Index of content:
Volume 21, Issue 2, 01 February 1953
21(1953); http://dx.doi.org/10.1063/1.1698854View Description Hide Description
Methods are presented which make possible quantitative spectroscopic measurements at liquid nitrogen temperatures from the visible region to as far as 1670A in the ultraviolet region.
21(1953); http://dx.doi.org/10.1063/1.1698855View Description Hide Description
It is pointed out that serious difficulties are encountered in preparing organic compounds of the transition elements. The hypothesis is proposed that these difficulties are caused by insufficient stability of the carbon‐metal bonds involved. Some rough calculations based on Mulliken's ``magic formula'' are used to test this hypothesis. The results obtained indicate that carbon‐metal bonds formed by the transition elements are extremely weak. The calculations further suggest that carbon‐metal bonds involving the alkali and alkaline earth metals are largely ionic.
21(1953); http://dx.doi.org/10.1063/1.1698856View Description Hide Description
Polar ions are compared with well‐known uncharged dipoles with respect to their polarizability and the contribution to the dielectric constant of their solutions. It is shown that (1) hydrodynamic forces exerted by the solvent on the solvated ion have to be taken into account in any kinetic discussion of dielectric behavior. (2) For ions having a symmetrical shape the polarizability is the same as that of an uncharged dipole obtained by adding, at the center of symmetry of the ion, a neutralizing charge. (3) The same effective dipole enters into the thermodynamic considerations of unsymmetrical ions. (4) The dipole analogy may be extended to the general case, leading to a rigorous but implicit expression. (5) As a first approximation, the polarizability of an ion is an indication of its dissymmetry.
21(1953); http://dx.doi.org/10.1063/1.1698857View Description Hide Description
The effect of interactions between the segments of a polymer chain on the size of the molecule has been calculated. Two approaches have been used. The first is a direct probability calculation using an exponential interaction energy between segments. The second follows the original approach of Flory but solves the problem to a different approximation. Both methods lead to essential agreement with Flory's result for small and moderate interactions. A probable reason for the spurious results found by others is proposed.
21(1953); http://dx.doi.org/10.1063/1.1698858View Description Hide Description
Second‐order transitions occur when a system which is completely ordered at low temperatures (degree of order s=1) changes gradually to a disordered state, ultimately reaching s=0 at the critical or Curie temperatureTc . Elementary theories always predict that s 2 ∝ (1 − T/Tc ) and predict limiting heat capacities at Tc which are finite and much smaller than those observed. Experimental data on liquid‐liquid and liquid‐vapor critical phenomena fit a relation in which the cube of s, not the square, is proportional to 1 − T/Tc . A crude argument, using this empirical equation, leads to the conclusion that the heat capacity rises sharply, probably reaching infinity at the Curie temperature. Experimental data on ammonium chloride and on quartz fit a relation in which the anomalous heat capacity is proportional to an inverse power of (1 − T/Tc ).
21(1953); http://dx.doi.org/10.1063/1.1698859View Description Hide Description
The entropies of alkyl and other simple free radicals may be estimated with sufficient accuracy to make informative calculations about the equilibria involved in the reactions of these radicals. Such calculations have been made for metathetical reactions, the decomposition of saturated molecules, and the decomposition of free radicals. The method provides a means of obtaining knowledge about reactions which are not readily susceptible to direct experimental investigation. The deviations of the frequency factors of some unimolecular decompositions from a value of 1013±1 are discussed.
21(1953); http://dx.doi.org/10.1063/1.1698860View Description Hide Description
Normal frequencies of molecular configurations of different azimuthal angles of internal rotation have been calculated for various molecules showing rotational isomerism. It has been shown that one can tell fairly accurately the value of azimuthal angle by comparing the observed frequencies (especially the skeletal deformation frequencies) with those calculated for different configurations. The sum rule for the rotational isomers similar to that for isotopic molecules has been derived and has been shown to be very useful in assigning the observed frequencies. The product rule for rotational isomers derived in our previous paper has also been applied to this problem and has been shown to be useful in determining the azimuthal angles of rotational isomers.
21(1953); http://dx.doi.org/10.1063/1.1698862View Description Hide Description
The infrared spectra, in the rock salt region, of oriented sections of single rhombic sulfur crystal have been obtained. The spectra of sections perpendicular to the c axis of the crystal, the 001 type planes, have been compared, under ordinary radiation with those of sections parallel to the c axis, 110 type planes. Most bands in the region investigated, 700 to 1500 cm−1, show differences in intensity in the two spectra. With a 110 section and radiation polarized at various angles, similar changes in intensity were observed. Both experiments lead to similar conclusions regarding the direction of the oscillating dipole in the vibrations giving rise to the observed bands. Assignments of the infrared active species (B 2 or E 1) in these bands are given but no attempt is made to explain them in terms of the fundamental vibrations.
21(1953); http://dx.doi.org/10.1063/1.1698863View Description Hide Description
The thermoluminescence resulting when calcium fluoride crystals colored by x‐ray irradiation are heated, has been investigated in the region from 3200 to 6500A. First, the form of the emission spectrum was determined by the use of a spectrophotometer in conjunction with a photomultiplier tube as a detector, and fifteen emission bands were observed of which seven weak ones had not previously been reported. The peaks of these bands occurred at 3430, 3650, 3710, 3790, 3810, 4150, 4270, 4370, 4570, 4780, 4890, 5230, 5350, 5430, and 5740A. Second, continuous annealing curves or ``glow curves'' were obtained for the strongest emission bands by raising the temperature of the crystal at a constant rate and recording the intensity of the emitted light at the peak of each band. The maxima of the glow curves thus recorded fall into two groups. In group 1, three maxima located at approximately 197, 285, and 310°C were observed. In group 2, five intensity maxima could be distinguished, the lowest occurring at approximately 70°C and the highest at 245°C. The corresponding activation energies were calculated, and in two cases were checked by carrying out the annealing at fixed temperatures. These results are consistent with the assumption that two distinct types of imperfections are responsible for the thermoluminescence of calcium fluoride. The optical absorption appears to be the result of the imperfections associated with group 2.
21(1953); http://dx.doi.org/10.1063/1.1698864View Description Hide Description
Neutron diffraction patterns have been obtained on liquid lead and liquid bismuth at two temperatures. The patterns differ only slightly at the different temperatures, the main peak being slightly shorter and broader and shifted slightly to smaller angles at the higher temperatures. The diffraction patterns have been analyzed to obtain information on the atomic distribution. The first main peak in the distribution curve for liquid lead is at 3.40A and represents about 9.4 atoms. The first main peak in the curve for liquid bismuth is at 3.35A and represents about 7.7 atoms. The second most prominent concentration of atoms occurs at 6.6A for both lead and bismuth.
21(1953); http://dx.doi.org/10.1063/1.1698865View Description Hide Description
The characteristics of a xenon resonance lamp were studied, using carbon dioxide as an actinometer, to determine the influence of the xenon pressure in the lamp on the intensity of the 1470A radiation. The xenon sensitized photosynthesis of ammonia was studied both in static and in flow systems. No ammonia could be detected. Some experiments on the photolysis of ammonia at 1470A are reported and discussed. The present work furnishes the first direct chemical evidence rejecting the value of 171.3 kcal per mole as the dissociation energy of nitrogen.
21(1953); http://dx.doi.org/10.1063/1.1698866View Description Hide Description
The equation of state of a gas in virial form is derived from the equilibrium constants for the formation of molecular clusters. It is indicated that by choosing effective Lennard‐Jones parameters independently for clusters of various numbers of molecules, the actual virial coefficients may be represented more satisfactorily. Results for the third virial coefficient are shown for A, N2, CO, CO2, CH4, C2H4, CF4, and Xe.
Substituted Methanes. X. Infrared Spectral Data, Assignments, Potential Constants, and Calculated Thermodynamic Properties for CF3Br and CF3I21(1953); http://dx.doi.org/10.1063/1.1698867View Description Hide Description
Infrared wave numbers, relative intensities, and percent transmission curves have been obtained for CF3Br and CF3I in the region 400–2200 cm−1. Assignments are given for the observed bands, in which the assignments of Plyler and Acquista for ν2 and ν3 are interchanged. Potential constants were calculated for both molecules by the Wilson FGmatrix method. And, finally, the heat content, free energy,entropy, and heat capacity for the ideal gaseous state at 1 atmos pressure were calculated for 12 temperatures from 100°to 1000°K, using a rigid rotator, harmonic oscillator approximation.
21(1953); http://dx.doi.org/10.1063/1.1698868View Description Hide Description
By use of thermal diffusion columns, the N2–H2 system has been separated continuously and its thermal diffusion ratio α has been determined with a method of least ambiguity under a pressure range of 1∼12.6 atmos. The α‐values thus obtained, based on the Furry‐Jones theory of the apparatus, have been found to be not constant and also to differ very much in most cases from those hitherto reported. However, the values corrected according to the theory modified by Drickamer et al. have been shown to be reasonable, supporting their modified theory.
21(1953); http://dx.doi.org/10.1063/1.1698869View Description Hide Description
A simple model of a solution is constructed, assuming that the interaction energies of the molecules depend on their mutual orientation. The thermodynamic properties of the model are calculated by the combinatory method of the strictly regular solution. Graphs of the excess free energy,entropy, and heat of mixing are obtained numerically for a number of typical cases.
21(1953); http://dx.doi.org/10.1063/1.1698871View Description Hide Description
The slight reverse curvature in the plot of logP versus 1/T (P,vapor pressure;T, absolute temperature) may be explained on the basis of the nonideal behavior of the vapor together with the change in heat of vaporization with temperature. If it is assumed that ΔH is linear with T and that the van der Waals a/V 2 term is a first approximation to the deviation from the ideal, the following equation may be obtained by integration of the Clapeyron equation:The last term is the nonideal gas correction with constant D=a/2.303R 2. With constants A, B, and C determined empirically, the equation reproduces experimental vapor pressures from the triple point to the critical point with an average deviation of 0.3 percent.
21(1953); http://dx.doi.org/10.1063/1.1698872View Description Hide Description
A method has been developed for measuringdiffusion coefficients in liquids to 10 000 atmospheres pressure.Measurements have been made over this pressure range at 0°, 20°, and 40°C for the system CS2–CSS35. The results are interpreted in terms of the enthalpy,entropy, and volume of activation as a function of temperature and pressure.
21(1953); http://dx.doi.org/10.1063/1.1698873View Description Hide Description
The virial equation of state for gases has been developed with the quantum‐mechanical grand partition function as a basis, instead of with the usual Slater sum or the density matrix. Attention is thereby focused on the energy levels of systems of one, two, three, etc., molecules rather than on their wave functions. An explicit, closed solution has been found for the nth virial coefficient in terms of the first n cluster integrals which is valid either classically or quantum mechanically. A simple generating function for the virial coefficients has been proposed.
21(1953); http://dx.doi.org/10.1063/1.1698874View Description Hide Description
Multiple magnetic resonance lines have been observed for H1, F19, and P31 nuclei in compounds such as PH3, PF3, F2PO(OH), and BrF5, in the liquid state. The multiplets consisted of two to seven equally spaced narrow components, symmetrically placed about a central frequency, and with splittings from 0.02 to 0.8 gauss. These multiplets arise from a new variety of interaction among the nuclear moments in a molecule. Resonance lines were found to be multiple either when a nucleus interacted with a different species of nucleus or when there was interaction between nuclei of the same species with resonance frequencies separated by a chemical shift. No compounds exhibited multiplets attributable to interactions among structurally equivalent nuclei. Nor were multiplets caused by nuclei whose electric quadrupole moments were coupled to a direction fixed in the molecule.
The number and relative intensities of the components of a multiplet were determined by the number and statistical weights of the various nuclear spin orientations of the nuclei causing the splitting. In a given molecule, the ratio of the multiplet splittings of the two different resonance lines was inversely proportional to the ratio of the gyromagnetic ratios of the interacting nuclei. The splittings were independent of applied magnetic field at 4180 and 6365 gauss; they were independent of temperature over ranges from 55°C to −130°C. In PF5, the one gas examined, the splitting of the doublet fluorine resonance was the same in the gas and liquid phases; also, the doublet was demonstrated to arise from coupling with the phosphorus nucleus, rather than from a chemical shift between the resonances from the apex and meridian fluorines in the bipyramidal structure, as proposed earlier.
All of the above characteristics are accounted for theoretically by assuming the magnetic nuclei interact via magnetic fields inside the molecule. The qualitative aspects are predicted by coupling of the form A12 μ 1·μ 2 between the nuclear momentsμ 1 and μ 2. The coupling constant A12 depends upon the detailed mechanism, which must involve the molecular electrons. Second‐order perturbation theory was used to calculate the relative magnitudes of coupling via the electron orbital and the electron spin magnetic moments. The electron spin mechanism was found to give splittings ten to twenty times the orbital. Approximate calculation of the electron spin mechanism in several simpler cases gave good agreement with experiment. The influence upon the splittings of electric quadrupole coupling and spin‐lattice relaxation was considered and is discussed briefly.