Volume 34, Issue 1

The crystal structure and electrical properties of [N,N′-di(n-propyl)-4,4′-bipyridylium] 2+ -(7,7,8,8-tetracyanoquinodimethane) 4 2- , (DPrBP) 2+ (TCNQ) 4 2- , are reported. The complex is monoclinic with the space group P2 1 /c and lattice constants a = 13.334(1), b = 25.954(3), c = 7.877(1) Å, β = 93.66°, Z = 2. The single crystal dc resistivities at 300 K along a, b and c are 3300, 200 and 800 Ω cm respectively with activation energies of 0.36, 0.22 and 0.30 eV. The existence of a hopping mechanism in addition to a frequency independent conduction process is indicated by a. c. conductivity data obtained on a compressed disc between 10 and 10 5 Hz.

On the basis of the phosphorescence spectrum of dinaphtho-(2′.3′:1.2); (2″.3″:6.7)-pyrene in n-dodecane (Shpolskii matrix) at T = 1.3 K a site and vibrational analysis was performed. With the help of ODMR measurements the symmetries of the vibrational bands were determined. In this way for most of the normal modes an assignment of the type of vibration became feasible. Because of their different zero-field splittings (ZFS) five sites could be identified. The signs of the ZFS parameters as well as the kinetic rate constants of population and depopulation of the zero-field levels (ZFL) of T 1 were determined. Like the ZFS the kinetic rate constants too were different for each site reflecting different interaction of the molecules with the host matrix. Radiative and radiationless processes are discussed in terms of spin-orbit coupling (SOC).

The charge-transfer effects of systems of the general type A-B-B are investigated, A being the donor. The importance of a one-to-one correspondence between Hückel-parameters and parameters such as effective nuclear charges and bond distances is pointed out. By using charge-transfer effects as an example it is shown that non-uniqueness leads to difficulties in interpreting charge-transfer effects.

The adsorption of hydrogen on a clean Ni(100) single crystal surface was studied in UHV between 120 K and 500 K means of LEED, thermal desorption spectroscopy and work function measurements. H 2 chemisorption occurs in two atomic states β 1 and β 2 with maximum desorption rates (at saturation) at 318 K and 348 K. The activation energy for desorption was determined to be 20 (± 2) kcal/mole for the β 1 -state and 23 (± 1) kcal/mole for the β 2 -state, in excellent agreement with previous results. Saturation at 120 K corresponds to a coverage of 0.9 20%). The initial sticking probability is 0.06 (± 0.01) and the rate of adsorption follows second order kinetics. No additional LEED spots arose from the H overlayer, indicating the absence of long-range order even at 120 K. The work function results clearly show the existence of two different binding states: Whereas the β 2 -population gives rise to a linear increase by 100 meV (at 120 K) the ß 1 -state leads to a decrease of ⊿ϕ to a final value of 30-40 meV.

The formation (by interaction with ammonia) and decomposition of nitrides on clean Fe surfaces was studied by means of Auger electron spectroscopy, x-ray photoelectron spectroscopy, thermal desorption spectroscopy, and scanning electron microscopy. The N atoms may exist in various forms with quite similar electronic properties, viz. as chemisorbed layer (= "surface nitride"), dissolved in α-Fe or γ-Fe, as γ′-nitride (= Fe 4 N) or as e-nitride, depending on temperature as well as pressure and duration of interaction with NH3. There is no noticeable chemical shift of the ionization energies of the Fe core levels, indicating that the bond is essentially covalent. The activation energy for the decomposition of e-nitride into Fe 4 N + N 2 is about 27 kcal/mole, that for the decomposition of Fe 4 N into Fe+N 2 ranges between 51 and 57 kcal/mole, depending on the mode of preparation. The latter values are identical to those found previously for the desorption of N2 from various Fe single crystal planes and indicate that the decomposition of the chemisorbed "surface nitrides" is the rate-limiting step which prevents the spontaneous decom-position of the metastable bulk iron nitrides.

The onset of the L III edge of the X-ray absorption spectrum of platinum in F-zeolites involves an electronic transition from core to band levels producing a so-called white line. The spectral area of this increases and the edge position shifts to higher energies, first when the platinum particle size becomes small enough so that the clusters are in the supercages, second when the clusters are covered with oxygen, and third when multivalent cations are introduced in the zeolite. Both features of the absorption edge indicate that under the three above circumstances, platinum becomes more electron deficient. This conclusion had been inferred earlier from the enhanced catalytic activity of Pt clusters in Y zeolites.

A Clusius’ separation column of hot wall catalytic activity was applied as a reactor to produce highly concentrated hydrogen. Satisfactory results are obtained in the decomposition of hydrogen sulfide and the conversion of carbon monoxide with water.

The onset, eδ 0 , in the energy distribution of field desorbed silver ions (from Ag deposit on a W-tip) has been measured at T > 400 K. With increasing field strength (0.5 < 10 10 V/m < F < 1·10 10 V/m) or with decreasing temperature (800 K > T > 500 K) eδ 0 increases up to 1.5 eV. For a comparison with the eδ 0 -values, activation energies, Q, for field desorption have been determined from Arrhenius plots (0.5 eV < Q < 2 eV). Two different kinetic regimes (desorption and diffusion) were identified. For field desorption, the sum eδ 0 + Q is constant, irrespective of F-values. In the frame of "image hump" or "charge exchange" models, field dependent activation barriers are discussed. Temperature dependencies are related to (i) the Fermi distribution of the emitter, (ii) the distribution of F and (iii) the distribution of surface bond energies. Results are consistently explained in a three-dimensional model with directionally distinguished desorption steps.

The phase diagram of TmSe has been measured in the composition range 0.6 ≦ x ≦ 1.2 (x = mole Tm/mole Se) up to 2300 °C. Both phase diagram and lattice constants show the existence of a gross homogeneity range at 0.90 < x < 1.05. X-ray and electron diffraction at room temperature confirm the homogeneity of this phase and show no superstructures in this range but only a correlation of vacancies. A dramatic change of the lattice constant (1.7%) with nonstoichiometry shows that the latter controls the valence of Tm between 3 + and 2.71+. Density measurements indicate that both the solution of excess Tm or Se in the lattice leads to the formation of vacancies in the other sublattice. The experimental results can be explained by the existence of approx. 1% Schottky defects for all samples and nonstoichiometric vacancies increasing with deviations from stoichiometry up to 10% for the Se-rich phase boundary (each tenth lattice site vacant!). With increasing temperature the nonstoichiometric samples show two phase transitions: the first at approx. 1100°C for Tm-rich samples or approx. 1700°C for Se rich samples (the stoichiometric sample shows both transitions); the second for all samples at 2003°. At 2020 °C a eutectic appears at the composition of x = 1.00, its components melting at 2032 °C and 2028 °C. This shows that in the temperature range between the phase transitions and the eutectic temperature a decomposition of the homogeneity range must take place. This decomposition might be due to a valence segregation, the two components of the eutectic possibly being Tm 2,71+ Se and Tm 3+ Se. Various measurements indicate changes of the lattice stability in TmSe as a function of stoichiometry. Thus stoichiometric samples show a minimum melting point and solution-reaction calorimetry (with 4N HCl) indicates that the composition with the highest melting point (Tm 3+ Se) has the highest stability. As it is briefly discussed the unstable states may have important implications as active states for surface phenomena and catalysis. Further change of the Tm-valence in a TmSe-containing compound exceeding the range Tm 3+ ↔Tm 2,71+ may be achieved by mixed-crystal formation. The lattice pressure in TmSe is changed by chemical substitution leading among others to the interesting mixed crystal systems TmSe 1-x Te x and Tm 1-x Eu x Se. The first shows two pressure-induced phase transitions accompanied by a change of the Tm-valence from 2.25 + → 2.52 + and 2.52+ → 2.93 + . Density measurements show that when the lattice collapses the number of vacancies nears 10%.

The diffusion of silica(Aerosil)-hydrosols has been studied by photon correlation spectroscopy. For data analysis the method of cumulants, proposed by Koppel, was used. Effects of charge and hydratation of the silica particles and their complicated aggregated structure required an extrapolation to the solconcentration c s = 0 and the scattering angle θ = 0. The described data analysis combined with the methodical proceeding - using Aerosil aggregates as a model system - shows how to evaluate the diffusion coefficient and its normalized variance. These data of characterization are important for many kinds of aggregated particles in technical applied dispersions.

Isosteric adsorption heats have been determined from the temperature dependences of equilibrium pressures, which in turn have been obtained from sets of very precisely measured adsorption isotherms. The dependence of the isosteric adsorption heat on the coating of the silver surface shows a remarkable structure: whenever a layer of adsorbate is filled up, inner rotations are inhibited, which results in a peak in the adsorption heat-coating curve. The adsorption isotherms of toluene, adamantane and benzotrifluoride on highly dispersed silver were measured with an ultra-sensitive adsorption balance. The nearly spherical adamantane enabled an exact determination of the specific surface area of the adsorbent. From their respective space requirements it followed that toluene is adsorbed in ortho- and meta-positions, whereas benzotrifluoride is adsorbed with its three F-atoms on the silver surface. The energy of rotation of the CH 3 -and CF 3 -groups could also be determined.

The reaction of oxygen and CO to form CO 2 has been investigated using an Ir (111) surface as an acting catalyst. Both instationary and stationary reaction processes have been established via separate gas exposing techniques. The instationary reaction process, achieved from coadsorbed CO and O which per se is an LH reaction is found to be controlled by an apparent activation energy of 10.7 kcal/mole. The stationary reaction with both CO and O 2 continuously present in the gas phase has been simulated using a proper computer program, involving both LH and ER reaction steps. By comparison with experimental results, close agreement is found when ruling out any ER reaction step from the reaction path.

From the sulfur-and selenium-concentration profiles recorded by an electron microprobe the chalcogen interdiffusion coefficient of the system Pb(S, Se) is determined as a function of mass fraction c Pbse and temperature. Undoped crystals and Pb-or chalcogen-saturated crystals are studied. The dependence of the interdiffusion coefficient on the deviation from stoichiometry shows that in the chalcogen-saturated part of the system the calcogen interdiffusion proceeds via chalcogen interstitials, whereas in the Pb-saturated part of the system the chalcogens diffuse via chalcogen vacancies. For pure PbSe it is shown that these vacancies are doubly ionized. The activation entropy of the diffusion of the chalcogens depends linearly on the ratio of activation enthalpy to solidus temperature.

Light and heavy ammonia were decomposed on a clean tungsten foil at 973-1123 K and at p NH₃ , ND₃ ≈ 0.2 Torr. The nitrogen uptake and the kinetics of the formation of nitride layers were examined during the course of the reaction. The amount of nitrogen chemisorbed (or the thickness of the nitride layers formed) at higher pressures was found to be appreciably larger from NH 3 than from ND 3 . Addition of hydrogen had no effect on the thickness of the nitride layer nor on the rate of ammonia decomposition. These facts support the concept of a dynamic balance between two consecutive rate determining steps: supply of chemisorbed nitrogen from ammonia into the surface and consumption of the chemisorbed nitrogen through its desorption from the surface to form nitrogen molecules.

The well known formulas for computing the partial molar functions from a given mean molar function are treated as deferential equations for computing the mean molar function from any given partial molar function. Solutions do not depend on the number of components, but only on the choice of three indices: the index d of the dependent mole fraction xa to be eliminated prior to any computations, the index j of a pivot mole fraction x j and the index i of the partial molar function yi. An arbitrary number of additional mole fractions of the other components safe x d may be linked to the pivot mole fraction x j . The simple solution: y = (x j - δ ij ) I ij , yi = (x j - δ ij ) 2 X ij and X ij = d I ij /dx j holds for an arbitrary number of components, if the (c - 2) mole fractions x j safe x d and x j are transformed to new variables found from the auxiliary equations. Three different cases arise if either i = d, i = j or i ≠ d, i ≠ j is chosen. Formulas for the three sets are provided. As an example a simple interpolation formula for ternary systems is discussed.

The fluorescence spectra of anthracene, 9-methyl- and 9,10-dimethylanthracene in n-hexane at 77 K were measured. The Franck-Condon-Factors are evaluated using equilibrium geometries, vibrational frequencies and normal coordinates of the molecules in the ground and first excited singlet state which were calculated by the QCFF/PI-method. These results are then compared with experimentally determined frequencies and estimated intensities. The new vibronic lines in the fluorescence spectra of the methylated anthracenes correspond to totally symmetric vibrations, up to 1000 cm -1 originated mostly by mixing with out-of-plane vibrations of the atoms of the anthracene skeleton. Above 1000 cm -1 these vibrations which have greater projections of their normal coordinates on the internal coordinates of the methyl groups, are not visible as vibronic lines in the fluorescence spectra. The multiplet structures of the O-O-transitions are ascertained.

Ab initio RHF-MO calculations are made for the linear NiCO and CuCO clusters. The ground states of them involve Ni 3d 9 4s 1 and Cu 3d 10 4s 1 electron configurations, respectively. The orbital energies of the CO 1 π-like levels in both clusters are closely similar to each other. The CO 5σ-like level shifts toward the CO 17r-like level due to the bonding interaction of the 5σ with the metal 4s and 3d Z 2 . The results are discussed in relation to the difference of the 3d-level positions and the similarity of 4s levels of Ni and Cu. For NiCO both the 4s and 3d orbitals contribute to the metal-CO binding while for CuCO only the 4 s orbital contributes to it. The present results are in agreement with the UPS experiments on adsorption.

Oblique illumination, although accompanied by distortion, provides a simple method of obtaining stereoscopic images in a shadow electron microscope. It is not necessary to tilt the specimen.

13 C/ 12 C variations of organic carbon have been investigated in an undisturbed forest soil with perched surface water. The measurements reveal a very distinct decrease of the 13 C-content with increasing depth of the soil horizons. Within 20 cm the δ 13 C-value is depleted by 1.5/ 0 00 . At present it is however not possible to verify, whether this behaviour is characteristic for undisturbed soils in general.