Elovich photovoltage decay kinetics observed for the organic semiconductor 3,4,9,10-perylenetetracarboxylic dianhydride

The decay of the photovoltage of thin films of 3,4,9,10-perylenetetracarboxylic dianhydride exhibits an unusual Elovich type kinetics: $\text{?dPV}\text{dt}\text{=B}{}_2\text{exp[}\text{(B}{}_1\text{PV?PV}{}_{\mathrm{sat}}\text{)}\text{PV}{}_{\mathrm{o}}\text{]}$, where B₁ and B₂ are constants. PV_o is the photovoltage at time zero when the light is turned off and PV_sat is measured by saturating the photovoltage with high intensity light and corresponds to the total amount of energy band bending present at the surface. The values B₁ = 5.31, B₂ = 0.109 V/s give the best fit to the data over a wide range of values of PV₀ which were obtained by varying the incident light intensity with neutral density filters. The detailed mechanism which accounts for this Elovich kinetic behavior is not yet known, but is thought to involve charge tunneling at the surface barrier and the presence of slow electronic traps.     

Electron escape effects in photoemission from Ag(001) surfaces

We have observed electron escape effects, separate from excitation probabilities, in photoemission from clean Ag(001) surfaces. By comparison of angle-resolved photoemission data for the (001) surface of silver at several photon energies with a direct transition model based on calculated band structures, we have determined the dispersion $\text{E}\text{(}\text{k}{}_{\mathrm{\perp }}\text{+}\text{k}{}_{\mathrm{\parallel }}\text{)}$ of final states involved in the photoemiss ion process. Observed photoelectrons from Ag(001) are emitted into a single final state band, closely parallelled by the $\text{G}{}_{\mathrm{\parallel }}\text{= 0}$ branch of the nearly-free-electron fcc band structure, even though other final states $\text{(}\text{G}{}_{\mathrm{\parallel }}\text{≠0)}$ are dipole-allowed. This is interpreted as the preferentially strong overlap of only one allowed final state band with states external to the crystal.     

The influence of atomic steps on the photoelectric properties of clean cleaved silicon as derived from the surface photovoltage

The correlation of structural and electrical properties of clean silicon surfaces cleaved in UHV was investigated quantitatively by the surface photovoltage, using light with an energy larger than the band gap of silicon. The surface photovoltage, which is a function of band bending and recombination probability, depends strongly on the appearance of atomic steps. The additional surface states vary with density and crystallographic orientation of the steps as well as with adsorption of oxygen. The experimental facts can be explained by accepting a shift of the Fermi level at the surface towards the valence band due to edge atoms. By measuring the change of sign of the surface photovoltage of crystals with various dopings an exponential temperature dependence of the ratio of the recombination probabilities r_v/r_c for transitions from and into the Surface states has been derived.     

Structure determination of the clean Co(112̄0) surface by LEED

The atomic structure of the (112&#x0304;0) surface of cobalt has been determined by LEED using six intensity spectra at normal incidence. The surface exhibits the truncated bulk structure with a contraction of the first interlayer spacing by about 8.5% with respect to the bulk value. Quantitative evaluation of the LEED spectra was done using Zanazzi and Jona's and Pendry's r-factors. The minimum averaged r-factors are $\text{r}{}_{\mathrm{<mtext>ZJ</mtext>}}\text{= 0.09}$ and $\text{r}{}_{\mathrm{<mtext>P</mtext>}}\text{= 0.22}$. No change of the interatomic distances within the plane could be detected and no rearrangement of the surface structure takes place up to temperatures shortly below the transition temperature.     

Relation between classical and quantum formulations of the Franck-Condon principle: The generalized r-centroid approximation

The classical transition point $\text{r}{}_{\mathrm{v\prime v″}}{}^1$ for the (v′, v″) band is introduced as the point where the kinetic energy does not change during the transition (classical statement of the Franck-Condon principle). It is shown that when the r-centroid approximation is valid, it implies that the r centroid $\text{r}{}_{\mathrm{v\prime v″}}$ equals $\text{r}{}_{\mathrm{v\prime v″}}{}^1$, allowing a new connection to be made between the classical and quantum statements of the Franck-Condon principle. The r-centroid approximation fails for some classically allowed bands having large Franck-Condon factors. This occurs when the (v′, v″) band has more than one classical transition point. A generalization of the r-centroid approximation permits such bands to be used in determining the variation of the electronic transition moment with internuclear distance.     

The surface structure of V(100)

The structure of the clean V(100)?(1×1) surface is determined, based on an r-factor comparison of experimental LEED intensity-energy spectra with the results of multiple-scattering model calculations. Minimization of the r-factor with respect to the calculational variables leads to optimum values of the first and second interlayer spacings of $\text{d}{}_1\text{=1.41 ± 0.01}\text{A}\text{?}$ and $\text{d}{}_2\text{=1.53 ±0.01}\text{A}\text{?}$, corresponding respectively to a contraction of 7% and an expansion of 1% with respect to the bulk value of $\text{d}{}_{\mathrm{<mtext>B</mtext>}}\text{=1.5141}\text{A}\text{?}$. Preliminary studies of the adsorption of O₂ and CO confirm that the V(100)?(5×1) structure observed during the process of cleaning the crystal is not characteristic of the clean surface, as suggested recently by Davies and Lambert (Surface Science 107 (1981) 391), but rather is associated with the presence of a significant concentration of oxygen in the surface region.     

Intensity measurements and relative band strengths of the fluorescent bands of the (B-X) system of I2

Relative integrated intensities of a few bands in the vibrational structure of the B-X fluorescent system of molecular iodine have been measured by the technique of photographic photometry. Reliable values of Franck-Condon factors and r-centroids given recently by Tellinghuisen have been employed to evaluate the variation of electronic transition moment with internuclear separation. The variation is found to be $\text{R}{}_{\mathrm{e}}\text{(}\text{r}\text{?}\text{) =}\text{const.}\text{(1?0.693}{}_0\text{r + 0.120}{}_1\text{r}{}^2\text{)}$ in the range 2.76 < r < 2.81 Å. Smoothed arrays of band strengths are presented for the bands. The effective vibrational temperature of the source is found to be 340 ± 55 K.     

The vibration frequency of an atom at the surface of a tight binding metal

The frequency of vibration ω⊥ of an atom normal to the surface of a tight binding metal is calculated in the Einstein model, and is compared with the bulk value ω₀. The value of $\text{(ohms}{}_0{}^2\text{?ohms}{}_{\mathrm{\perp }}{}^2\text{)}\text{ohms}{}_0{}^2$ is found to be greater than the value expected from a nearest neighbour central force model for a small number of electrons or holes in the band, whilst for a half filled band it is lower. Equality is obtained at the band filling for which there is zero surface relaxation. For the (001) surface of a simple cubic lattice this occurs for 0.5 and 1.5 electrons per atom in the band. These results appear to be consistent with the behaviour of Cr in the middle of the transition metal series, and of Ni, Pd and Pt at the end. The possible effect of the spill out of s electrons at the surface, not treated in the model, is briefly discussed.     

Derivation of the classical theory of correlations in fluids by means of functional differentiation

The k-particle, infinite-volume distribution functions $\text{n}\text{?}{}_{\mathrm{k}}\text{(}\text{r}{}_1\text{, …,}\text{r}{}_{\mathrm{k?1}}\text{, γ)}$ and modified Ursell correlation functions $\text{U}\text{?}{}_{\mathrm{k}}\text{(}\text{r}{}_1\text{, …,}\text{r}{}_{\mathrm{k?1}}\text{, γ)}$ of a classical system of particles with the two-body potential $\text{q(}\text{r}\text{) + γ}{}^{\mathrm{\nu }}\text{K(γ}\text{r}\text{)}$ are considered. The limiting values of the functions $\text{n}\text{?}{}_{\mathrm{k}}\text{(}\text{r}{}_1\text{, …,}\text{r}{}_{\mathrm{k?1}}\text{, γ),}\text{n}\text{?}{}_{\mathrm{k}}\text{(}\text{S}{}_1\text{/γ, …,}\text{S}{}_{\mathrm{k?1}}\text{/γ, γ)}$ and $\text{γ(}{}^{\mathrm{1?k}}\text{)ν}\text{U}\text{?}{}_{\mathrm{k}}\text{(}\text{S}{}_1\text{/γ, …,}\text{S}{}_{\mathrm{k?1}}\text{/γ, γ)}$ in the limit γ → 0 are calculated, under fairly weak conditions on q and K, by a method involving functional differentiation. These limiting functions are used to describe the molecular structure of the various states of the system both in the range of the potential q(r) and in the rage of the potential γ^νK(γr). The direct correlation function c? (r, γ) is also considered and it is shown that for $\text{S}\text{≠ 0}$, $\text{lim}{}_{\mathrm{\gamma \to 0}}\text{γ}{}^{\mathrm{?\nu }}\text{c}\text{?}\text{(}\text{S}\text{γ}\text{, γ) = ?βK (}\text{S}\text{)}$, for all one-phase states, where β is the reciprocal temperature. Special cases of our results confirm those of other authors, including the well-known results of Ornstein and Zernike.     

Indium antimonide energy band parameters derived from diamagnetic exciton spectra

The oscillatory absorption edge spectrum of thin InSb crystals in which the discrete structure of the Wannier-Mott exciton had been revealed at H=0 for the first time has been studied at 1.8 K in magnetic fields of up to 80 kOe. The experimental data were analyzed with a computer taking into account the exciton nature of the phenomenon and the effect on the Landu levels of the nonlocality of potential resulting from the electron-electron exchange interaction. A set of band parameters has been obtained providing a minimum discrepancy between the theoretical and experimental spectra by not more than 0.65 meV per point: $\text{E}{}_{\mathrm{p}}\text{= 23.42 eV, F = -1.18, N}{}_1\text{= -0.33,}\text{r}{}_1\text{= 36.41,}\text{r}{}_2\text{= 15.95,}\text{r}{}_3\text{= 16.99,}\text{r}\text{= 15.22, q = 0.31}$ for $\text{h}\text{= 17.9}$. The reasons for and the extent of discrepancy between the InSb parameters derived by other authors from interband magnetoabsorption experiments are discussed.     

Determination of the deformation potential of shallow donors in silicon

The stress dependence of the energy of the ground state of group V impurities (P, Sb) in silicon was investigated by measurement of the Hall effect in a wide range of pressures. A conclusion was reached that the deformation potential of the lowest impurity state of shallow donors (Ξ¹_u) in silicon differs from the deformation potential of the conduction band (Ξ_u), the value of this difference being dependent on the type of the impurity. According to our data, the most probable values for ($\text{Ξ}{}_{\mathrm{u}}\text{ Ξ}{}^1{}_{\mathrm{u}}$) are 0.12 eV for phosphorous-doped silicon and 0.06 eV for antimony-doped silicon.     

Two-body density matrix of bose fluids

A detailed cluster analysis of the two-body density matrix $\text{n(}\text{r}{}_1\text{,}\text{r}{}_2\text{,}\text{r}\text{′}{}_1\text{,}\text{r}\text{′}{}_2\text{)}$ and the associated pairing matrix elements is performed for a strongly interacting system of bosons described by a Jastrow wave function. The study employs a multiplicative cluster procedure as suggested by Feenberg. It is demonstrated that the density matrix has a structure $\text{f(r)f(r′)n(r)n(r′)}\text{exp}\text{[?R(}\text{r}{}_1\text{,}\text{r}{}_2\text{,}\text{r}\text{′}{}_1\text{,}\text{r}\text{′}{}_2\text{)]}$ where f(r) is the conventional Jastrow factor and n(r) the one-body density matrix corresponding to the Jastrow trial state $\text{(r  |}\text{r}{}_1\text{?}\text{r}{}_2\text{|, r′  |}\text{r}\text{′}{}_1\text{?}\text{r}\text{′}{}_2\text{|)}$. The function R may be expressed by a series of irreducible cluster contributions which are given by a compact expressions in terms of spatial distribution functions. The density matrix exhibits off-diagonal long range order of the type $\text{n(}\text{r}{}_1\text{,}\text{r}{}_2\text{,}\text{r}\text{′}{}_1\text{,}\text{r}\text{′}{}_2\text{) → f(r)n(r)f(r′)n(r′)}\text{as}\text{|}\text{r}{}_1\text{?}\text{r}\text{′}{}_1\text{| → ∞,r,r′ < ∞}$. The associated pairing matrix elements and pairing energy of a Bose fluid are derived upon proper specialization of our results for the two-body density matrix.     

Diode laser spectroscopy of the CCl radical

Vibration-rotation transitions of the fundamental band have been observed for both C³⁵Cl and C³⁷Cl in the ${}^2\text{Π}{}_{\mathrm{<mtext>1</mtext><mtext>2</mtext>}}$ and ${}^2\text{Π}{}_{\mathrm{<mtext>3</mtext><mtext>2</mtext>}}$ states by using an infrared diode laser spectrometer with Zeeman modulation. A few lines of the “hot” band (v = 2 ← 1) have also been recorded for C³⁵Cl. From an analysis of the observed spectra improved values were obtained for the vibrational harmonic frequency and anharmonicity constant, rotational constants, and Λ-doubling parameters. It was found necessary to take into account centrifugal distortion effects on the spin-orbit coupling constant A in the analysis, which gave $\text{(}\text{dA}\text{dr}\text{)}{}_{\mathrm{<mtext>e</mtext>}}\text{r}{}_{\mathrm{<mtext>e</mtext>}}$ to be ?176 ± 38 or ?125 ± 38 cm^?1, depending upon whether ²Σ^? or ²Σ⁺ states contribute more to the Λ-type doubling. The equilibrium internuclear distance r_e was calculated from the derived rotational constants to be 1.64506 ± 0.00016 Å.     

Ab initio studies on all-trans-polyene

Ab initio crystal orbital calculations have been performed on all-trans-polyene using a minimal STO-3G basis set. Several ground state properties such as the equilibrium structure, the most important force constants, the band structure and the density of states are reported. An alternant structure with $\text{r}{}_{\mathrm{<mtext>C=C</mtext>}}\text{= 1.327}\text{A}\text{?}$, $\text{r}{}_{\mathrm{<mtext>C-C</mtext>}}\text{= 1.477}\text{A}\text{?}$, $\text{?}\text{CCC}\text{= 124.2°}$ and $\text{r}{}_{\mathrm{<mtext>C-H</mtext>}}\text{= 1.085}\text{A}\text{?}$ was found to have the lowest energy.     

A LEED determination of the structures of Ru(001) and of CORu(001)−√3 × √3 R30°

The structures of Ru(001) and of the √3 × √3 R30° overlayer of CO on Ru(001) have been determined by LEED I–V measurements and comparison to calculations. Special attention was paid to accurate angular alignment, selection of a well-ordered portion of the surface, and avoidance of beam-induced changes of the CO layer. Five orders of reflexes over a range of 300 eV each were used for the clean surface and 7 orders over 200 eV each for the CO superstructure. For the clean surface, a slight contraction of the first layer spacing (by 2%) was found which gave r-factors of 0.04 (Zanazzi-Jona) and 0.16 (Pendry) for 5 non-degenerate beams. For the CO structure the most probable geometry is the on-top site with spacings $\text{d(}\text{RuC}\text{) = 2.0 ± 0.1}\text{A}\text{?}\text{and}\text{d(}\text{CO}\text{) = 1.10 ± 0.1}\text{A}\text{?}\text{(r}{}_{\mathrm{<mtext>ZJ</mtext>}}\text{= 0.21; r}{}_{\mathrm{<mtext>P</mtext>}}\text{= 0.51)}$. The two threefold hollow and the bridge sites can be clearly excluded.     

Infrared spectrum of acetonitrile: Analysis of Coriolis resonance

The Coriolis resonance between ν₄ and ν₇ in CH₃CN and between ν₁ and ν₅, ν₃ and ν₆, and ν₄ and ν₇ in CD₃CN has been analyzed, applying the technique developed by DiLauro and Mills, to obtain the signs of [$\text{ζ}{}_{\mathrm{r,sa}}{}^{\mathrm{y}}\text{(}\text{?p}\text{?Q}{}_{\mathrm{r}}\text{)(}\text{?p}\text{?Q}{}_{\mathrm{sa}}\text{)}$] and the ratio of $\text{?μ}\text{?Q}{}_{\mathrm{r}}$ to $\text{?μ}\text{?Q}{}_{\mathrm{s}}$ for the interacting pairs in CD₃CN. For (ν₄, ν₇) in both CH₃CN and CD₃CN, the sign of [$\text{ζ}{}_{\mathrm{r,sa}}{}^{\mathrm{y}}\text{(}\text{?p}\text{?Q}{}_{\mathrm{r}}\text{)(}\text{?p}\text{?Q}{}_{\mathrm{sa}}\text{)}$] is found to be negative as it is also for (ν₁, ν₅) in CD₃CN. For (ν₃, ν₆) the sign of this interaction term is found to be positive. For a given definition of normal coordinates the signs of these interaction terms give the relative signs of $\text{?p}\text{?Q}{}_{\mathrm{r}}$ and $\text{?p}\text{?Q}{}_{\mathrm{sa}}$; our study also gives approximate values for the corresponding ratio [$\text{(}\text{?p}\text{?Q}{}_{\mathrm{r}}\text{)}\text{(}\text{?p}\text{?Q}{}_{\mathrm{sa}}\text{)}$]     

Photoemission of Xe and Kr adsorbed on the W(110) plane

The UPS and XPS spectra of Xe adsorbed on clean, O, CO, and Xe covered W(110) surfaces and the UPS spectrum for Kr on clean and O covered W(110) surfaces have been investigated. On clean W, Xe and Kr show a splitting of the $\text{5p}{}_{\mathrm{<mtext>3</mtext><mtext>2</mtext>}}$ and $\text{4p}{}_{\mathrm{<mtext>3</mtext><mtext>2</mtext>}}$ hole states respectively. For Xe the coverage dependence of this splitting was investigated in detail; neither the positions nor the intensity ratio of the substates are coverage dependent for θ ? 0.04, suggesting that splitting is due to differences in the image interaction of the $\text{m}{}_{\mathrm{j}}\text{= ±}\text{3}\text{2}$ and $\text{m}{}_{\mathrm{j}}\text{= ±}\text{1}\text{2}$ components. For Xe equal shifts, relative to vacuum, of ~1.0 eV were observed for 5p, 4d, and 3d levels, suggesting that initial state effects are small. Image interaction for Xe and Kr on clean W could best be fitted by assuming an increase, rather than a decrease in the effective hole-image separation from the nominal value, suggesting that the image plane is moved back into the metal by a screening length. For Xe adsorbed on $\text{Xe}\text{W}$(110), or on virgin-$\text{CO}\text{W}\text{(110)}$ polarization of the intermediate layers was found to contribute significantly to relaxation. Coadsorbed oxygen broadened Xe 5p and Kr 4p peaks. There was an almost linear relation between O 2p UPS intensity at the energies of the various peaks and the amounts of broadening, suggesting that the latter results from resonance neutralization by electrons from the O 2p states.     

Laser-induced fluorescence spectroscopy of the E band of the Ã-X̃ transition of SO2 cooled rotationally in a supersonic jet

The rotationally resolved, laser-induced fluorescence spectrum of the E band of the $\text{A}\text{?}{}^1\text{A}{}_2\text{-}\text{X}\text{?}{}^1\text{A}{}_1$ transition of SO₂ seeded in a supersonic jet was observed, and each rotational line was assigned on the basis of the ground state combination differences and the relative intensity data as a function of the rotational temperature. It was demonstrated that the line congestion was reduced significantly in the spectrum of the jet, and some of the lines, e.g., ^rR₀(0), were assigned unambiguously. This makes it possible to determine the vibronic band origin with an error of less than 0.2 cm^?1.     

On the order of levels in charmonium

We prove a theorem concerning the energies of the 2S and 3D states in a potential $\text{V(r) =}\text{?g}{}^2\text{r}\text{+ V}{}_{\mathrm{<mtext>c</mtext>}}\text{(r)}$, where V_c is a non-singular confining potential. If $\text{(}\text{d}\text{d}\text{r)}{}^3\text{(r}{}^2\text{V}{}_{\mathrm{<mtext>c</mtext>}}\text{)}$ is positive, then the 3D state lies above the 2S state, provided

$\text{d}\text{d}\text{r}\text{1}\text{r}\text{d}\text{d}\text{r}\text{2V}{}_{\mathrm{<mtext>c</mtext>}}\text{+r}\text{d}\text{V}{}_{\mathrm{<mtext>c</mtext>}}\text{d}\text{r}\text{< 0, ?}{}_{\mathrm{r}}\text{>0.}$

For V_c = r^α, this corresponds to 0 < α < 2.     

A variation-iteration method for the three-body problem

We construct a practicable formulation of the variation-iteration method for the calculation of properties of a tri-nucleon system interacting via a modern potential. Our general approach is to take some integral form of the Schrödinger equation and reduce it to a set of simpler integral equations which may be solved by iteration. We concentrate particularly on the simplicity of the kernels which appear in this last set. The best method appears to be a two-step one, essentially equivalent to the Green function. In constructing the formalism, we are led to define a set of functions $\text{?}{}_{\mathrm{MN}}{}^{\mathrm{L}}\text{(k}{}_{12}\text{, k}{}_{23}\text{, k}{}_{31}\text{, r}{}_{12}\text{, r}{}_{23}\text{, r}{}_{31}\text{)}$ which play the role of iteration kernels. We give various properties of $\text{?}{}_{\mathrm{MN}}{}^{\mathrm{L}}$ and indicate very briefly how $\text{?}{}^{\mathrm{L}}{}_{\mathrm{MN}}$ may be reduced to a function of only two variables.