Search

Detailed insights into the structural features of γ-zirconium phosphate pillared with butyl chains were obtained by molecular dynamics simulations (NPT and NVT type) of models. The available X-ray powder diffraction structure provided reasonable initial conditions (interlayer distances and unit cell parameters) to generate the models submitted to the simulations (timing 1 ns). Using the results of the molecular dynamics trajectories, the X-ray structure of the compound was adequately reproduced by an average of the structural models corresponding to the four accessible orientations of the butyl chains toward the a and b cell axes. This result showed that, in the crystallite, the chains were arranged along the four possible orientations. The predominant structural motif of the molecular assembly in the bulk of pillared γ-zirconium phosphate butyldiphosphonate was found to be composed of layers laid one upon the other with the pillars of each layer pointing in the same direction, with rare interruptions of this sequence by opposite positioning of the butyl chains of two interfaced layers. The molecular dynamics technique was thus found to be a valid approach to solving otherwise intractable issues concerning structural features of γ-zirconium phosphate derivatives pillared with aliphatic chains, and it is likely that this promising technique can be extended to a variety of similar compounds.

The structure and conformation of semicrystalline poly[bis(phenoxy)phosphazene] (PBPP) was studied using molecular mechanics including ad hoc quantum mechanically derived force field (FF) parameters, in combination with the energy-dispersive X-ray diffraction (EDXD) technique. The atom−atom radial distribution function (RDF) curves were calculated for the various models of backbone conformations and for the structure in the crystalline α-form of PBPP. On the basis of comparison between theoretically calculated RDFs with the RDF obtained from the EDXD experiment, a model was proposed. The structural features of this model are (i) the polymer backbone adopts a low-energy, planar, trans−cis [TC]n conformation, (ii) each chain is, on average, comprised of 16 monomeric units, and (iii) the unit cell may be restricted to contain two such chains running antiparallel to each other and with their backbones aligned parallel to the (1, 1, 0) Miller plane. The results demonstrate the capability of the combined use of computational chemistry (molecular modeling) and X-ray diffraction techniques such as EDXD to provide insights, otherwise experimentally inaccessible, into the conformational and structural features of semicrystalline polyphosphazenic materials.

When soybean [Glycine max(L.) Merr.) is grown in soil containing few Bradyrhizobium japonicum(Kirch.) Jordon, the limited mobility of the symbiont results in the formation of a small number of nodules distributed mainly on the primary root of the plant near the inoculation point. The authors conducted a 3‐yr study aimed at increasing nodulation and N2fixation in ‘Hodgson’ soybean as a second crop, by cover inoculating B. japonicumon the soil with irrigation water at the time of sowing or at the three‐node (V3) stage. Inoculation treatments were: (i) seed inoculation; (ii) cover inoculation at sowing; (iii) seed inoculation plus cover inoculation at sowing; (iv) cover inoculation at the V3stage; (v) seed inoculation plus cover inoculation at the V3stage; (vi) uninoculated control. The seed plus cover inoculation at sowing or the V3stage produced, on the whole, the greatest nodulation, 1.4 to 2.4 times that for seed inoculation alone. The addition of cover inoculation at sowing increased seed yield by 50 g m−2in 1985 and 90 g m−2in 1986, regardless of whether the cover inoculation was performed at sowing or the V3stage. In 1988, a year in which the soybean was sown about 1 mo later than in the previous years and subjected to higher temperatures, seed and cover inoculation increased N2fixation and plant‐N concentration, but not yield.

In the framework of the molecular dynamics modified-neglect-of-diatomic-overlap approximation we have studied the cis-trans isomerization process of polyacetylene (PA) induced by doping. We have considered the following four different models: 1H, the addition of a single atom H to a positive ionized PA chain; 2H, the addition of two H atoms to the neutral PA chain; 1I, the addition of an I atom to a positive ionized PA chain; and 2I, the addition of two I atoms to the neutral PA chain. In the cases 1H and 1I we have found only partial isomerization of the chain. In the cases 2H and 2I we have found complete isomerization. These results are in agreement with the experimental data which show that complete isomerization can be achieved only for doping levels greater than 7%. The isomerization mechanism involves first a transformation from the cis-transoid to the trans-cisoid (TC) form, followed by the transition from TC to the trans-transoid form. The subtraction of some charge from the double bonds, due to the dopant presence, is the essential process to induce the isomerization of the cis-rich PA film.

A relationship is explored between correlation energies of light diatomic molecules, both hetero and homo, correlation energies of the constituent atoms and the overlap population S. While S is quantitatively sensitive to the choice of basis set, it is demonstrated that the general trends revealed are largely independent of such a choice. A weak residual dependence on molecular chemical potential is also indicated by the present study.

It has been known since the work of March and White that the simplest nonrelativistic density functional theory, namely, the statistical method of Thomas, Fermi, and Dirac, sums subseries of the so-called 1/Z expansion to yield, for heavy neutral atoms, the ground-state energy E= − a0Z^7/3+ a1Z²−a2Z^5/3+···. The term of O(Z^5/3) is the Dirac–Slater exchange energy Eexc, and it is of considerable interest to know at what order the correlation energy Ecorr enters this expansion. Dimensional scaling considerations by Kais et al. suggested Ecorr∝Z^4/3 in the limit of large Z. Here, attention is focused on whether this can be distinguished empirically from a term of the form (aZ ln Z+bZ) for neutral atoms. If the latter term is correct, then a relationship between Ecorr and the Shannon information entropy can be forged analytically for large atomic number Z in nonrelativistic theory. © 1998 John Wiley & Sons, Inc. Int J Quant Chem 69: 721–726 (1998)