Index of content:
Volume 32, Issue 2, 01 February 1960
32(1960); http://dx.doi.org/10.1063/1.1730693View Description Hide Description
The trace of the derived polarizabilitytensor has been measured on an absolute scale for the ions CO3 =, NO3 —, PO4 —3, SO4 —2, and ClO4 — in aqueous solution. From these and the known normal coordinate matrix elements, the values of the derivative with respect to bond length have been calculated. Results are interpreted in terms of bond order, and indicate appreciable π bonding in the ions NO3 —, SO4 —2, and ClO4 —. In these cases, as opposed to hydrocarbons, a π electron makes a larger contribution to the derived bond polarizability than does a σ electron.
32(1960); http://dx.doi.org/10.1063/1.1730694View Description Hide Description
A method for growing large single crystals of metal chelate salts such as NaNi triacetylacetonate·p‐dioxane and NaNi triacetylacetonate·benzene has been devised. The growth habits of these crystals are described. X‐ray studies of three of the NaNi and NaMg salts indicate a diffraction symmetry consistent with space group , with two molecules per unit cell. Paramagnetic resonance measurements by M. Peter indicate that the benzene compounds are truly trigonal, while the dioxane compounds are macroscopically trigonal as a result of disordering among unit cells that are individually monoclinic.
32(1960); http://dx.doi.org/10.1063/1.1730695View Description Hide Description
Hall effect and electrical resistivitymeasurements as a function of temperature to ∼450°C have been carried out on a polymer of copper phthalocyanine. From the Hall effect data it is concluded that intrinsically the polymer of copper phthalocyanine is P type. Two states of this polymer have been determined, one has an activation energy of 0.26 ev and a carrier density ∼1016/cm3 (the carrier concentration varying with temperature), a second, a heat‐treated polymer, also shows P type conductivity but the carrier density is ∼1018/cm3 and is constant in the range of temperaturemeasured. The mobility in the heat‐treated polymer is found to increase with increasing temperature.N type conductivity observed on the first structure is believed to be a consequence of impurities.
32(1960); http://dx.doi.org/10.1063/1.1730696View Description Hide Description
The Kirkwood nonlinear integral equation for determining the optimum free volume of a liquid is integrated by an iterative procedure with the approximation that the cell distribution function is spherically symmetrical. The potential of the system is assumed to be pairwise additive and the interaction between two molecules obeys the Lennard‐Jones 12–6 potential. To simplify the computations, the contributions to the cell potential by molecules beyond the first shell of nearest neighbors are calculated as though the molecules were located at their lattice points. The method of Levine, Mayer, and Aroeste was used to optimize the lattice dimensions and take into account the existence of holes. Extensive tables are given for the configurational internal energy, compressibility factor, and configurational entropy. Comparison of the configurational internal energy and compressibility factor with the Monte Carlo calculations indicates that our isotherms correspond to extensions of the crystalline phase into a metastable low density region. Comparison of our configurational entropy with experimental data indicates that we are lacking the entropy of fusion. Our values of the configurational internal energy agree rather well and our values of the compressibility factor agree very poorly with the experimental data. Comparisons are also made with the results of other theoretical treatments.
32(1960); http://dx.doi.org/10.1063/1.1730697View Description Hide Description
Molecular weight decrease by mechanical shearing results when solutions of around 10% of polyisobutene having average molecular weights above 500 000 are forced through a capillary at nominal rates of shear above 10 000 sec—1. Comparison of observed plots of shear load vs duration of shearing at fixed rates of shear during this degradation process with the corresponding estimated plots which would be expected to obtain if degradation did not occur provide a means of evaluating the amount of applied shearing energy which is dissipated by the degradation process. The result is several hundred thousand kilocalories per mole of broken bonds, which is several thousand times the bond energy of carbon‐carbon bonds. This finding is consistent with the hypothesis that whenever a bond breaks the system loses much of the free energy temporarily stored in bonds and macromolecular chains located in a comparatively large volume surrounding the broken bond, these bonds and chains having been involved in concentrating the required activation energy into the ruptured bond.
32(1960); http://dx.doi.org/10.1063/1.1730698View Description Hide Description
The emission spectrum of the anthracene‐sym‐trinitrobenzene complex has been investigated in a solid solution in a diethylether‐isopentane (EP) glass at 77°K. The observed luminescence consists of two distinct electronic transitions: one transition is the reverse of the charge‐transfer E←N absorption of the complex; the other transition is a phosphorescence very similar to the T→S emission of anthracene in EP, but blurred somewhat in vibrational structure, blue shifted by 113 cm—1, decreased in lifetime by a factor of about 2 and showing decreases in the ground state vibrational frequencies of a few percent. These results, as opposed to previous conclusions (vide infra), would suggest that the phosphorescence is truly a T→S process of the complex which is largely localized on the anthracene component.
Vibrational analyses of T→S spectra of anthracene in EPA and in EP and of the T→S and E→N spectra of the complex in EP at 77°K are given.
32(1960); http://dx.doi.org/10.1063/1.1730699View Description Hide Description
A collision cross section for radial combination reactions is derived which indicates that translationally hot free radicals combine faster than the average thermal radicals. The cross section is used to calculate the disturbance of the equilibrium distribution of velocities by radical combination reactions. It is concluded that the equilibrium assumption of chemical kinetics is justified for photochemical systems where the radical concentrations are low, but may be violated in systems where free radical concentrations approach 0.1 mole fraction.
32(1960); http://dx.doi.org/10.1063/1.1730700View Description Hide Description
Average concentrations of trapped free radicals condensed from a gaseous mixture are calculated on the basis of a statistical model that assumes these concentrations to be those obtaining in the solid for which the free radicals and diluents are distributed in such a way that the entropy of the system is a maximum. This entropy is calculated in an independent pair, Bethe approximation. All kinetic and energetic considerations are neglected. The results are given as a function of diluent concentration and of the number of nearest neighbors in the solid. When condensed from a completely dissociated gas without diluents the free radical fraction ranges from 0.12 for a simple cubic lattice to 0.08 for a face‐centered cubic lattice.
32(1960); http://dx.doi.org/10.1063/1.1730701View Description Hide Description
The radicals produced by gamma irradiation of polyvinyl chloride powder in various gases and in vacuum are observed by electron spin resonance. Different patterns are observed for the irradiation in vacuum, air, and sulfur dioxide. No electron spin resonance absorption is obtained by the irradiation in nitrogen oxide, halogens, and hydrogen sulfide. The recombination reaction velocity of the radical produced in vacuum is measured at different temperatures to get the activation energy 38 kcal/mole.
32(1960); http://dx.doi.org/10.1063/1.1730702View Description Hide Description
Computations of the interaction between a hydrogen atom and a hydrogen molecule, where one of the three atoms is in a two‐quantum electronic state, have been made using a modified Heitler‐London procedure. The calculated interaction energies of the system including a degenerate state for which the magnetic quantum number is m=0 are about 5 ev lower than the previous Heitler‐London results in which no excited state were considered. The results are compared with experimentally determined energies, and a possible explanation of the discrepancy between theoretical and experimental values is presented. Calculations are also presented for cases where m=1.
32(1960); http://dx.doi.org/10.1063/1.1730703View Description Hide Description
The threshold law for processes of excitation is difficult to measure directly and lack of experimental evidence has prevented a choice being made between different theoretical treatments. The ionization efficiency curves for the ion‐pair processes, I+ from I2, O+ from O2, and C+ from CO, and for what is here claimed to be auto‐ionization in N2 + and O2 + have been measured. It is shown that these curves measure excitation probabilities and that the threshold law is in all cases a step function of the energy. The implications of this with regard to the assignment of dissociation products and the measurement of appearance potentials are discussed.
32(1960); http://dx.doi.org/10.1063/1.1730704View Description Hide Description
The ionization of HCl by electron impact is studied in a mass spectrometer with a pseudomonoenergetic electron beam. The ionization potential of HCl is determined as 12.56±0.1 ev. The ionization curve at threshold exhibits excessive curvature which is attributed to the 2Πi doublet ground states of the ion estimated to be separated by 0.3 ev. A sharp change in slope of the ionization curve at 1.6 ev above threshold is attributed to the onset of ionization of the A 2Σ+excited state of HCl+.
32(1960); http://dx.doi.org/10.1063/1.1730705View Description Hide Description
The angular dependence of the protonmagnetic resonance absorption of a Rochelle salt single crystal has been measured. The experimental second moments are compared with the theoretical curves, as calculated for the structures proposed so far. Lösche's model was found to reproduce the qualitative form of the angular dependence, whereas the other structures gave no agreement at all.
The occurrence of transitions in the proton magnetic absorption line widths at +24°C and below —20°C demonstrates that the assumption of protonic motion, upon which the dynamic theories of ferroelectricity in Rochelle salt are based, is essentially correct.
32(1960); http://dx.doi.org/10.1063/1.1730706View Description Hide Description
Galvanostatic measurements, made on spherical silverelectrodes in highly purified aqueous acidsolutions of silver perchlorate at current densities up to 10 amp cm—2, gave overpotential as a function of time and Faradaic current density at constant total current. Variables were total current density, concentration of silver ions and state of the electrodesurface.
The theory of the surfacediffusion kinetics was developed to encompass high current density regions and the surface adion concentration (c ad 0) evaluated.
A theoretical analysis of the steady state of surfacediffusion control of the metal‐ion exchange reaction shows that the current is distributed nonuniformly between growing sites, with constant potential over the electrodesurface.
An equation previously developed by Rojter, Juza, and Polujan has been modified and shown to be applicable to galvanostatic transients when the transfer reaction is rate controlling. The symmetry factor β, of electrode kinetics, is shown to be potential dependent, and, in systems of sufficiently high i 0 values, to tend to zero at high overpotential. This results in a limiting current density for a transfer controlled process. The tendency towards such a current was observed and shown to be quantitatively consistent with the discussion of β.
32(1960); http://dx.doi.org/10.1063/1.1730707View Description Hide Description
Chlorine pure quadrupole spectra have been observed in silver chlorite and sodium chlorite at room temperature. The observed coupling constant of 108.16 mcps for silver chlorite is discussed in terms of a simple model of the electronic structure, in which the magnitude of the apparent coupling constant may be understood as resulting from a highly asymmetric field gradient ellipsoid and from residual positive charge on the chlorine atom.
32(1960); http://dx.doi.org/10.1063/1.1730708View Description Hide Description
Transport and thermodynamic calculations for gases at high temperature call for reliable interaction potentials at close approach. Potentials for the noble gases can be extrapolated approximately, yet with some reliability, from the empirical fit of properties in the ordinary range, using a screened Coulomb repulsion and a dispersion attraction reduced by a suitable factor at close approach. Parameters for this potential can be obtained approximately from the known parameters for the exp‐six potential. The resulting empirical coefficient for the Coulomb term is found to be of the general order of magnitude given by the product of the nuclear charges. If the theoretical Coulomb repulsion is used, the screening exponent may alternatively be chosen on the basis of other empirical parameters.
32(1960); http://dx.doi.org/10.1063/1.1730709View Description Hide Description
A detailed analysis of the electron spin resonance spectrum for the potassium salt of m‐dinitrobenzene negative ion in DME is presented. Isotopic substitution has been performed in various positions in the molecule. m‐Dinitrobenzene negative ion exhibits only one nitrogen hyperfineinteraction of 9.0 gauss, and possesses protonhyperfineinteractions of 4.6 gauss for the number 2, 4, and 6 protons and 1.2 gauss for the number 5 proton. The hyperfine spectrum is different for each of the alkali metals, Li, Na, K, Rb, and Cs, but still retains only one nitrogen hyperfineinteraction. An enriched Li6metal salt of m‐dinitrobenzene indicates no alkali metalhyperfineinteraction.
32(1960); http://dx.doi.org/10.1063/1.1730710View Description Hide Description
The spectra of positively charged fragments from beta decay of tritium in primary and secondary monotritiated propane and in o‐, m‐, p‐, and α‐monotritiated toluene have been determined by mass spectrometric techniques. Although the number of products from decomposition of each molecule is quite large, the mass patterns from the two propanes are fairly similar while the four mass spectra from the isomeric tritiated toluenes are very nearly the same. These results suggest that the influence of the location of the radioactive atom in the molecule on the fragmentation following beta decay is often obscured by secondary processes, such as radiationless transitions, H‐atom migration, and other rearrangements of the excited molecular ion before breakup. However, there are significant differences within the two groups of mass spectra. Some of the differences in the propane patterns are consistent with the greater stability of the secondary propyl structure as compared to that of the primary isomer. Energetically plausible decomposition schemes for the tritiated propanes are proposed. The distributions of excitation energies made available to the daughter molecular ions (C3H7He3)+ in the beta transformation of H3 to He3 are estimated.
32(1960); http://dx.doi.org/10.1063/1.1730711View Description Hide Description
Thermal conductivity of the binary mixtures of O2–He, O2–Ne, O2–Kr, and O2–Xe have been measured at 30°C and 45°C for various compositions by using the thick‐wire‐variant of the hot‐wire method. For pure O2 the experimental value of the thermal conductivity is lower than that given by the recent theory of Hirschfelder based on the assumption of local chemical equilibrium, and consequently the values calculated from Hirschfelder's formula for the mixture conductivities are also expected to be higher than the experimental values which is actually found to be so for the mixtures investigated here. The agreement becomes excellent when the experimental values of the conductivity of pure O2 are used to calculate the mixture conductivities. The formula for the mixture conductivities has been tested in detail and was found to give satisfactorily the composition dependence. The simpler formula for the thermal conductivity of gas mixtures given by Mason and Saxena has also been tested.
Molecular Diffusion Studies in Gases at High Temperature. IV. Results and Interpretation of the CO2–O2, CH4–O2, H2–O2, CO–O2, and H2O–O2 Systems32(1960); http://dx.doi.org/10.1063/1.1730712View Description Hide Description
Experimental measurements of the binary diffusion coefficients for the CO2–O2, CH4–O2, H2–O2, CO–O2, and H2O–O2 systems over the nominal temperature range 300–1000° K are reported. The measurements were made by the point source technique. The data have been analyzed in terms of intermolecular potential energies and the rigorous kinetic theory by using the Lennard‐Jones (12–6), modified Buckingham (Exp‐6), point center of repulsion (inverse power), and the exponential repulsion potential energy functions. It is shown that all four of these functions are capable of fitting the diffusion data with about the same precision (within experimental error). However, the potential energy values themselves which are calculated from these four different fitted models are shown to be widely at variance; especially in the range of interaction corresponding to the lower temperatures. Thus, these results provide an example of the high degree of ambiguity (previously noted by several others) involved in deducing intermolecular potential energies from transport data. The use of such fitted potentials to extrapolate transport data to higher temperatures, on the other hand, is probably a more reliable procedure, and diffusion coefficients for the above five gas pairs (in their unexcited, undissociated states) have been estimated to 3000° K from the potentials based on the 300–1000° K measurements.