Read an Excerpt
Inorganic Chemistry of the Main-Group Elements Volume 5
A Review of the Literature Published Between October 1975 and September 1976
By C. C. Addison The Royal Society of Chemistry
Copyright © 1978 The Chemistry Society
All rights reserved.
ISBN: 978-0-85186-792-2
CHAPTER 1
Elements of Group I
BY P. HUBBERSTEY
1 Introduction
The definition of the limits of the literature search pertinent to the present Report is complicated by the extensive role of the alkali metals as simple counter-cations. In general, papers have been abstracted which are relevant to a number of broad subject groups in which the role of the alkali metals is unique. Consequently, the format of this Chapter is such that the inorganic chemistry of the alkali metals is considered collectively in sections which reflect topics presently of interest and importance.
For certain topics (e.g. cation solvation, molten salts, crown and cryptate complexes), the chemistry of the Group I and II metals is closely interwoven; in these cases, the data abstracted are considered once only in the relevant section in this Chapter.
The extraction of alkali-metal cations from salt solutions into organic solvents has been the subject of four papers. The ion [π-3-1,2-B9C2H11]Co- has been proposed as a nearly ideal hydrophobic anion for extraction of M+ ions into C6H5NO2via formation of ion pairs. Li+ has been selectively extracted from nearly neutral aqueous solutions of alkali-metal salts via the formation of the trioctylphosphine adduct of a lithium chelate of fluorinated β-diketones; although high separation factors were obtained from Na+, K+, Rb+, and Cs+, selectivity from the alkaline-earth-metal cations was found to be poor. The extraction of M+ into PhNO2 and MeNO2 using hexafluoroacetylacetonate has also been investigated. Dissociation constants of the alkali-metal enolates were determined, the extent of association of enolate ion with enol to give a dimeric ion was deduced, and the latter's formation constant calculated.
2 The Alkali Metals as Solvent Media
The role of liquid sodium as a heat-exchange medium in the fast breeder reactor, and that of liquid lithium as a prime candidate far use as the blanket medium in a deuterium–tritium-fuelled thermonuclear reactor, has maintained interest in the solution chemistry of these liquid metals.
Phase equilibria for Li–Li3N dilute solutions have been investigated by two independent groups of authors. Pulham et al. have determined the hypoeutectic and hypereutectic liquidi by thermal and by electrical resistance methods, respectively. The freezing point of Li (453.64 K) is depressed by 0.25 K to 453.39 K at the eutectic composition 0.068 mol % N. The depression was used to calculate the solid solubility of Li3N in Li (0.O24 mol%N) at the eutectic temperature. The solubility of Li3N in liquid Li increases smoothly from the eutectic to 2.77 mol % N at 723 K. Over a wide temperature range, the data can be represented by equation (1). These latter data are corroborated by those of Veleckis et al. [equation (2)], who used a direct sampling technique. This agreement resolves the problem of the earlier inconsistent data referred to in the previous Report. Veleckis et al. also measured the equilibrium nitrogen pressure over solid LiN at temperatures between 933 and 1051 K. From a thermodynamic analysis of the solubility and decomposition data, the standard free energy of formation of solid Li3N (ΔG[??]f/kJ mol-1) was estimated to be 138.9 × 10-3T/K – 163.6. For dilute solutions of Li3N in Li, the Sieverts law constant ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is given by -13.80+14590 (T/K)-1. The melting point of Li3N was found to be 1086 K, in good agreement with the previously reported value of 1088 K.
Phase equilibria of Li–LiH and Li–LiD dilute solutions have also been studied by Pulham et al. The maximum depression of the freezing point of Li by LiH (LiD) is 0.08 K (0.075 K), corresponding to a eutectic composition of 0.016 mol%H (0.013 mol%D). These data, which indicate negligible solid solubility of the salts in Li, have been used to show that both hydrogen and deuterium dissolve in liquid Li as monatomic solute species. Typically, the depression caused by small LiH concentrations (Figure 1) follows quite closely the line derived theoretically for monatomic solutes. The theoretical line for a diatomic species is included in the Figure for comparison. The solubilities of LiH and of LiD in liquid Li have been determined by electrical resistance methods at temperatures up to 824 K (5.68mol%H), and 729 K (2.63 mol%D), and can be represented over a considerable part of the temperature range by equations (3) and (4), respectively. The hydrogen-deuterium isotope effect has been discussed and the experimental data have been extrapolated to predict the behaviour of tritium in liquid Li.
New solubility data for NaH in liquid Na have been determined by Whittingham in a detailed study (610 — 677 K) of the thermodynamic and kinetic properties of the liquid Na–H2 system. Comparison with some previous data has been effected and a composite solubility equation (5) formulated.
These new solubility data for hydrogen isotopes have been collated and compared to the corresponding solubilities in NaK and K; surprisingly, hydrogen is least soluble in sodium.
Solubility data have been used to determine solvation enthalpies, Ux, defined as in equation (6)] for N3-, O2-, H-, and D- in Li and for H- in Na and K. The values of Ux are collected in Table 1. Those for H- and D- in Li are lower than those for O2- and N3- by factors of ca. 22 and 32, respectively, corresponding to increasing Ux with increasing charge of solute. Those for H- in Li, Na, and K are very similar, that in Li being the greatest. Solvation enthalpies have been derived in ab initio M.O. calculations of solvation clusters in Li and Na. By comparison with experimental data, the best model was deduced to be that of a tetrahedral solvation sphere of cations supplemented by a further metal tetrahedron positioned on the three-fold axes of the first solvation sphere. Other incidental results to emerge from the calculations are the effective radii for Li (0.1675 nm), Na (0.1715 nm), and H (0.0525 nm in Li and 0.0535 nm in Na) and the effective charges on the H (–0.45 in Li and –0.25 in Na).
The chemistry of liquid alkali metal–hydrogen solutions has been surveyed. Whereas hydrogen and nitrogen act independently in Li at 693 K, hydrogen and oxygen interact in Na at 673 K, according to equilibrium (7). Hydrogen-oxygen interactions in the other alkali metals are also considered and rationalized in terms of the enthalpy changes of the corresponding solid-state reaction. Furthermore, Y has been shown to react with hydrogen in Li at 673 K to form a mixture of Y(H) solid solution and YH2 according to reaction (8).
Enrichment of deuterium in the gaseous phase above dilute Li–LiD solutions (xD = 10-5) has been observed by Ihle and Wu at temperatures above 1240 K. This supports the contention that deuterium can be removed from highly dilute solutions in Li by distillation. The results are of importance in the context of the technology of thermonuclear reactors and have been extrapolated to Li–LiT solutions.
Several papers pertinent to the elucidation of the corrosive properties of very dilute solutions of non-metals in liquid alkali metals have been published. The corrosion of V, Nb, Ta, Mo, and W plates suspended in dynamic liquid sodium, containing more than 5 p.p.m. oxygen, has been examined at 873 K; the products were analysed through a matrix of Na by X-ray diffraction techniques. The ternary oxides Na4VO4 and NaVO2 were formed on V, together with a V(O) solid solution. For Nb and Ta, only a single ternary oxide Na3MO4 (M = Nb or Ta) was observed, together with a M(O) solid solution. Although corrosion of Mo was found to be independent of oxygen concentration, no ternary oxide products being observed, that of W was found to be strongly influenced by initial oxygen concentration in the Na. At low oxygen levels, the cubic phase Na3WO4 was identified; at very high oxygen levels in static Na, however, the orthorhombic phase Na6WO6 was observed. Inclusion of labile carbon in the system containing Mo caused the formation of Mo2C. The closely related solid-state reaction of Na2O with Mo and W under vacuum gave the ternary phases Na4MoO5 and Na4WO6, respectively, together with unreacted refractory metal and Na vapour.
Barker and Hooper have reinvestigated the products of the reaction of liquid Na with CrO3 at temperatures up to 873 K; CrO2, Cr2O3, and Na2CrO4 were also studied as substrates. The ternary oxide NaCrO2 is found in each case in which reaction took place. The previously accepted reaction product, Na2CrO3, was not formed; the error has been rationalized in terms of the experimental procedure, and improved techniques have been developed. Gellings et al. have also studied the reaction of CrO3 with liquid Na, their results corroborating the identification of NaCrO2 as product. The product of these reactions, NaCrO2, together with the other ternary oxides Na3CrO4 and NaCrO4, has been prepared by Barker et al. by the reaction of Na2O and Cr2O3 or Cr, and it has been characterized by X-ray powder diffraction techniques. NaCrO2 decomposes reversibly to the simple oxides at ca. 1068 K.
The reaction of pure liquid Li with MO2 (M = Ti, Zr, Hf, or Th) has been shown to follow thermodynamic predictions. TiO2 and ZrO2 give rise to Li4O and the appropriate transition metal; HfO2 yields Hf metal, Li2O, and a tetragonal phase, which may be the ternary oxide LiHfO2; ThO2 does not react. Reaction with liquid Li doped with dissolved nitrogen, however, converts all four oxides, in differing degrees, into either the mononitride or a ternary nitride Li2MN2 (M = Zr, Hf, or Th).
Liquid K reduces NiO to Ni metal at 458 K with the concomitant formation of the ternary oxides K2NiO2 and K2NiO3; thermomagnetic analysis indicates that the reaction occurs in a single step. K2NiO2 was also prepared by the reaction of equimolar quantities of K2O and NiO; K2NiO3 was produced by the reaction of K2O and NiO in O2 or by heating K2NiO2 in a stream of O2.
The reaction between Ba and N in liquid Na has been investigated at 573 K. Solubility studies showed that the reaction of a 4.40 mol % Ba solution occurs in two stages; (i) dissolution of N (N is insoluble in pure liquid Na), and (ii) precipitation of Ba2N, the initial product of the reaction. The occurrence of these two processes is reflected in the resistivity studies effected on a number of Na–Ba solutions (between 0.34 and 6.89 mol % Ba). The extent of the solution process was found to be a linear function [equation (9)] of the initial Ba concentration, the solubility limit corresponding to an overall reaction composition approximating to Ba4N. This ratio, and the decrease in resistivity which invariably occurred during the solution process, leads to the concept of strong preferential solvation of the nitride ion by Ba cations, perhaps in the form of a 'Ba4N' solvated unit.
The reaction of C2H4 with liquid K has been studied in the range 503 — 671 K. At low temperatures, self-hydrogenation occurs precisely according to equation (10). The surface reaction is explained by dissociative adsorption of C2H4 into H adatoms, which are subsequently employed in hydrogenation. With increasing temperature, progressively less C2H6 is produced, which is attributed to the loss of H from the surface by solution in the metal.
3 Metallic Solutions and Intennetallic Compounds
The nature of the bonding in intermetallic phases has been discussed, and an attempt has been made to demonstrate qualitatively the dependence of both the number of phases in a binary system and their relative thermal stabilities on the electronic configurations of the component atoms. Particular attention has been devoted to compounds of the alkali metals with Hg, Sn, Pb, Sb, and Bi.
The preparation of the novel compounds K2Cs and K7Cs6 by precipitation from solid K–Cs solutions at temperatures below 183 K has been reported. Structural analysis has shown that K2Cs (a = 0.9065, c = 1.4755 nm at 178 K) is isotypic with the hexagonal Laves phase Na2K, whereas K7Cs6 (a = 0.9078, c = 3.2950 nm at 178 K) forms hexagonal crystals with a novel kind of Frank-Kasper structure. Although the K atoms in K7Cs6 are sited in two 12-co-ordination polyhedra, the Cs atoms occupy one of four sites with 14-fold, 15-fold (×2), and 16-fold co-ordination. The K ··· K, Cs ··· Cs, and K ··· Cs distances vary from 0.454 to 0.461, from 0.501 to 0.546, and from 0.466 to 0.574 nm.
The Li–In phase diagram has been exhaustively re-examined by Alexander et al., using thermal and X-ray diffraction analysis. The work has confirmed the liquidus data of Grube and Wolf and delineated the solid-state relationships. Eleven new phases (Table 2), together with the previously known Liln phase (which extends from ca. 46 to between 55 and 63 mol % Li, depending on temperature), have been observed. The discovery of new phases, of which only five are stable at room temperature, has removed the apparent anomaly between the Li–In and the Li–Ga and Li–Tl systems. The solid solubility of Li in In is low (ca. 1.5 mol % Li at 432 K) and that of In in Li is very small.
Intermetallic phases of the Li–Pd and Li–Pt systems have been synthesized and their structures elucidated; pertinent structural data for seven Li–Pd phases (including Li2Pd and LiPd), as determined in X-ray diffraction studies, and for Li2Pt and LiPt, as determined using neutron-diffraction techniques, are collected in Table 3.
Thermodynamic properties of liquid Li–Tl and of liquid Na–X (X = Cd, Hg, In, Tl, Sn, Pb, Sb, Bi, S, Se, or Te) have been studied. The unsymmetrical form of the nature of the dependence on concentration of the thermodynamic characteristics of the Li–Tl system, which exhibits negative deviations from Raoult's Law, is thought to be consistent with the equilibrium diagram. The dependence on concentration of the entropy of mixing in the Na–X systems is S-shaped, the point of inflexion corresponding to formation of intermetallic compounds. This behaviour is attributed to a high degree of short-range order in the liquid, and of partial ionic character of the bonds in these intermetallic compounds. Short-range order has also been studied in liquid Li–Pb solutions by neutron-diffraction techniques. The data indicate a preference for unlike nearest neighbours; this is manifest in a reduction of distance between unlike neighbours (0.295 nm) as compared with the mean distances between the pure components (Li ··· Li = 0.300 nm; Pb ··· Pb = 0.340 nm). It has been suggested that the short-range order is probably due to salt-like Li ··· Pb bonding. No evidence for the existence of isolated Li4Pb clusters was obtained; indeed, in liquid Li4Pb, each Pb atom is surrounded by ca. 10 Li atoms.
(Continues...)
Excerpted from Inorganic Chemistry of the Main-Group Elements Volume 5 by C. C. Addison. Copyright © 1978 The Chemistry Society. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
