Search

We report improvement in the precision of certain of the polarized neutron diffraction data for Cs 3 CoCl 5 . The improvement allows us to analyse the data using a chemically based model of the spin-density distribution that is equivalent to a conventional multipole treatment to fourth order on the cobalt, and to second order on the chlorine atoms of the CoCl 2– 4 ion. To test the completeness of the model and to understand the meaning of the parameters in terms of the wavefunction, we have used it to analyse a set of theoretical magnetic structure factors. These are obtained from the wave-function of a Hartree–Fock calculation on the CoCl 2– 4 ion. We obtain an excellent fit to the theoretical ‘data’ and a much improved fit to the experimental data when the new model is used. We confirm the main features of the spin- and charge-density distributions deduced in our previous study, and we are now also able to interpret the experimental parameters in terms of the wavefunction by analogy with the fit to the theoretical data. We find that there is ca . 3 % of the total spin delocalized onto each chlorine atom of the CoCl 2– 4 ion, dominantly via σ - rather than π -bonding. There is a well defined diffuse spin density on the cobalt atom of 4p symmetry, and strong evidence for 3d–4p mixing. The spin density, in this almost cubic ion, has distinct non-cubic symmetry, which may arise from longer-range effects due to the rest of the tetragonal crystal.

Spin-only magnetic form factors and X-ray form factors for Mn, Mn+, Co, Co2+, Ni and Ni2+ are calculated using configuration interaction wavefunctions. The magnetic form factors for Co2+ are used in the analysis of polarized neutron diffraction data for Cs3CoCl5, to give spin populations. The inclusion of correlation functions in the wavefunction leads to changes in Hartree-Fock form factors of transition metal atoms, which maximize at K/4π ~ 0.4 A-1, where they amount to approximately 5% of the form factor at that value of K.

The structure of K3[Cr(CN) 6] at 4.2 K has been determined by a single-crystal neutron diffraction study. The crystal was found to consist of four twins together with stacking disorder [triclinic, P1, a = 8.450 (5), b = 10.576 (8), c = 13.557 (4)/k, (~ = 90.0 (1),/~ = 90.0 (1), y = 90-0 (1)°1. Refinement of atom parameters with four extra population (twin and disorder) parameters gave R(F 2) = 0.053 for 1575 reflections, and x(F 2) = 3.73 at convergence. The Pi structure is related to the 295 K Pcan structure by symmetry-determined translations, but not rotations, of the constituent K ÷ and Cr(CN)~- ions, together with some small additional distortions.

Polarised neutron and X-ray diffraction experiments have been performed on Ni(NH3)4(NO2) 2, to study respectively the spin and charge distributions in the molecule. The results have been compared with an ab-initio M.O. calculation. 27 % of the spin is transferred to the ligand atoms by covalent interactions. A simple ligand field model, with four empirical parameters, agrees qualitatively, but not quantitatively with the combined spin and charge results. The U.H.F. ab-initio calculation results are also in broad agreement with the experiments, but correspond to a good deal less covalency in the bonding. This emphasises the need to consider configuration interaction and/or a long range "crystal field" in describing bonding in crystalline transition metal compounds.

The temperature dependence of the paramagnetic susceptibilities of single crystals of Cs 3 CoCl 5 , Cs 2 CoCl 4 and K 2 Co(CNS) 4 . 4H 2 O have been measured over the range 80 to 300 °K. The splitting of the excited 4 T 2 term by low symmetry crystal fields is of the order of 1000 cm -1 , and the extent of t 2 g electron delocalization in each complex corresponds to a delocalization parameter k of approximately 0.9. The magnitudes of the observed 4 T 2 term splittings are calculated in a semi-quantitative way through a comparison of the details of the molecular geometry with crystal field theory for a distorted tetrahedron. Magnetic exchange effects in the complexes, corresponding to exchange integrals of several degrees, are sufficiently large that the zero field splitting effect is relatively unimportant. The geometrical arrangement of adjacent (CoCl 4 ) 2- tetrahedra taken with the delocalization parameters allows an order-of-magnitude calculation of anisotropic magnetic exchange in crystals of Cs 3 CoCl 5 and Cs 2 CoCl 4 . A new X-ray analysis of the co-ordination geometry of Cs 3 CoCl 5 is also summarized.

Single crystal X-ray analyses of the structure of potassium ferricyanide at ca. 300 and 95° K are reported. Least-squares refinements within the monoclinic space group, P 2 1 / c , have converged with the residuals R 300°K = 0·092 and R 95°K = 0·077. The unit cell dimensions at 300°K are: a = 7·06, b = 10·38, c = 8·40 Å, β = 107·0°, and at 95°K are: a = 7·03, b = 10·31, c = 8·35 Å, β = 107·2°. The coordination octahedra are only slightly distorted, both analyses independently revealing a small tetragonal elongation of the molecule along Fe-CN bonds lying nearly perpendicular to the crystallographic c axis. The principal crystal paramagnetic susceptibilities are characterized by an essentially unique, and lower, moment parallel to the c axis. These two observations are to be reconciled only in relation to the role of second-nearest neighbour effects of either antiferromagnetic or crystal-field character. Exchange is ruled out by earlier e.s.r. work and the present susceptibility dilution experiments. The crystal anisotropy appears to be dominated by the coulombic field set up by adjacent potassium ions in the lattice. While all potassium ions in the structure will play some role in this second-order perturbation, those forming ‘chains’ with the octahedra parallel to the c axis are likely to be the more important. A simple crystal-field approach, using symmetry-adapted, zero-order wavefunctions on the iron atoms, has shown how the correct direction and relative magnitudes of the crystal magnetic anisotropies may be predicted by this model. The splitting of the ground 2 T 2g term of the complexion is estimated as 150 to 300 cm -1 .

[ND 4 ] 2 [Fe(D 2O ) 6 ](SO 4 ) 2 cristallise dans P2 1 /a avec affinement jusqu'a 0,024. Structure isomorphe a celle d'autres sels de Tutton formes a partir d'ammonium

The spin distribution in the Cr(CN)63− ion in Cs2KCr(CN)6 has been determined by polarised neutron diffraction and analysed in terms of atomic orbital populations. It does not agree with simple models of covalence in metal-ligand bonding, strong spin polarisation obviously being present. An unrestricted DV Xα calculation shows features in fair agreement with our results.

Analyse de 207 facteurs de structure magnetique, obtenus par diffraction de neutrons polarises par la phase de groupe d'espace Pccn, pour en deduire les populations de spin des atomes: dans le cation, 15% des spins se trouvant sur les atomes des coordinats (atomes Cl plus particulierement) et, dans l'anion, 17% sur les atomes Cl

The spin-density data previously obtained by a polarised neutron diffraction experiment on K2Na[CrF6] have been reanalysed. A chemically-based model for the CrF63– ion, involving conventional atomic orbitals, has been fitted to the 92 observed magnetic structure factors using a least-squares procedure. The best refinement, with seven variable parameters, converged with the agreement factors R 0.0861, R′ 0.0579, and χ2 1.569. The spin density in chromium 3d orbitals has t2g symmetry [t2g2.66(5), eg–0.06(5)]. There is also a region of spin density centred on the chromium atom which is radially much more diffuse than the theoretical 3d orbitals and contains 0.4(1) spins parallel to the spin density of t2g symmetry closer to the chromium centre. The model indicates a small parallel spin population [0.020(5)] in each 2pπ orbital of fluorine, and an antiparallel spin population [–0.02(1)] in a fluorine 2pσ orbital. The model also indicates further antiparallel spin density along the Cr–F vector, and no significant spin population of the fluorine 2s orbital. The spin-transfer coefficients fs*, fσ*, and fπ from the present modelling [0.001(3), –0.021(12), 0.020(5) respectively] are shown to be in good agreement with the values determined from other resonance and neutron diffraction techniques.

The crystal structure of the polymer potassium bis(carbonato)cuprate(II), K2[Cu(CO3)2], has been investigated in terms of the electronic structure of the copper atom. Single-crystal X-ray diffraction data with (sinθ)/λ⩽ 10.8 nm–1, which were previously used in a 3σ(I) refinement of the crystal structure, were employed. Crystals are orthorhombic, space group Fdd2, Z= 8, a= 1.142 5(3), b= 1.765 8(4), c= 0.615 4(2) nm. All 1 693 independent reflections were used in a least-squares refinement on intensity values. Final agreement indices were R 0.029 and χ 2.027. The copper atom is in an essentially square-planar environment of four oxygen atoms from the bridging carbonato-groups [Cu–O 192.9(1) and 193.3(1) pm]. The final difference-Fourier synthesis showed non-centrosymmetric areas of positive and negative deformation charge density on opposite sides of the CuO4‘plane.’ These regions of charge were included in the scattering model as spherical scatterers, and their populations refined to ±0.31 (4) electrons at fixed axial positions 55 pm from the copper atom. The dipole moment on the copper atom, parallel to the c crystal axis, is of magnitude 0.5(1)× 10–29 C m. The symmetry of these deformation density features does not follow the local (bonded) square-planer symmetry of the copper atom environment. Two Cu ⋯ CO contacts of ca. 280 pm, which occur in only one axial direction, seem to be involved in the polarisation of the valence electrons of the copper atom. The non-centrosymmetry of these deformation density features precludes their modelling in terms of 3d or 4s atomic orbital populations.

High-quality X-ray data sets were collected on the new and simple molybdenum(III) complex Cs3Mo(NCS)6 at 295 and 115 K. The 1 102 unique data at 115 K were analysed using a multipole model of the charge density to give R(I) 0.015, χ2 1.38. The structure consists of [Mo(NCS)6]3– octahedra compressed down a three-fold axis along c. There is disorder over the two equivalent rotations of the octahedron about c. Deformation density maps show strong peaks at the Cs and Mo positions which probably arise from the disorder and/or anharmonic thermal motion. A valence-orbital analysis of the data restricted to (sin θ)/λ < 7 nm–1 because of the disorder/anharmonicity is able to define the covalent charge transfers to and from the Mo atom and within it, the NCS group, and the Cs atoms. The Mo configuration is 4d2.5(3) 5(s/p)2.0(4), with a strong contraction of the 4d radius, to about 2/3 of the free-ion value. The high diffuse population and 4d orbital contraction are believed to reflect high covalence in the bonding to the NCS– ion, which has comparable σ-donation and π-back-bonding components. The electronic structure of the NCS– ion is well defined and shows good agreement with an ab initio calculation. The magnetic susceptibility of the compound shows evidence of substantial magnetic exchange [θ=–20.1(l) K] and a pathway for this is provided by the close intermolecular S ⋯ S contacts.

We present an interpretation of the polarised neutron diffraction data for Cs3CoCl5 in terms of a simple molecular-orbital description of the CoCl42– ion, with direct modelling of the spin-density features. A direct observation of the overlap spin density in each Co–Cl bond has been obtained, and a probable overlap population of –0.13 ± 0.01 spins has been deduced.We have found that the cobalt atom 3d orbital populations are essentially t2 2.71 ± 0.15, e 0.0 ± 0.1 spins, no statistically significant differentiation of the t2 orbital populations on account of departure from cubic symmetry being observed. There is a further significant centrosymmetric spin population (0.50 ± 0.10) on the cobalt atom, but in a diffuse orbital (maximising at ca. 100 pm from the nucleus) whose angular variation cannot be specified. The spin population of each chlorine atom is, as expected, mainly in the pσ orbital (0.062 ± 0.006 spins) with a total pπ contribution less by a factor approaching three. Fairly minor changes in Co2+ 3d and Cl0 3p free-ion orbital wavefunctions are deduced for the atoms in the complex ion.In general, it is found that our data are sufficiently numerous and accurate that we are able to test the validity of simple ligand-field and molecular-orbital models. A ligand-field treatment of CoCl42– ion describes the data well, but not completely, even if we allow for the occurrence of the diffuse cobalt-centred spin distribution. In terms of a simple molecular-orbital model a mixing parameter Aσ 0.15 ± 0.05 is deduced. The model is further qualitatively successful in accounting for the overlap spin population provided the ‘diffuse’ orbital is considered to contribute significantly to the metal–ligand overlap integrals. However, it is not possible to quantify this aspect nor to exclude spin-polarisation effects as a major contribution to the observed apparent overlap population of –0.13 spins. The total cobalt spin population is greater than 3.0 (3.18 ± 0.06) which is evidence for spin-polarisation effects. In terms of usual fractional spin-transfer coefficients, f, we have fσ 6.2, fπ 1.1, and f8 0.3% for spin transferred to the bonded chlorine atoms.

A high-precision X-ray diffraction data set on the compound Cs2KCr(CN)6 has been collected at 120 K, and analysed in terms of the charge-density distribution and chemical bonding features. It has been useful to model the density only at the level of cubic symmetry, viz. a regular octahedral [Cr(CN)6]3– ion, because small effects, apparently arising from anharmonic motion, perturb the data with high values of h, restricting the resolution along the a axis for charge-density purposes. Molecular form-factors have been derived for the CN– ion by fitting the theoretical electron density by a multipole analysis, enabling a more critical comparison between experiment and theory than is usual. In the region of the chromium atom and its six bonded carbon neighbours the density may be described by the superposition of a ‘prepared’t2g3eg0 Cr3+ ion and the unco-ordinated cyanide ions, with some σ donation and π back donation resulting in t2g2.3(4)eg0.6(5). On the nitrogen atoms of the cyanide ions significant charge [0.3(1) e] appears to have been polarised from the π orbitals into the σ‘lone pair,’ probably by the ‘intramolecular’ influences of the cations of the crystal. These cations are themselves perturbed away from simple spherical 1+ charged species by the contacts with the nitrogen atoms.

The crystal structure of the title compound (CloH9Nz) [Fe(CloH8N2)3] (C104), has been determined at c. 295 K from diffractometer data and refined by least squares to R 0.056 (7018 reflections with I z 3a(I)). Crystals are triclinic, Pi, a 16.607(3), b 13.496(3), c 10.254(2) A, a 101.13(2), B92.08(2), y 10277(1)", Z 2. The iron atom is surrounded by three bidentate ligands, the six Fe-N distances being equal ((Fe-N) 1 .963 A). The monoprotonated bipyridylium cation adopts the unusual cis configuration, with the proton clearly located on one of the nitrogen atoms. The e.s.r. g-tensor is approximately axial (g, 1 .474, g, 2.69& but its axes do not coincide with obvious geometric features of the cation.

The crystal and molecular structure of the β-polymorphic form of phthalocyaninatocobaIt(II), [Co(pc)], has been determined by single-crystal neutron-diffraction methods at 4.3 K. Crystals are monoclinic, space group P21/c, with a= 14.495(5), b= 4.742(4), c= 19.107(5)A, β= 120.76(2)°, and Z= 2. Automatic diffractometry has provided Bragg intensities for 1 261 independent reflections. Data are of limited resolution along the b axis direction (hkl with k⩽ 2). The structure has been refined by full-matrix least-squares methods to R 0.031. Bond lengths at 4.3 K show systematic deviations from those determined at 295 K by X-ray diffraction data of moderate resolution. This is explained primarily in terms of errors in the X-ray determined atomic positions due to the effects of asphericity of the valence electrons, with thermal motion at 295 K having only a small effect. Within the central cavity of the macrocyclic phthalocyaninate ligand, the cobalt atom co-ordinates to a precisely planar array of nitrogen atoms, with Co–N bond lengths 1.908(2) and 1.915(2)A at 4.3 K which differ significantly from the values obtained at 295 K by X-ray diffraction. The neutron-scattering length of cobalt has been found by least-squares refinement to be 0.252(7)× 10–l2 cm.

A charge density study of cobalt(II) phthalocyanine based upon 12,163 unique reflections within a (sin /theta/)//lambda/ range 0-10.8 nm/sup /minus/1/, collected by X-ray diffraction at 115 K, was performed. The data was modeled by a conventional multipole analysis as well as by a valence orbital population analysis. When the latter approach was used and data was employed to a (sin /theta/)//lambda/ limit of 8.2 nm/sup /minus/1/, a value of 0.031 was obtained for R/sub w/(I) and 0.015 for R(F). The deformation density maps show bond-centered peaks and lone-pair peaks of ca. 500 e nm/sup /minus/3/ within the phthalocyanine ligand molecule, such as are typical of most good light-atom charge density studies. The rearrangement of the electron density around the cobalt atom is substantial, both in angular function and in radial extent, and suggests a depletion and contraction of the 3d orbitals, combined with a population of more diffuse cobalt-based functions. The valence orbital population analysis quantifies the suggestion, giving a total 3d population of 4.9 (1)e with a radius reduced 35% from the free ion value and a diffuse population (modeled as 4p) of 2.6 (1)e. Those populations are strongly anisotropic: 3d/sub x/sup 2/-y/sup 2///sup 0.0/3d/sub xz,yz//sup 2.4/3d/sub z/sup 2///supmore » 0.8/3d/sub xy//sup 1.8/4p/sub z//sup 1.1/4p/sub xy//sup 1.4/. There seems to be little charge flow from the ligand to the metal atom (0.5 (1)e), but the bonding is obviously quite covalent in character. This is manifest in the distribution of charge between the 3d /pi/ and /sigma/ systems and the diffuse outer regions of the cobalt atom.« less

The crystal structure of the title compound, [Co(C9H13N5)2] I2,H2O, has been determined by single-crystal X-ray diffraction at 295 K and refined by least squares to a residual of 0.058 for 3255 'observed' reflections. Crystals are triclinic, Pī, a 14.934(5), b 11.315(4), c 8.090(3)Ǻ, α 85.19(3), β 87.76(3), γ 74.10(3)°, Z 2. The distances from the cobalt to the central trimethine nitrogen atoms are 1.893(8) and 1.903(8)Ǻ consistent with the almost low-spin behaviour of the complex; the distal Co-N distances range between 2.045(8) and 2.140(8)Ǻ. One of the iodide anions appears to be disordered over two sites populated in the ratio 0.86:0.14; occupancy of the second site together with the magnetic moment of 2.1 BM may indicate a small proportion of the high-spin isomer.

A new analysis of the 116 K single-crystal X-ray diffraction data set of good accuracy on β-phthalocyaninatomanganese(II) has yielded individual 3d and 4s orbital populations on the manganese atom of significant accuracy and qualitative estimates of bonding electron density in the macrocycle ring. The manganese orbital populations (x and y along Mn–N vactors, z perpendicular to the square plane of co-ordinated nitrogen atoms) have been determined by a least-squares refinement procedure as dxy1.5(2), dxz0.6(2), dyz0.7(2), dz20.9(2), dx2–y20.1(2), and 4s2.0(3). Difference-Fourier syntheses indicate the aspherical nature of the valence charge density on the manganese atom, and the presence of bonding charge density in the ligand. In the later stages of the structural and orbital population analyses, allowance was made in the scattering model for overlap charge density by placing small spherical charges of variable population and radial extent between C–C and C–N atom pairs and in the Mn–N vectors. The present treatment of the data differs from the previous analysis in that refinement of atomic positional and thermal parameters has been based upon intensities (I), and all 2 265 observations with (sinθ)/λ⩽ 0.662 A–1, rather than on structure factor amplitudes (|F|) and only the 1 608 reflections with I > 3σ(I). Full-matrix least-squares refinement on the basis of a spherical atom model yielded R 0.045, and bond lengths and angles of improved accuracy are reported. The atomic positional and thermal parameters thus obtained have been used to generate a set of ‘difference structure factors,’Fd, containing, principally, information about the valence charge density on the manganese atom. These Fd values have been analysed by least-squares refinement in terms of aspherical density arising from 3d and 4s orbital populations on the manganese atom. The present analysis has established that worthwhile information on chemical bonding in a molecule as large as a transition-metal macro-cycle can be obtained from a low-temperature X-ray data set of good quality, and acts a guide for a program to involve the collection of low-temperature data sets of excellent quality for this purpose.

The crystal structure of the title compound has been determined by the heavy-atom method from diffractometer data and refined by least-squares to R 0·082 for 3449 reflections. Crystals are monoclinic, space group A2/a, a= 23·252(6), b= 18·342(4), c= 17·824(5)A, β= 92·45(2)°, Z= 8. The iron atom is surrounded by three bidentate ligands, the six Fe–N distances being equal (mean 1·973 A). The geometries of two of the perchlorate ions are as expected but the third ion is badly disordered, and there may be additional partial occupancy of the lattice by further water molecules. The e.s.r. spectrum has been determined at ca. 85 K and the principal g values are found to be along the molecular pseudo-trigonal axis (g1 1·459(5)) and in the plane normal to it [g2, g3 2·615(10), 2·727(10)]. The splitting of the 2T2g ground term of the Fe3+ ion (ΔAggca. 800 cm–1) is deduced to be opposite in sign to that predicted by crystal-field theory.

The preparation of the title compound [Co(NH3)6] (CrO4)Cl,3H2O is reported together with the determination of its crystal structure at 295(1) K. Crystals are orthorhombic, Pcmn, a 8.594(5),b 8.598(5), c 18.126(9) A, Z 4.

The charge density observed by X-ray diffraction at 140 K in [Ni(tu)4Cl2](tu = thiourea) is reported. The values for the chloride, thiourea, and NiII fragments closely resemble those of the free ion or molecule, except (a) each chloride σ-donates 0.16(14) e and π-donates 0.35(15) e; (b) each thiourea σ-donates 0.17(20) e to the nickel; (c) NiII, in its 3B1 ground state, has a 3d configuration dxy1.91(14), dxz,yz3.08(18), dx2–y21.12(14), dz21.39(14) with a substantial [1.6(7) e] 4p population. These figures are consistent with a conventional bonding model, in which 3dσ orbitals on NiIIσ-bond almost equally to chlorine and thiourea, and in which 3dπ and 4pπ orbitals π-bond only to chlorine. The residual density maps suggest a differential nephelauxetic effect is occuring, with metal-centred π orbitals larger than σ. A polarisation of the entire NiCl2fragment is also observed, perhaps caused by intermolecular effects. Previous results show that thiourea has a much larger effect on the spectra and magnetism than the chloride. This contrast with the charge-density results is explained by the greater dominance of overlap in spectroscopy, and of ligand energy levels (or crudely, ionisation potentials) in charge densities. This illustrates experimentally the theoretical observation that covalence (i.e. molecular orbital) parameters are best ‘directly’ observed by diffraction, since spectroscopy is often dominated in more ionic complexes by the less interesting terms due to orthogonalisation of metal and ligand orbitals.

The crystal structure of deuterated trans-tetraamminedinitronickel(II), Ni(ND3)4(NO2)2, has been determined by single-crystal neutron diffraction methods at 4·2 K. Crystals are monoclinic, C2/m, a 1058(l), b 672(1), c 586.3(3) pm, β 114.82(5)o, Z = 2. Diffractometry has provided Bragg intensities for 219 independent reflections; and the structure has been refined by full-matrix least-squares methods to R(F2) 0·070 and χ2 3·8. There are slight differences in the molecular geometries determined by neutron diffraction and earlier X-ray determinations of Ni(NH3)4(NO2)2 at 295 and 130 K. Small, but significant, decreases are evident in all non-hydrogen bond lengths on decrease in temperature from 295 to 4·2 K, up to a maximum of 2·0(4) pm for the Ni-NH3/ND3 bond. The magnitudes of these decreases are correlated with the force constants of the bonds. The intermolecular geometry and thermal parameters show that in the ab plane there is a network of relatively strong, linear N-D···O hydrogen bonds. In the c* direction there is a slightly bent, longer, N-D(1)···O bond which is weaker. This causes a large amplitude of rigid-body translational motion in the c* direction, together with high thermal motion of D(1) in the b axial direction.

Cristallisation dans P2 1 /n avec a=11,302, b=8,303 et c=7,761 A, β=90,30°, affinement jusqu'a R=0,042. Le compose est isomorphe avec d'autres elpasolites Cs 2 M I M III (CN) 6 dans leurs phases monocliniques. Les cartes de difference residuelle presentent des pics coherents avec l'anharmonicite attendue a partir d'une transition vers les phases de symetrie plus elevee caracteristique des alpasolites

The crystal structure of deuterated trans-tetraamminedinitronickel(II), Ni(ND3)4(NO2)2, has been determined by single-crystal neutron diffraction methods at 4·2 K. Crystals are monoclinic, C2/m, a 1058(l), b 672(1), c 586.3(3) pm, β 114.82(5)�, Z = 2. Diffractometry has provided Bragg intensities for 219 independent reflections; and the structure has been refined by full-matrix least-squares methods to R(F2) 0·070 and χ2 3·8. There are slight differences in the molecular geometries determined by neutron diffraction and earlier X-ray determinations of Ni(NH3)4(NO2)2 at 295 and 130 K. Small, but significant, decreases are evident in all non-hydrogen bond lengths on decrease in temperature from 295 to 4·2 K, up to a maximum of 2·0(4) pm for the Ni-NH3/ND3 bond. The magnitudes of these decreases are correlated with the force constants of the bonds. The intermolecular geometry and thermal parameters show that in the ab plane there is a network of relatively strong, linear N-D···O hydrogen bonds. In the c* direction there is a slightly bent, longer, N-D(1)···O bond which is weaker. This causes a large amplitude of rigid-body translational motion in the c* direction, together with high thermal motion of D(1) in the b axial direction.

The crystal structure of the title compound (C10H9N2) [Fe(C10H8N2)3] (ClO4)4 has been determined at c. 295 K from diffractometer data and refined by least squares to R 0.056 (7018 reflections with I > 3σ(I)). Crystals are triclinic, Pī, a 16.607(3), b 13.496(3), c 10.254(2) Ǻ, α 101.13(2), β 92.08(2), γ 102.77(1)°, Z 2. The iron atom is surrounded by three bidentate ligands, the six Fe-N distances being equal ( 1.963 Ǻ). The monoprotonated bipyridylium cation adopts the unusual cis configuration, with the proton clearly located on one of the nitrogen atoms. The e.s.r. g-tensor is approximately axial (g|| 1.474, g, 2.690) but its axes do not coincide with obvious geometric features of the cation.

The magnetic susceptibility of K3[Fe(CN)6], [Fe(phen)3][ClO4]3,3H2O, [Fe(phen)3][PF6]3,[Fe(bipy)3][ClO4]3, 3H2O, and [Fe(bipy)3][PF6]3 have been measured between 4 and 300 K. The e.s.r. g-values of the compounds, other than the first, were measured at ca. 85 K. The results are interpreted in terms of a three-parameter model for the splitting of the 2T2g, ground term of the t2g5 configuration. The spin–orbit coupling constant is ca. 270 cm–1 in each compound. The t2g, orbital angular-momentum reduction factor is 0.75 in K3[Fe(CN)6] and ca. 0.89 in the other compounds. The splitting of the t2g orbital set by an axial ligand-field component is small for K3[Fe(CN)6](130cm–1) and may not reflect details of the ligand positions. The splitting is larger in the other compounds (ca. 850 cm–1) and appears to be an intrinsic property of the ion, not much affected by the nature of the anion or by the presence of solvent molecules.

The crystal structure of the polymer potassium bis(carbonato)cuprate(II), K2[Cu(CO3)2], has been investigated in terms of the electronic structure of the copper atom. Single-crystal X-ray diffraction data with (sinθ)/λ⩽ 10.8 nm–1, which were previously used in a 3σ(I) refinement of the crystal structure, were employed. Crystals are orthorhombic, space group Fdd2, Z= 8, a= 1.142 5(3), b= 1.765 8(4), c= 0.615 4(2) nm. All 1 693 independent reflections were used in a least-squares refinement on intensity values. Final agreement indices were R 0.029 and χ 2.027. The copper atom is in an essentially square-planar environment of four oxygen atoms from the bridging carbonato-groups [Cu–O 192.9(1) and 193.3(1) pm]. The final difference-Fourier synthesis showed non-centrosymmetric areas of positive and negative deformation charge density on opposite sides of the CuO4‘plane.’ These regions of charge were included in the scattering model as spherical scatterers, and their populations refined to ±0.31 (4) electrons at fixed axial positions 55 pm from the copper atom. The dipole moment on the copper atom, parallel to the c crystal axis, is of magnitude 0.5(1)× 10–29 C m. The symmetry of these deformation density features does not follow the local (bonded) square-planer symmetry of the copper atom environment. Two Cu ⋯ CO contacts of ca. 280 pm, which occur in only one axial direction, seem to be involved in the polarisation of the valence electrons of the copper atom. The non-centrosymmetry of these deformation density features precludes their modelling in terms of 3d or 4s atomic orbital populations.

The crystal structure of the title compound, [CO(C15H11N3)2](ClO4)2.c. 1.3H2O has been determined by single crystal X-ray diffraction methods at 295 K. The structure is described in terms of atetragonal P42/n sub-lattice structure similar to that of [Co(terpy)2]I2.2H2O, with a 8.86(1), c 20.19(2) �. Perchlorate orientation and disorder, however, demand consideration of superlattice structures in I 41/a//F41/d with 4 and 8 times the P41/n cell volume. The relationship between the general structural features of the [Co(terpy)2] X2 salts is described and discussed. The cobalt-nitrogen(central, distal) distances (2.03, 2.14 �) are the longest observed in the [Co(terpy)2]2+ system, and are consistent with the unusually high magnetic moment of the salt (4.2 BM).

The preparation of the title compound [Co(NH3)6] (CrO4)Cl,3H2O is reported together with the determination of its crystal structure at 295(1) K. Crystals are orthorhombic, Pcmn, a 8.594(5),b 8.598(5), c 18.126(9) �, Z 4.

The magnetic susceptibility and anisotropy of cis-bis(2,2'-bipyridyl)dichloroiron(III) tetrachloroferrate(III), cis-[Fe(bpy)2Cl2]+ [FeCl4]-, have been measured between 1.8 and 300 K at magnetic fields up to 1 T. The magnetization on the cation and anion sites has been estimated by means of polarized neutron diffraction measurements at 1.5 and 4.6 T, and at temperatures from 2.1 to 41 K. These magnetic results may be fitted to a 'mean-field' model of magnetic exchange in which the total exchange integral parameter, ΣijJij, is c. -1.1 K. The magnetic results and structural details for the compound suggest that the strongest exchange pathway connects chains of alternating cations and anions. The inter-chain exchange is small. However, the substantial magnetic anisotropy and the much stronger decrease with temperature of the FeCl4- ion site moment, relative to that of the cation, both indicate a complex magnetic exchange network and consequent magnetic behaviour. No sign of magnetic order is observed to at least 2.9 K.

The crystal structure of the title compound, [Cr(OC(NH2)2)6] [Cr(CN)6].2(CH3)2SO.2CH3CH2OH, has been determined from a single-crystal X-ray diffraction study at 295(1) K. It was refined by least-squares methods to a residual of 0.039 for 4203 independent 'observed' reflections. Crystals are monoclinic, P21/n, a 17.449(6), b 10.315(4), c 11.726(3) �, β 102.51(2)�, Z 2. Both chromium atoms lie on crystallographic centres of symmetry. Cr-O(urea) are 1.960(2),1.962(2), 1.982(1) � and Cr-C(cyanide) 2.074(2), 2.080(2), 2.081(2) �; the geometries of the ligands and solvent molecules are correspondingly precise.

The magnetic structure factors of 101 unique hkl reflections of Cs3CoCl5 have been measured at 4.2 K with a magnetic field of 4.6 T along the crystal c axis. A further 110 unique reflections with a magnetic field of 1.5 T along the a axis have also been obtained. A statistical analysis of the errors in the data sets has been made so that the reliability of their interpretation in models of chemical interest may be assessed. Ten of the c-data reflections, all based on low nuclear structure factors, were rejected on statistical grounds. The major sources of error in the data were found to be extinction and multiple scattering, with counting statistics and machine instability playing a lesser role. The two data sets agree well, no reflection exceeding 3σ from its scaled value. Using a free-ion Co2+ magnetic form factor, the data are reproduced to within 5% by a single spherical model of spin occupation 3d2.4x0.6, where x represents orbitals much more diffuse than Co2+ 3d functions. That is, the data indicate that ca. 80% of the spin is located in Co2+ 3d-like orbitals and 20% is in more diffuse orbitals, whether cobalt-centred or delocalised onto the chlorine ligand atoms. Covalence is indicated by the non-zero intensities of the l-odd data. Small, but statistically significant deviations from the above model are expected to be correlated with, for example, the concentration of the Co2+ 3d spin into the t2 orbitals and the details of the distribution of the diffuse ‘x’ population.

The crystal structures of (ND4)2V(SO4)2.6D2O and (ND4)2V0.45Zn0.55(SO4)2.6D2O have been determined at 5.8 K by neutron diffraction. The 663 and 857 unique reflections measured respectively refined to give R(F2) 0.035 and 0.045. The crystals are monoclinic, P21/a, isostructural with the other members of the series of Tutton salts. The V-O bond length of 211.8 pm [210.4 pm for mean (V/Zn)-O in the solid solution with Zn] is intermediate between these for the Mn (214.9 pm) and the Ni(205.5 pm) salts. It is shorter than the value determined by X-ray diffraction at 295 K. The major departure from a regular VO6 octahedron is a shortened V-O(9) bond, such as is observed in the other Tutton salts, to give a tetragonal distortion.