two complex modulated structures are found to exist in a limited area of the solid solubility range, described by compositions around Ni{sub 1.1}Sn{sub 0.7}P{sub 0.3}. This region of the high temperature Ni{sub 1+m}Sn{sub 1{minus}x}P{sub x}, B8-type phase field had previously been found to separate out as an isolated B8 type phase region at low temperatures. A complex, long range ordered triclinic superstructure phase is found for specimens quenched from lower temperatures and a plausible structural model, based on ordering of NiP{sub 3}Sn{sub 2} trigonal bipyramidal clusters, proposed. The underlying B8 subcell remains metrically hexagonal so that there are at least 12 distinct twin variants whose B8 sublattice reflections overlap exactly. The diffraction patterns characteristic of specimens quenched from higher temperatures are compatible with a multiple twinned microdomain model.

TheB8-type structure field of the Mn–Sn system has been investigated. Two high temperature phases (HTP1 and HTP2) and one low temperature phase (Mn3Sn2) were found. They all crystallize with the NiAs structure type with part of the trigonal bipyramidal interstices filled by manganese atoms in an ordered manner. The ordering as well as the manganese content is different for the three phases, giving rise to three different orthorhombic superstructures. Mn3Sn2seems to have the lowest manganese content, since the corresponding basal unit cell is smaller than for HTP1-2. Structural models of the phases are based on selected area electron diffraction, X-ray powder diffraction, and preliminary single crystal X-ray measurements. The ideal cell parameters found are (a=7ahex, b= 3a hex ,c=chex), (a=5ahex, b= 3a hex ,c=chex), and (a=2ahex, b= 3a hex ,c=chex) for HTP1, HTP2, and Mn3Sn2, respectively. The crystal structure of Mn3Sn2has been refined by means of the Rietveld method from X-ray powder diffraction data. Mn3Sn2is orthorhombic,Pnma,a=7.5547(2),b=5.4994(2),c=8.5842(2) A,Z=4. (Pbnmin the setting above.) The compound is isostructural with Ni3Sn2andγ′-Co3Sn2(H. Fjellvag and A. Kjekshus,Acta Chem. Scand.A40, 23–30 (1986)). FinalRp=8.97%,Rwp=11.44%, GOF=2.86, andRBragg=4.11% using 43 parameters and 5701 observations and 330 Bragg reflections.

The high temperature modification of η-Cu6Sn5 (reassigned η-Cu5Sn4) was by means of electron diffraction found to exhibit structural modulations. The diffraction patterns may be indexed using the basic NiAs-cell and a modulation vector q ≈ (δ a */24, δ a */24, c */2), 8 ≤ δ ≤ 9. Models for the crystal structures of the two commensurate modulations δ = 8 and δ = 9 are suggested. Both superstructures are caused by the ordered occupancy of Cu in the trigonal bipyramidal sites. The model given for η 6-Cu5Sn4 (corresponding to δ = 8, found above 350° C) has the space group C2 and cell parameters a = 12.60 Å, b = 7.27 Å and c = 10.20 Å and β = 90.0°. η 8-Cu5Sn4, corresponding to δ = 9, has the unit cell (in P21/c) a = c = 9.83 Å, b = 7.27 Å and β = 62.5°.

This survey covers the NiAs-Ni 2 In-type structures and ordered superstructures in the T-B systems. Only compounds with a c/a value between 1.2 and 1.3 are treated. Twenty-three reports of superstructures were found, exhibiting 16 different unit cells; 14 of these structures are reported as solved and they form 11 distinct structure types.

Crystals of two new phases of the compositions NiCr5Sn12 and NiCr3Sn8 were grown from a 660 °C tin melt with Ni and Cr present in equal amounts. Crystals of the new composition CrSn2 were grown from a tin melt with Cr present. The crystal structures were determined. The structure of NiCr5Sn12 (space group no. 180, P6222, a = 5.4944(4), c = 13.883(5), Z = 1) belongs to the NiMg2 structure type. NiCr3Sn8 crystallizes in the space group I4/mcm (no. 140, a = 6.6092(3), c = 5.3460, Z = 1) and is isostructural with CuAl2. CrSn2 is isostructural with CuMg2. The three structures NiCr3Sn8, CrSn2 and NiCr5Sn12 form a series: they are built up by different stackings of the same layer consisting of rows of tetragonal antiprisms. Letting A, B and C denote three orientations of this layer, each differing from the other two by a rotation of 2π 3 , NiCr3Sn8 is a simple AA′A sequence, CrSn2 is an ABA′B′A stacking and NiCr5Sn12 corresponds to the sequence ABCA. Adjacent layers are connected via hexagons, and the structure models may be idealized with respect to this figure. The parameters of the idealized structures agree well with the real structures.

The cell of the low-temperature modification of Cu 6 Sn 5 , the η-phase, has been determined by means of electron diffraction of single domains. The basic hexagonal NiAs (B8) type cell is pentupled, reflected in the reciprocal lattice by a fivefold superstructure running along [1121]. The extensive domain-twinning macroscopically gives rise to perfect hexagonal symmetry, which explains the previously proposed hexagonal cell [a=5a NiAs , c=5c NiAs ; Bernal (1928), Nature, 122, 54]. The structure was solved by occupying every fifth trigonal bipyramidal site of the NiAs-type structure and shifting the atoms surrounding these additional Cu atoms to form Edshammar 11 polyhedra with Cu-Sn and Cu-Cu distances in the range 2.60-2.74 A

The title compound was synthesized according to Andersson Acta Chem. Scand. 18 (1964) 2233. The structure is an incommensurately modulated variation of the UO3 structure, in a (3+1)-dimensional super space description isostructural to the composite phases (1−x)Ta2O5•xWO3 Acta Crystallogr., Sect. B 52 (1996) 223 and Acta Crystallogr., Sect. B 51 (1995) 709. The compound crystallizes in the orthorhombic system with cell parameters 3.936(3), 6.153(2), 3.656(1) A. The superspace group is X:Pmmm(00γ):1s 1 , where X refers to a non-standard superspace centering with the lattice vector (0 1 2 1 2 1 2 ) , and γ=1.61 is the modulation wave vector.

The structural and associated ferroic properties of the stuffed tridymite-type compounds ${\text{BaAl}}_{2}{\text{O}}_{4}$ and ${\text{SrAl}}_{2}{\text{O}}_{4}$ have been investigated by means of ab initio calculations. Structures and energy landscapes have been analyzed in terms of symmetry-adapted distortion modes. Despite their rather different room-temperature symmetries, a triply-degenerate unstable antiferrodistortive rigid unit mode (RUM) of the ${\text{Al}}_{2}{\text{O}}_{4}$ tetrahedral framework is shown to be the dominant instability for both compounds. An orthorhombic configuration resulting from a single wave $(1\mathbf{q})$ distortion mode competes with a hexagonal configuration resulting from three superposed such waves $(3\mathbf{q})$. The very small energy difference between these two configurations in the case of ${\text{BaAl}}_{2}{\text{O}}_{4}$ would explain the recent electron microscope observation of orthorhombic symmetry on the nanoscale. A second unstable mode of polar character (also a RUM for the tetrahedral framework) is also present. While this second instability is too weak to condense in ${\text{BaAl}}_{2}{\text{O}}_{4}$, in the case of ${\text{SrAl}}_{2}{\text{O}}_{4}$ it is fundamental to make the $1\mathbf{q}$ configuration prevail. The condensation of this additional instability is the cause of the symmetry reduction in ${\text{SrAl}}_{2}{\text{O}}_{4}$ to monoclinic. In contrast with previous literature, ${\text{SrAl}}_{2}{\text{O}}_{4}$ is a proper ferroelectric and a pseudoproper ferroelastic, with an uncommon bilinear coupling between its spontaneous polarization and shear monoclinic strain. One expects direct switching of spontaneous polarization through a shear stress, and conversely switching of the ferroelastic spontaneous shear strain through an electric field. This property must be related to the elasticoluminescent and electroluminiscent properties reported in Eu-doped ${\text{SrAl}}_{2}{\text{O}}_{4}$. The calculated energy maps have been systematically compared with variations in the so-called global instability index, introduced within the empirical bond valence model. The similarity of the variation in both quantities (for the most unstable distortion modes) reported in other systems is also observed here.

The P 6 3 ( a =2 a p , b =2 b p , c = c p ) crystal structure reported for BaAl 2 O 4 at room temperature has been carefully re-investigated by a combined transmission electron microscopy and neutron powder diffraction study. It is shown that the poor fit of this P 6 3 ( a =2 a p , b =2 b p , c = c p ) structure model for BaAl 2 O 4 to neutron powder diffraction data is primarily due to the failure to take into account coherent scattering between different domains related by enantiomorphic twinning of the P 6 3 22 parent sub-structure. Fast Fourier transformation of [0 0 1] lattice images from small localized real space regions (∼10 nm in diameter) are used to show that the P 6 3 ( a =2 a p , b =2 b p , c = c p ) crystal structure reported for BaAl 2 O 4 is not correct on the local scale. The correct local symmetry of the very small nano-domains is most likely orthorhombic or monoclinic.

The precipitation of calcium carbonate in alkaline silica solutions results in the formation of complex curvilinear forms if aragonite formation is encouraged by growth at an elevated temperature (80 °C). The resulting coralline self-assembled silica-calcium carbonate particles are "biomorphs", bearing a striking resemblance to natural coral forms. These materials, comprised of calcium carbonate nanocrystals and an amorphous silica matrix, have a complex ultrastructure, made of clusters of gathered sheets of variable curvatures formed by successive curling. The nanocrystals within these "ruled surfaces" are thin, elongated, densely packed needles of aragonite. These clusters are outgrowths from central saddlelike cores that resemble developable petaloid surfaces. The size, shape, crystallography, and chemical composition of the resulting biomorphs were examined by optical microscopy, field emission scanning electron microscopy (FE-SEM), powder X-ray diffractometry (XRD), Fourier transform infrared spectroscopy (FT-IR), transmission electron microscopy (TEM and HRTEM), and energy dispersive X-ray analysis (EDX).

The possibility of pseudofossils is a well-known obstacle to the identification of fossilized microorganisms (Cloud, 1973; Westall, 1999) and distinguishing abiotic from biotic origins is still a hotly debated topic (Dalton 2002; Schopf et al., 2002; Brasier et al., 2002). Here we report silica-carbonate aggregates, so-called ‘biomorphs’ that mimic—both morphologically and chemically-primitive microfossils.

The crystal structure of Ni13Sn8P3 was elucidated from High Resolution Electron Microscopy Images. It was found to be a superstructure of the B8-type (NiAs-type) structure with Sn and P atoms ordered at the hexagonally closepacked array and Ni atoms in all octahedral and in two out of every eleven trigonal bipyramidal sites. The structure motif within a NiSn B8-type matrix comprises triangles of P atoms with two of the three edges capped by Ni atoms in trigonal bipyramidal sites. This motif is repeated along the [0 1 1]B8 zone axis so that the structure can be envisaged as pairs of face-sharing Ni centered Edshammar polyhedra corner-connected along [0 1 1]B8. The reciprocal lattice ( H ) can be described as H = G +m q 1 +n q 2 (where G refers to the Bragg reflections of the underlying B8-type structure, q 1 =1/11 [11 2 1]B8, q 2 =1/22 [ 6 515]B8 and m and n are integers. The resultant triclinic ( P 1 ) unit cell parameters are a=6.456 A, b=21.291 A, c=13.247 A, α=81.052°, β=56.260° and γ=68.221°.

N-Nb2O5 [C2/m (No. 12), a = 28.51, b = 3.830 and c = 17.48 Å, and β = 124.8°] has been investigated by means of selected-area electron diffraction (SAED) and high-resolution transmission electron microscopy (HRTEM). N-Nb2O5 is domain twinned, with the twin plane perpendicular to the c* axis. The domains are rather small and the domain twinning can sometimes be best explained as stacking faults. A second type of coherent twinning at an angle of 90° to the other two domain directions was also found. These domains are linked together by areas containing blocks of different sizes, similar to the disordered block arrangement observed in M-Nb2O5.

The ReO3-type structure NbO2F, niobium dioxyfluoride, has been studied at high pressures using diamond anvil cells and synchrotron X-ray radiation. High-pressure powder diffraction measurements have been performed up to 40.1 GPa. A phase transition from the cubic (Pm3¯m) ambient pressure structure to a rhombohedral (R3¯c) structure at 0.47 GPa has been observed. Rietveld refinements at 1.38, 1.96, 3.20, 6.23, 9.00 and 10.5 GPa showed that the transition involves an a − a − a − tilting of the cation–anion coordination octahedra and a change of the anion–anion arrangement to approach hexagonal close packing. Compression and distortion of the Nb(O/F)6 octahedra is also revealed by the Rietveld refinements. At 17–18 GPa, the diffraction pattern disappears and the structure becomes X-ray amorphous.

High-resolution electron-microscopy (HREM) images from different hydroxyapatite (OHAp) samples showed p3 projection symmetry along [001] instead of the p6 projection symmetry compatible with the space group P6_3/m of OHAp. Image processing was used to establish without ambiguity that threefold symmetry dominates the images along [001]. OHAp crystals decompose in the transmission electron microscope and it is concluded that the threefold symmetry observed corresponds to an early step in the decomposition process, the dehydration of OHAp to oxyapatite (OAp). A structural model for OAp where every second O atom along the 63 axis in OHAp is removed has the maximal space-group symmetry P\bar6. This is compatible with the p3 projection symmetry observed. Atomic shifts in this OAp model compared to the OHAp structure were estimated using the HREM images and geometric optimizations of the atomic structure. No refinements of the atomic coordinates against diffraction data were possible but the simulated HREM images of this crude model fit well with the experimental images.

A non-technical account of the links between two-dimensional (2D) hyperbolic and three-dimensional (3D) euclidean symmetric patterns is presented, with a number of examples from both spaces. A simple working hypothesis is used throughout the survey: simple, highly symmetric patterns traced in hyperbolic space lead to chemically relevant structures in euclidean space. The prime examples in the former space are derived from Felix Klein's engraving of the modular group structure within the hyperbolic plane; these include various tilings, networks and trees. Disc packings are also derived. The euclidean examples are relevant to condensed atomic and molecular materials in solid-state chemistry and soft-matter structural science. They include extended nets of relevance to covalent frameworks, simple (lattice) sphere packings, and interpenetrating extended frameworks (related to novel coordination polymers). Limited discussion of the projection process from 2D hyperbolic to 3D euclidean space via mapping onto triply periodic minimal surfaces is presented.