ALBERT

Notes: IR spectra of jet cooled ArHF are obtained via direct absorption of a high resolution tunable difference frequency laser in a 2.54 cm path length, slit supersonic pulsed expansion at 〈10 K. Detection limits of 2×109 molecules/cm3/quantum state permit observation of the high frequency ν1 fundamental stretch (1000) ← (0000), the ν1+ν2 van der Waals bend plus stretch combination band (1110) ← (0000), as well as transitions to the (1002) triply vibrationally excited state that are weakly allowed via Coriolis interactions with the Π+ component of the (1110) manifold. The ground state (0000) molecular constants are in excellent agreement with previous microwave data. From the changes in rotational and centrifugal distortion constants, the vibrationally averaged van der Waals well depth is estimated to increase (+15%) with ν1 excitation, but decrease dramatically (−42%) upon subsequent excitation of the l=1 ν2 bend. L-doubling in the ν1+ν2 (1110) perpendicular bending state is large and negative [−69.8(18) MHz] and indicates the presence of a near resonant Coriolis coupledvibration of Σ+ symmetry at lower energy. A second, localized Coriolis perturbation is observed in the (1110) state and assigned to the near resonant (1002) Σ+ fundamental plus van der Waals stretch overtone at higher energy. Analysis of this Coriolis interaction indicates that coupling can be significant even for a three quantum change in vibration. However, a perturbative, small amplitude oscillator model predicts Coriolis matrix elements only 18% of the observed values, suggesting that large amplitude, bend–stretch interactions can strongly enhance Coriolis coupling. The decrease in the B rotational constant and the vibrationally averaged well depth upon ν2 excitation confirms the strong coupling between van der Waals stretch and bend coordinates. The slit expansion geometry quenches perpendicular velocity distributions and therefore offers intrinsically sub-Doppler resolution in an unskimmed molecular beam. Residual linewidths in the ArHF spectra are all below the apparatus resolution limit of ±25 MHz, which translates into a lower limit for the predissociation lifetime of 3 ns, i.e., in excess of 2×106 ν1 vibrational periods.

Notes: Measurements of angular momentum alignment in molecular beam scattering are presented for scattering of Na2 by He, Ne, and Ar. A mechanism for the production of this alignment is discussed, based on mJ conservation for quantization along the momentum transfer vector. The experimental results differ in several aspects from the predictions of this simple mechanism. It is found that the differential cross sections are strongly dependent on mJ. Furthermore, it is found that the principal axis of the orientational distribution does not coincide with the momentum transfer vector. The discrepancy is attributed to a nonstrict mJ conservation in the collision. These experiments are therefore expected to provide a sensitive probe to the nonsudden aspects of molecular scattering.

Notes: The nascent rotational state distribution of CO(v‘=0,J‘) following excimer laser photolysis of ketene at 351 nm has been determined under collisionless conditions in a flow cell. At this low excitation energy dissociation can only take place on the triplet potential surface leading to CH2(X˜ 3B1) and CO(X˜ 1Σ+). The available energy permits only the vibrational ground state of CO to be populated. The observed rotational distribution of CO(v‘=0,J‘) deviates drastically from a phase space theory statistical distribution as well as from a thermal one. A Boltzmann plot of this distribution exhibits a population inversion for J‘〈13. The nonstatistical behavior is attributed to a barrier along the dissociation path. The fragments are repelled too rapidly for energy to be randomized between them. Thus the photofragmentation dynamics of triplet ketene contrasts markedly with dissociation on the singlet surface which has no barrier and gives a statistical CO rotational state distribution. An impulsive model calculation for the ab initio transition state geometry is in surprisingly good agreement with the experimental energy partitioning among the fragment degrees of freedom. This suggests that the CCO bond angle is strongly bent at the top of the barrier and that the barrier height is a substantial fraction of available energy.

Notes: A pulse sequence designed for the excitation of multiple quantum transitions in magic angle spinning solid state NMR spectroscopy is presented. It is shown that under the action of the standard time-reversal pulse sequence, the change in the sign of the dipole coupling (which is used to generate the multiple quantum coherences) upon rotation causes the multiple quantum intensity to vanish after each rotor period. This effect is demonstrated both in calculations and in experimental 1H spectra of adamantane. A modification of the time-reversal pulse sequence, which involves switching the phase of the rf pulses every half-rotor period causes the spin part of the Hamiltonian to switch sign in synchrony with the modulation of the spacial part. This allows the creation of multiple quantum coherences in solids with magic angle spinning. The effectiveness of this pulse sequence is demonstrated through calculations and experiments.

Notes: By slowly cooling under pressure diluted solutions of HCl molecules in argon or krypton we have obtained different aggregation states of these molecules: dimers, trimers, and even aggregates and microcrystals. Thick solid samples (up to 14 mm) have been cooled down to 4.2 K and compressed within the range 0–5 kbar using a piston–cylinder device. The new spectroscopic data on dimers and trimers, in connection with some results previously obtained at low pressure in various matrices, lead us to confirm that the band frequencies of dimers and trimers are mainly sensitive to the intermolecular separation and do not appreciably depend on the nature of the rare gas matrix. The HCl aggregates are clearly distinguished from the microcrystals because of the quite distinct behavior observed for their corresponding translational modes. Finally, emphasis is made on the first observation of an unexpected and strong absorption band, centered at 200 cm−1, which disappears irreversibly under the effect of pressure. Its possible origin is discussed and tentatively ascribed to structure defects.

Notes: Experimental results on KBr@B:ClO−4 single crystals, before and after x irradiation at 77 K, are reported. Infrared and polarized Raman spectra show clearly that the perchlorate molecules are isolated in the KBr matrix and possess Td symmetry. EPR spectra, at 9.5 and 36 GHz, of irradiated crystals at 4.2 K indicate the formation of ClO2−4 in trigonal symmetry. The g and 35Cl hyperfine tensors are found to be strictly axial and collinear. Temperature variation of the EPR spectra presented no motional averaging while uniaxial stress applied to the crystals yielded negative results. The structure of the ClO2−4 molecule is discussed in terms of a trigonal Jahn–Teller effect and a probable dissociation into (ClO−3–O−) complex.

Notes: A pulsed electric discharge in a pulsed supersonic jet of N2 is found to produce high densities of metastable E3Σ+gN2. Subsequent laser irradiation of these metastables results in single-photon excitation to autoionizing states. Singly charged cations are detected by time-of-flight mass spectrometry or alternatively, ejected electrons are detected by TOF electron energy analysis. From both the laser excitation spectra and photoelectron spectra we have identified a number of autoionizing Rydberg states, including a triplet Rydberg series (np=8−26) 3Πu←E 3Σ+g converging to the ground state ion, the first member of an s-type triplet Rydberg series 3Σ+u←E converging to the B state of the ion and the lowest member of a d-type triplet Rydberg series 3Πu and 3Σ+u←E converging to the A state of the ion. Some of these transitions involve core excitations of a 3s Rydberg state configuration, resulting in a doubly excited electronic configuration [He(2sσg)2(2sσu)(2pπu)4 (2pσg)2(3sσg)]. Other transitions confirm that the E state is a mixture of a 3s Rydberg state with an X ionic core and a 3p Rydberg state with an A ionic core.

Notes: Potential energy surfaces and spectroscopic properties were calculated for H2S and H3S+ from highly correlated SCEP-CEPA wave functions. The equilibrium geometry of H3S+ is predicted to be re =1.350 A(ring) and θe =32.2°. The vibrational frequencies of H323S+ (in cm−1) were calculated to be 2529 (ν1), 1050 (ν2), 2527 (ν3), and 1208 (ν4) which are all in close agreement with experimental values obtained for solid H3S+SbF−6. The computed proton affinity for H2S of PA298=716.7 kJ mol−1 is in very good agreement with experiment.

Notes: The complex formed from hydrogen fluoride and the hydrogen molecule has been investigated theoretically in two ways. Large basis, well-correlated ab initio electronic structure calculations have been used to map out regions of the potential energy surface and an electrical interaction model has been used to find the classical intermolecular interaction effects. From the ab initio potential surface, the fundamental vibrational transition frequencies of hydrogen and hydrogen fluoride are predicted to be red shifted by 20 and 15 cm−1, respectively. The Liu and Dykstra theory of vibrational frequency shifts that uses the intermolecular electrical interaction yields shifts of 31 and 19 cm−1, respectively. The equilibrium structure of the molecule is T shaped, a feature that is determined by electrical interaction, and the well depth is around 300 cm−1. A significant fraction of the well-depth results from electron correlation effects.

Notes: Energy gradient equations are presented for the coupled-cluster model with all possible excitations. By taking advantage of the equations for the coupled-cluster amplitudes, the gradient formulas may be expressed without explicit reference to the first-order changes in the amplitudes, in contrast to all earlier work. The coupled-cluster doubles (CCD) and coupled-cluster singles, doubles, and triples (CCSDT) models are treated as special cases of the general theory. Finally, by limiting the model to finite orders in perturbation theory, the gradient equations for the full fourth-order many-body perturbation energy are derived. Like the fourth-order energy itself, the gradient procedure is shown to be an n7 process in the number of basis functions. The computational implementation of this fourth-order energy gradient is discussed in detail.

Notes: Theoretical potentials for the X 1∑+ states of NaH, KH, and RbH are reported using near Hartree–Fock quality Slater basis sets and incorporating electron correlation through the coupled-pair formalism (CPF). Excellent agreement is obtained with the available RKR potentials even though the CPF formalism is based on the SCF reference that dissociates incorrectly. Electric dipole moment functions are reported for NaH–RbH, as well as vibrationally averaged dipole moments, Einstein coefficients, and radiative lifetimes for the first ten vibrational levels. For KH, an extensive study is made of the computational requirements of obtaining an accurate permanent dipole moment. For CsH, re, ωe, De, and μe are determined.

Notes: Optimized geometries and relative energies for three states of the C4 molecule have been obtained from single-reference configuration interaction (SRCI) calculations. The 1Σ+g state, which is formed without activation from the dimerization of ground state C2 molecules, is calculated to lie approximately 25 kcal above the 3Σ−g state. At the SRCI level, a rhombic form is calculated to lie 1.2 kcal below the triplet form; consideration of the Davidson correction reduces this difference to 0.4 kcal, inclusion of a second set of diffuse d functions increased the difference only by about 0.2 kcal. Consideration of these effects, the difference in zero-point energy and previous results for methylene leads to a final estimated separation of 4.9 kcal, favoring the rhombus. Electron density distribution analysis for the rhomb is consistent with the existence of a bond between inverted sp2 carbon atoms. To aid the detection of this unusual molecule, preliminary estimates of the lowest optical transitions were obtained from SRCI calculations and vibrational frequencies were obtained from SCF calculations. Comparison of the calculated results with experimentally obtained spectra suggest the possibility that both the linear triplet and the rhombus may have already been observed.

Notes: A pump–probe two-color experiment has been performed to elucidate the formation mechanism of O2 X 3Σ−g in the multiphoton absorption of NO2 over the region 460–540 nm. A probe dye laser was employed to excite O2 X 3Σ−g into the B 3Σ−u state and the UV emission intensity of Schumann–Runge bands was measured under the various experimental conditions. The maximum vibrational level of O2 X 3Σ−g formed is v‘max =24 which corresponds to Evib=30 968 cm−1. The rotational distribution of O2 X 3Σ−g (v‘=24) was almost of Boltzmann with Trot=1300 K at low pressures. The isotopic 1:1 mixture of N16O2 and N18O2 has been photolyzed to test whether the O2 molecules are formed by unimolecular dissociation or through chemical reactions. From the product branching ratio of 16O2:16O18O:18O2 and the maximum vibrational levels observed, the vibrationally excited O2 molecules are concluded to be mainly generated by the chemical reaction of O(1D)+NO2→O2+NO, ΔH=32 000 cm−1. The O(1D) atoms are formed by a sequential three-photon absorption of NO2, where the initial two-photon absorption occurs through the 1 2B2←X˜ 2A1 transition in a cyclic manner and a certain collision-induced process takes place in a dense system of the predissociative states locating in 11 900–19 400 cm−1 above the dissociation threshold of NO(2Π)+O(3P).

Notes: Selective photolysis at room temperature of IO−4 embedded in KClO4 crystals and subsequent x irradiation at low temperature lead to formation of IO2, which may be detected by ESR. Detailed comparison of the spin Hamiltonians of ClO2, BrO2, and IO2, all trapped in KClO4, indicates that the distributions of electronic charge and spin in these molecules are very similar.

Notes: Threshold photoelectron spectra of 12CO2 and 13CO2 have been obtained from the onset at 900 A(ring) (13.75 eV) to 620 A(ring) (20 eV) at a 9 meV resolution using dispersed synchrotron radiation with a 0.3 A(ring) bandpass. A number of forbidden and/or previously unobserved transitions have been assigned for the X˜, A˜, B˜, and C˜ states of CO+2. Photoelectron spectra by electron time of flight at selected excitation wavelengths in the Franck–Condon gap region below the A˜ state, reveal the autoionization of neutral valence states as well as Rydberg series converging to the A˜ 2Πu and B˜ 2Σ+u states. The photoelectron energy distribution obtained upon excitation of the Rydberg states is consistent with that given by the Bardsley–Smith model for autoionization, while the superexcited valence states decay via a resonance autoionization process proposed previously by Guyon, Baer, and Nenner. These valence states are also responsible for the strong enhancement of vibrationally forbidden transitions, such as (000)→(010). Finally, several threshold electron peaks in the region between the A˜ and C˜ states, whose origin can be attributed to autoionization, are shown to be highly sensitive to isotopic substitution.

Notes: The photoion yield curve of SeH, prepared by the reaction H+H2Se is presented. The adiabatic I.P. is 9.845±0.003 eV, and autoionization structure is observed, from which higher I.P.'s are inferred. The photoion yield curves of H2Se+, SeH+, and Se+ from H2Se are also measured. The fragmentation thresholds, together with I.P. (SeH), enable one to infer the bond energies D0(HSe−H)=78.99±0.18 kcal/mol and D0(SeH)=74.27±0.23 kcal/mol. The adiabatic I.P. for H2Se (X˜ 2B) is 9.886±0.003 eV.

Notes: The cross section for the formation of H(n=2) or D(n=2) by photodissociation for the three isotopes H2, HD, and D2 has been measured using synchrotron radiation between 500 and 200 A(ring). The comparison with computed cross section curves shows the importance of the Q11Πu Q21Πu(1) neutral states of the 2pπ2Πu ionization continuum. Contribution from Rydberg states converging to the 2pπ2Πu state of H+2 is suggested. No isotopic effect can be confirmed.

Notes: The electron energy loss spectra of gaseous thiophene and thiolane in the regions of S 2p, S 2s, and C 1s are presented along with the x-ray photoelectron yield (NEXAFS) spectra of both gases in the region of S 1s excitation. The thiophene spectra are compared to the corresponding NEXAFS spectra of solid (multilayer) and monolayer thiophene on Pt (111). MS-Xα calculations of the C 1s, S 2p, and S 1s excitation spectra of free thiophene are also reported. Intercomparison of the gas, surface, and calculated spectra allows a complete interpretation of the spectral features and facilitates determination of the molecular orientation of thiophene with respect to the surface in both the compressed [thiophene on Pt (111) at 150 K] and relaxed [thiophene on Pt (111) at 180 K] monolayer phases.

Notes: The vacuum ultraviolet absorption cross sections of SiH4, GeH4, Si2H6, and Si3H8 are reported for the wavelength region 107–220 nm using synchrotron radiation as a light source. Absorption maxima of these compounds were found at the exciting wavelengths of 115–119 nm. Broad peaks observed were mostly assigned as primarily Rydberg transitions of the σSiH and σSiSi bonding electrons to the 4s, 4p, and 4d orbitals. The absorption features of germane resemble those of monosilane. In the photoexcitation of monosilane, the emission of the SiH(A˜ 2Δ→X˜ 2Π) transition was observed and its onset was found to be 132±2nm. The absorption spectrum of disilane showed five peaks. They were mostly assigned as 2a1g→4s, 2a1g→np(n=4−6) transitions and the strongest band was overlapped by 1eg→4d and 1eu→4p Rydberg transitions. In trisilane molecules three very weak and broad peaks were recognized and assigned as 3b2→4s, 4p and 4a1→4s, 4d Rydberg transitions. The strongest band was tentatively assigned as the superposition of 1a2→4d and 3a1→4p Rydberg excitations.

Notes: In zero field NMR, the spectra are characterized purely by the quadrupolar or dipolar interactions, which are sensitive to molecular motion. An analytical theory is presented which describes the effect of two-site flips of a deuteron on its zero field NMR spectrum and demonstrates the potential of this novel technique in the investigation of molecular dynamics.

Notes: The cluster formation of hydrogen cyanide (HCN) in carbon tetrachloride (CCl4) matrix was studied by IR and far IR spectroscopy. This included three different phases of the matrix material: (a) The liquid (conventional solution studies), where only the formation of linear HCN dimers occurs. (b) The plastic crystalline solid (matrix isolation studies in the high temperature modification with randomly oriented rotating CCl4 molecules) where linear dimers and trimers of HCN appear to be stable. (c) The crystalline solid (matrix isolation studies in the low temperature modification with orientationally ordered CCl4 molecules) where long linear HCN chains of unspecified degree of aggregation are observed. For dilute solutions the phase transition from (a) to (b) induces a thermodynamically unstable state which relaxes towards equilibrium on a time scale of minutes depending on the temperature applied. This relaxation process is accessible to time-resolved Fourier transform IR and far IR spectroscopy.

Notes: A very detailed analysis of the fine structure of the 4E(G) state of Mn2+ in cubic and nearly cubic clusters is presented to account for the unusual fine structure observed in ZnSe:Mn. First, experimental results are reported which show that the spin-orbit interaction in the 4E(G) state of Mn2+ in ZnSe is about one half of that previously observed in cubic or nearly cubic clusters as MnS4, MnBr4, and MnCl4 and that the state Γ8 is neatly shifted with respect to the Kramers doublets Γ6 and Γ7. Second, a model is elaborated which shows that second-order spin-orbit interactions reduced by a Jahn–Teller coupling to E vibrational modes accounts well for the observed overall splitting of the 4E(G) state in the studied clusters and that fourth-order spin-orbit interactions correctly describe the shifts of the Γ8 state. The validity of the perturbation schemes is discussed and the results are compared to those obtained from a complete diagonalization in the d5 configuration.

Notes: We demonstrate that in a previously studied model of the stretching modes of the water molecule rotational motion in the plane of the molecule tends to decouple the stretches. For rotational angular momentum near J=18 (h-dash-bar), the two local mode stretches are almost entirely decoupled. The source of this decoupling is the centrifugal distortion which stabilizes the asymmetric stretch and effectively cancels the G-matrix coupling. This cancellation is clarified using three different methods: Direct examination of the numerically computed matrix elements, exact analytic matrix elements of an approximate Hamiltonian, and solutions of a Mathieu equation formulation of a classical resonance Hamiltonian. The importance of this result is discussed in light of the fact that strong rotational excitation can occur in infrared multiple photon excitation. If such rotational decoupling occurs in real systems, then intramolecular energy transfer would be diminished thus holding open the possibility of mode specific infrared excitation. The calculations were carried out by numerically evaluating matrix elements between a basis of Morse oscillator eigenstates using an efficient Gaussian quadrature scheme based on associated Laguerre polynomials.

Notes: Using a point-dipole model, we discuss short- and long-range interactions in a 2D array representing a monolayer. Long-range forces may be treated exactly if the film is homogeneous at the wavelength scale and if the effective polarizability including short-range forces is known. Reflection and transmission amplitudes for polarized light are calculated for two monolayer models taking the subphase into account for arbitrary incidence angle and reflection strength. Experimental results are then compared to the theory. The good agreement shows that oblique incidence reflection and transmission spectra are valuable tools for investigating orientation and organization of chromophores in monolayers. Although in most of the studied cases—including J aggregates—the dye chromophore lies flat on the water surface, two instances are presented of molecular aggregates whose transition moments are vertical.

Notes: The intensity of the (0,0,5)–(0,0,0) transition at 15 552 cm−1 in HCN has been measured using a cw laser dye laser, and a White cell with an optical path length of about 0.5 km. From a photoacoustic spectrum of HCN, the relative band intensity of the (0,0,5)–(0,0,0) transition to the (0,1,5)–(0,1,0) hot band transition has been determined. The integrated band intensity of the (0,0,5)–(0,0,0) band is 2.91(7)×10−23 cm−1/(molecule cm−2) or 17.5(4) cm/mol, which implies a dipole transition moment of 6.70(8)×10−5 D. The (0,1,5)–(0,1,0) band is equal in intensity to the (0,0,5)–(0,0,0) band within experimental error. The self-broadening and pressure shifts of these bands have also been examined. In the (0,0,5)–(0,0,0) band, the self-broadening coefficients for both the P and R branch for J between 0 and 19 were measured. For the P branch, the self-broadening coefficient is 23.7 MHz/Torr for J=1, increases monotonically to a maximum of 48.9 MHz/Torr at J=9, and then decreases monotonically to 21.7 MHz/Torr at J=19. For the R branch, the coefficient is 24.1 MHz/Torr for J=0, rises to a maximum of 45.4 MHz/Torr at J=7 and then falls to 14.7 MHz/Torr at J=19. For both branches, the variation of the self-broadening coefficient with J follows the J thermal population factor, which suggests that the collisional broadening is dominated by rotational resonances. Even though the vibrational density of states in the upper state is much larger than the density of rotational states, the line broadening appears to be fully accounted for by rotational relaxation.

Notes: The rate of bond inversion at the nitrogen atom in aziridine (dimethylene-imine) was measured in the gas phase for a range of pressures and temperatures. 1H NMR spectra were recorded of samples at 5–1100 Torr, and from 298 to 388 K. Rate constants for inversion were derived by complete line shape analysis of the NMR spectra. Their temperature dependence at the high pressure limit (1100 Torr) gave Ea =15.8±0.4 kcal mol−1; ΔH

Notes: Exothermic gas phase electron transfer reactions: A−+B=A+B−, where A and B are polyatomic molecules with positive electron affinities, generally proceed at collision rates. However, reactions involving A=SF6 or perfluorocycloalkanes have rates which decrease with the exothermicity of the reaction, becoming very slow at low exothermicity. Earlier work attributed this behavior to the presence of an energy barrier, due to a large geometry change for A− to A. The reaction coordinate used also involves the bond energies A− ⋅ B and A ⋅ B−. These were measured in the present work. It was found that the bond energy in SF−6 ⋅ B is much larger than in SF6 ⋅ B− . This difference increases the energy barrier very significantly and is thus an additional cause for the slow electron transfer. The bond energies for several other complexes like: Cl− ⋅ B, Cl− ⋅ SF6, Cl− ⋅ perfluorocyloalkanes, and Cl− ⋅ C6F6 were measured. These provide insights into the nature of the bonding involved. The work was performed with a pulsed electron high pressure mass spectrometer.

Notes: Following multiphoton or nonlinear UV excitation (22 ps, 266 nm, ∼2.5 mJ) of simple molecules at low pressure, we have measured the temporal evolution of emissive product populations with subnanosecond time resolution. For several parents, our data support the importance, beneath the I. P., of fragmentation to yield neutral species competitive with further excitation to yield parent ions. The primary neutrals and ions are subject to further excitation. Channels of dissociation are recognized as prompt (unimolecular) and delayed (ion–molecule) components of fragment emission. Representative data for CH2CO are presented. A discussion of some recent REMPI results on this molecule is also given. The question has arisen whether rapid production of neutrals, energetically requiring nhν〉I.P. of the parent, occurs through superexcited species. We suggest that a conclusive demonstration of such pathways has yet to be achieved.

Notes: The small-curvature semiclassical adiabatic (SCSA) approximation, which is based on a reaction-path Hamiltonian, is used to calculate the tunneling splitting due to the degenerate rearrangement of hydrogen fluoride dimer. The calculation employs a semiempirical potential energy surface which approximates the HF molecules as rigid rotators, and for which accurate tunneling splittings have been previously calculated. The semiclassical method is shown to be accurate within 33%. The internal motion of the dimer along the reaction path and the contributions of the generalized normal mode vibrations to reaction-path curvature in the tunneling region are also discussed.

Notes: Picosecond infrared pump–probe experiments determined the vibrational population lifetimes (T1) of the hydroxyl fundamental stretching mode OH(v=1) in 12 alcohols (R3COH) and 8 silanols (R3SiOH) in dilute room temperature CCl4 solutions. T1 for the silanols is in the range 185〈T1〈292 ps, while T1 for the alcohols is much less (T1〈80 ps). The deuterium-exchanged analogs (COD and SiOD) exhibit population relaxation times similar to protonated hydroxyls. An analysis of the vibrational energy levels corresponding to modes involving the four bonds nearest the hydroxyl groups of these molecules is used to qualitatively explain the trends of the observed T1 lifetimes for these systems. Solution T1 lifetimes are also compared to those previously measured for OH(v=1) on the surface of silica and in other condensed-phase, room temperature systems.

Notes: An iterative procedure is proposed for determining increasingly accurate effective Hamiltonians for use in the adiabatically reduced coupled equations approach to intramolecular dynamics calculations [J. Chem. Phys. 84, 2254 (1986)]. The relationships between this iterative determination of the effective Hamiltonian, which is based on an adiabatic approximation, and some other partitioning methods for determining an effective Hamiltonian are discussed. The present iterative procedure provides accurate agreement with the exact dynamics for the two specific model systems studied.

Notes: An expression for the distribution of quantum states of the reaction products of unimolecular dissociations is obtained, based on statistical theory. A recently formulated RRKM-type treatment of unimolecular reactions with highly flexible transition states is used to obtain a distribution of quantum states of the products, by introducing an adiabatic approximation for motion from transition state to products. Any impulsive (nonadiabatic) exit channel effects are neglected thereby. Both the final yields of the quantum states of the products and the time evolution of these states are considered. The time evolution of the yield of the products can permit a direct test of non-RRKM effects and, additionally via the long-time component, of other aspects of RRKM theory. The long-time component of the yield of individual quantum states of the products then provides a test of the additional (here, adiabatic) approximation. Such tests are the more definitive the narrower the distribution of initial E's and J's of the dissociating molecule.

Notes: The question of intensity vs energy fluence dependence in the IR excitation of a polyatomic molecule in the quasicontinuum is addressed. It was found that the energy fluence scaling law breaks down for laser pulses shorter than ∼100 ps for C3F7I. This is consistent with similar results obtained previously for SF6 and C2F5Cl. The reason for the increased intensity dependence is physically due to an increase in the intrinsic small signal absorption cross section of the polyatomic molecules for short interaction times. This increase in the absorption coefficient is not intensity related and is a fundamental property of a dense ensemble of interacting states. The change in the absorption cross section for all molecules studied reveals a nonequilibration of energy relaxation among all the states even for 100 ps pulses. This nonequilibrium situation can be exploited to achieve ps nonstatistical photochemistry.

Notes: Static dissipative structures are obtained analytically for multiple time scale systems with folded slow manifolds. We find that the associated free boundary problem displays much of the richness of the full reaction–diffusion system. For a simple one-dimensional structure the pattern is shown to be linearly stable. Spoke patterns in a disk are calculated exactly. Since one type of static pattern in a disk was shown earlier to unfold into a rotating wave, these patterns are also of interest as the basis for the analysis of a class of multiple arm rotors. In a second special limiting case the patterns are shown to be in the form of narrow "double layers'' in two-dimensional systems. Double layer patterns in a disk are obtained.

Notes: Coupled-Hartree–Fock and equations-of-motion calculations at ab initio level for the boron–boron and boron–proton indirect nuclear spin–spin coupling constants in a representative series of polyboranes (B2H6, B2H4, B2H−7, B3H−8, B4H10, and B5H9) are presented. The boron–hydrogen–boron coupling constant is very sensitive to molecular structure and shows neither a unique sign nor a constancy of absolute value. It is found to be negative in B2H6. On the other hand, the coupling constant for a two-center two-electron boron–boron bond is predicted to be invariably positive in sign. Results show that the positive sign must also be attributed to 1J(BH), irrespective of the bridged or normal nature of the boron–hydrogen bond.

Notes: Nucleation was used to study the concerted unimolecular rearrangement reactions of low concentrations of different solutes. Those solutes that exhibited a polar transition state or a polarizable intermediate during thermal rearrangement caused a solvent, when supersaturated in the vapor phase in a thermal diffusion cloud chamber, to nucleate at supersaturations smaller than those needed for homogeneous nucleation. This thermally induced nucleation decayed over time. The solutes that did induce nucleation were: benzocyclobutenol, 2-methoxybenzocyclobutene, allyl vinyl ether, and an unknown decomposition product of allyl vinyl ether. However, 1,5-cyclodecadiene, whose transition state has no polar character, did not induce nucleation. Most of this research used benzocyclobutenol as the dopant. Evidence was found which confirms the suggestion made previously that the thermal ring opening of benzocyclobutenol to its E-enol leads to induced nucleation. Added 1,4-naphthoquinone successfully quenched nucleation by trapping the E-enol in solution. Activation energies of 30.9±0.6, 33.5±1.6, and 34.7±2.0 were obtained for the overall decomposition of benzocyclobutenol to o-tolualdehyde using bottom plates having copper, gold, and aluminum surfaces, respectively. Monitoring of the liquid phase concentration of benzocyclobutenol indicated a third order dependence of the nucleation rate on concentration.

Notes: Lithium sulfate monohydrate has been studied at 80 and 298 K by x-ray diffraction. The monoclinic crystal with space group P21 has lattice dimensions at 298 K of a=5.4553(1), b=4.8690(1), c=8.1761(2) A(ring), and β=107.337(2)°; lattice dimensions at 80 K were reported in our neutron study [J. Chem. Phys. 80, 423 (1984)]. Least-squares refinement based on 3486 (80 K) and 3390 (298 K) independent reflections, assuming a spherical atom model, results in final R( F 2) values of 0.024 (80 K) and 0.026 (298 K). Static deformation and charge density model refinement, based on Hirshfeld-type multipole functions, greatly reduces the residual electron density and gives R( F 2) values of 0.017 (80 K) and 0.016 (298 K). Refinement of the resulting multipole parameters within the Gaussian radial dependence model allows a qualitative estimation of the differences in electron densities between 80 and 298 K. A change of about 0.2 e A(ring)−3 in the deformation density of the O–H bonds in the water molecule over this temperature range is related to the contraction in hydrogen bonding at 80 K. Changes in the sulfate oxygen atom lone-pair deformation densities, caused by contractions in the SO2−4–Li+ contacts between 298 and 80 K, are also found between the two temperatures. X-ray–neutron diffraction deformation density calculations indicate a substantial electron deficiency in the sulfur valence shell. Kappa refinement leads to a point charge model with positive atomic charge close to 1.8 on sulfur, about −1.0 on each sulfate oxygen atom, and −0.8 on the water oxygen atom. Interatomic distances at 80 and 298 K, particularly those based on a Gaussian radial dependence for the deformation density function, are in excellent agreement with the corresponding neutron diffraction values.

Notes: Bloomfield, Geusic, Freeman, and Brown have used the Smalley synchronized laser and supersonic quenching carrier gas pulsed method to obtain spectra for Ge+n similar to pulsed spectra obtained previously for Si+n. Their data are presented, compared to data obtained with electron beam ionization, and discussed extensively in order to separate effects due to cluster stabilization energies, growth kinetics, and ionization fragmentation. Differences in chemical bonding and growth kinetics between Si+n and Ge+n microclusters are discussed.

Notes: A novel moment-gradient expansion scheme, expressing the microscale probability density P as an infinite sum of global-space gradients of its corresponding macroscale density P¯ multiplied by coefficients formed from its local and total moments, is employed to derive an asymptotic long-time macrotransport equation from its more detailed microtransport predecessor. Particular emphasis is paid to third- and higher-order gradient terms in the expansion. These are shown to result in non-Gaussian behavior of the macroscale probability density P¯ governing convective–diffusive transport processes.

Notes: The palladium LIII absorption and chlorine K absorption spectra from [Pd(NH3)4]Cl2 ⋅ H2O and Pd(OCOCH3)2 have been reported for the first time and discussed in connection with the Pd LIII absorption and Cl K absorption spectra of K2PdCl4, reported previously. The influence of the ligands Cl−, OCOCH−3, and NH3 has been observed in the pallaldium LIII absorption spectra of these compounds. The observed chemical shift of a strong "white line'' at the Pd LIII absorption threshold can be explained in terms of the spectrochemical series of the ligands. An intense white line disappears at the Cl K absorption threshold of [Pd(NH3)4]Cl2 ⋅ H2O in contrast to the Cl K absorption spectrum of K2PdCl4.

Notes: A study of the convergence properties associated with direct lattice summations of the exchange contributions to the quantities needed in a restricted Hartree–Fock calculation for a polymer, leads to a detailed analysis of the analytic and asymptotic properties of the Fock–Dirac density matrix in the LCAO representation. The results, obtained simply by means of Fourier analysis, provide an important characterization of the restricted Hartree–Fock method as applied to chain-like systems. As a by-product we obtain a general proof in direct space that a partially filled band, which influences drastically the analyticity properties of the density matrix in reciprocal space, leads to a logarithmically diverging derivative of the orbital energy at the Fermi level. This in turn gives a vanishing density of states at that level. This result, well known for the electron gas model, is thus an inherent property of the restricted Hartree–Fock approximation.

Notes: The problem of a macromolecule adsorbed on a single surface is treated by means of a generating function technique. The basic method has the virtues of simplicity and flexibility. The statistical weights that appear in the generating functions for tails, trains, and loops can be calculated with use of a variety of models. The essential physics of the adsorption process, and the occurrence of a critical point, are transparent in this treatment. Illustrative calculations are done for the simplest case, in which tails have unit weight, trains have binding energies proportional to their lengths, and loops are weighted by lattice walk statistics. Methods for treating more realistic models for chains, and for handling their interactions when there is multiple chain adsorption, are discussed.

Notes: Starting from the generalized entropy of extended irreversible thermodynamics, we study the influence of the electric current on the probability of finding a system in a state of given energy and electrical conductivity. As an application, we study the probability of open and closed pores in a membrane.

Notes: The previously observed phenomena of temporal and spatial self-organization during the catalytic oxidation of CO on a Pt(100) surface were computer simulated by use of the cellular automaton technique. The underlying model is footed on the knowledge about the individual reaction steps (adsorption, desorption, surface structural transformation, etc.), which in turn formed the basis of a previous theoretical treatment in terms of the formulation and solution of a set of coupled differential equations. The present result nicely show the formation and propagation of two-dimensional patterns, and they reproduce qualitatively well all the experimental observations. The development of macroscopic patterns even with an a priori perfectly homogeneous surface is a particularly interesting effect.

Notes: Transmission electron microscopy and optical absorption were used to examine small clusters of the layered semiconductors, PbI2 and BiI3. In both systems, a layer of metal is sandwiched between two hexagonally closed-packed layers of iodine. We describe a simple solution preparation which gives rise to clusters corresponding to single layer sandwiches, roughly 7 A(ring) thick, whose lateral dimensions vary from 12 to 90 A(ring) depending on the solvent and the nature of the growing cluster interface. The cluster size distributions are markedly different for PbI2 and BiI3 reflecting the different structure in the metal planes of these systems. PbI2 cluster sizes are determined by hexagonal symmetry, with cluster growth achieved by placement of lead atoms symmetrically about a smaller cluster. In BiI3, whose metal plane has a honeycombed structure like graphite, clusters grow to be much larger with their sizes determined by the closure of six-membered rings. The optical absorption spectra of PbI2 and BiI3 can be quantitatively understood in terms of the measured cluster size distributions.

Notes: Laser induced holographic grating relaxation studies of camphorquinone (CQ) diffusing in liner polystyrene (PS) (with both narrow and wide molecular weight distribution) and crosslinked PS (with bis-phenol-A-dimethacrylate as the crosslinking agent) are carried out. Diffusion coefficients of CQ in the polymer hosts are obtained from the decay rate of the intensity diffracted from the laser induced holographic grating. The diffusion coefficients (D) of CQ in polydispersed PS samples are compared with that in nearly monodispersed sample. Diffusion coefficients are found to be equal in both types of polymer samples over the whole temperature range studied, thereby indicating that at high molecular weight the molecular weight distribution has little effect on the diffusion coefficient of CQ. The diffusion coefficient of CQ in crosslinked PS is found to decrease with increasing concentration of the crosslinking agent. It decreases approximately 30 times as the concentration of the crosslinking agent increases to 20%. The relationship between the concentration of the crosslinking agent and polymer free volume is discussed. The diffusion coefficients of CQ in linear and crosslinked PS are also measured as a function of temperature. It is found that the temperature dependence of both D and D/T can be described by a Williams–Landel–Ferry (WLF) equation. However, the WLF constants C1 and C2 differ in both fits. The C2 constant obtained from the D/T vs T fit for linear PS sample agrees with that extracted from the viscoelastic measurement.

Notes: We report here the first direct measurement of intermolecular forces between a polymer layer and a solid surface. When solid surfaces bearing adsorbed polymer approach one another, interactions develop that may be categorized as polymer–polymer and polymer–solid (and a negligible contribution of solid–solid interactions at the large solid–solid separations imposed by adsorbed polymer). Direct measurements of the forces between polymer layers adsorbed on mica have been made recently with apparatus of the type developed by Israelachvili. Attractive forces are attributed to a combination of osmotic (polymer–polymer) and bridging (polymer–solid) interactions. Bridging occurs if macromolecules adsorbed one one surface can reach a second surface. In the work presented here, poly(α-methylstyrene) (PαMS) adsorbed on one mica sheet was brought into contact in cyclohexane with a bare mica sheet. Adhesion was measured as a function of separation and contact time and found to be more than 20 times greater than between two polymer–coated surfaces under otherwise identical conditions. This direct measurement of molecular forces between dissimilar surfaces enables estimation of the PαMS segmental sticking energy on mica in cyclohexane as (1)/(3) kT.

Notes: The reactions between aluminum cluster ions Al+n (n=4–25) with oxygen have been studied using a low energy ion beam apparatus at a center of mass collision energy of 1 eV. Some extraordinary reactions were observed. Oxygen very specifically cleaves the aluminum clusters, Al+n, resulting in Al+n−m cluster ions where m=4, 5 and for the larger clusters 10. Al+ is also an important product ion for clusters with n≤13. No oxygen containing product ions were observed. We suggest that addition of oxygen to the cluster raises the ionization potential and makes AlxO+ and AlxO+2 products energetically unfavorable, thus neutral oxide species are formed.

Notes: The absolute integral cross sections for oxygeg atom collisions with helium atoms have been obtained. From these cross sections, the interaction between He and ground state oxygen was calculated.(AIP)

Notes: Radiative lifetimes for CH3S(A˜ 2A1) in the υ3 =0, 1, and 2 levels are 760±60, 810±60, and 197±15 ns, respectively. These results are in disagreement with previous work on the υ'3 =0 and 1 levels although in agreement for the υ3 =2 level. Since the previous work was in a beam, it is possible that loss from the viewing region contributed to the shorter lifetimes reported previously for the lower levels. Quenching by (CH3S)2 is fast for all three levels.

Notes: The theory of magneto-chiral birefringence and dichroism is extended to systems composed of diamagnetic molecules acted on by an optical field. For molecules with the point group symmetries 3, 4, 6, ∞, 32, 422, and 622 expressions are derived for the magneto-chiral birefringence and dichroism in the presence of weak as well as very strong optical molecular reorientation caused by a linearly polarized, circularly polarized, or unpolarized laser beam. Measurements of magneto-chiral birefringence and dichroism at optical saturation provide direct information on the respective magnetic dipole and electric quadrupole polarizability tensor components.

Notes: The vibration–rotation spectrum of the amide anion (NH−2 ) has been measured in the 3 μm region by velocity modulation laser spectroscopy. The ν1 and ν3 bands have been simultaneously fit to a Watson S reduced Hamiltonian; the band origins are (in cm−1) 3121.9306(61) and 3190.291(14), respectively. The r0 structure has been determined: rNH=1.0367(154), θHNH=102.0(3.3). Under appropriate discharge conditions, NH−2 concentrations near 4×1011 molecules/cm3 can be produced in cells with metal-coated walls. Comparisons with condensed phase measurements and ab initio calculations are presented.

Notes: The photoabsorption and fluorescence cross sections of H2CO were measured in the 105–180 nm region using synchrotron radiation as a light source. Vacuum ultraviolet (VUV) emission from excited photofragments was dispersed and identified to be the CO(A 1Π→X 1Σ+) system. The vibrational population of CO(A) was determined from the fluorescence spectrum and was used to study the photodissociation mechanism. The threshold for the production of the VUV emission is at 140.3 nm. UV emission from HCO* was observed below a threshold of 147.5 nm. The upper limit of the dissociation energy, D0(H–HCO), determined from the HCO emission threshold is 3.61±0.03 eV. The quantum yield for the production of either CO* or HCO* is a smooth function of the excitation wavelength, from which two dissociative states with vertical energies at 8.69 and 10.7 eV are derived. The VUV and UV fluorescences have maximum quantum yields at 116 nm of about 1.6% and 0.23%, respectively. The photodissociation process of H2CO in VUV is discussed.

Notes: The OH2− defect in CaO has been studied by EPR. The 77 K static spectrum has been reinterpreted and new values are given for the spin-Hamiltonian parameters and principal axes orientation. The averaging features due to the motion of the defect at higher temperatures are studied by means of the stochastic Liouville formalism. Inclusion of pseudosecular Hamiltonian terms within the theory explains most of the experimental details. The hopping process follows the classical exponential temperature dependence, with an activation energy of 0.23±0.01 eV. A superhyperfine (shf) structure resolved at room temperature permits identification of a nearest-neighbor substitutional Mg2+ ion as the perturbation that stabilizes the defect. The 25Mg shf and quadrupolar parameters, including their absolute signs, are also given.

Notes: The absolute Raman cross sections of the 992 cm−1 (ring breathing) and 3060 cm−1 (C–H stretching) modes of benzene are presented at excitation wavelengths of 514.5, 488.0, 441.6, and 325.0 nm. The technique employs a straightforward determination of the total (unrestricted energy bandwidth) scattering cross section along with a simple relative measurement. The effect of the dispersion of the local field correction (via the frequency dependence of the index of refraction) is found to significantly affect the preresonance benzene cross sections and can distort the normally employed "A-term'' analysis. When the local field effect is incorporated into the data analysis, the results involving neat benzene and previous studies involving 1% benzene in acetonitrile are brought into reasonable agreement. The "n2'' solid angle correction is also verified experimentally using air, water, and benzene as test substances. The preresonance enhancement of the 992 cm−1 mode is found to be consistent with coupling to electronic excitations centered at ∼80 kK which presumably involve the π-electron system. In contrast, the 3060 cm−1 mode is coupled to electronic states centered in the deep ultraviolet, at ∼150 kK, which presumably involve the more energetic σ-electron excitations.

Notes: The photoabsorption cross sections of C2H6O isomers (ethyl alcohol and dimethyl ether) and C3H8O isomers (n-propyl alcohol, isopropyl alcohol, and ethyl methyl ether) have been measured in the wavelength region from 30 nm to the ionization potential of each molecule (about 120–130 nm) using an ionization chamber. The obtained photoabsorption cross sections show a maximum value at about 70–80 nm for each molecule. In the wavelength region shorter than about 50 nm, the cross sections are almost the same among isomers and equal to the sum of the cross sections for the constituent atoms. In the longer wavelength region, however, the cross sections show different peaks and shoulders depending on its molecular structure. Especially in the cases of dimethyl ether and ethyl methyl ether vibrational structures of the cross section are observed in the wavelength region from 70 to 100 nm.

Notes: Classical and quantum theories of ensemble averaged fluorescence polarization are presented for the asymmetric rotor in the regular, rigid rotor limit and in the limit of statistical vibration–rotation energy transfer in the isolated molecule. Planar asymmetric rotors are explored in detail, with special emphasis on p-difluorobenzene and pyrimidine. The classical polarizations are nearly identical to the quantum results, but are 1000 times faster to calculate. For nearly all molecular geometries, our results predict the transition from regular to statistical rotational motion will sharply reduce the fluorescence polarization, signaling the onset of intramolecular vibration–rotation energy transfer.

Notes: We have used the method of quasiclassical dynamics to investigate intra- and intermolecular energy transfer in H2+OH collisions. Energy transfer has been investigated as function of translational temperature, rotational energy, and vibrational energy. The energy transfer mechanism is complex with ten types of energy transfer possible, and evidence was found for all types. There is much more exchange between the translational degree of freedom and the H2 vibrational degree of freedom than there is between translation and OH vibration. Translational energy is transferred to the rotational degrees of freedom of each molecule, and this occurs more readily for OH than H2. Both molecules exhibited intramolecular energy transfer from vibration to rotation, and this was a major pathway for vibrational deactivation. Evidence was also found for the intermolecular transfer of energy from vibration to the rotational and vibrational degrees of freedom of the other molecule.

Notes: A two-step photodissociation method has been used to measure the photodissociation cross section of vibrationally excited OCS(ν1) at λ=249 nm. The method employs an NH3 laser to pump OCS molecules to the ν1=1 vibrational state, followed by photodissociation by a KrF excimer laser and mass spectrometric analysis of the chemically scavenged photodissociation products. The experiments determine the photodissociation cross section ratio σ d(ν1)/σ d295 K =4.5±0.9, where σ d295 K is the 249 nm photodissociation cross section of thermal (295 K) OCS. An analysis of our previously reported data for the ν2=2 vibrational state of OCS gives the photodissociation cross section ratio σ d(2ν2)/σ d295 K =8.4±1.1. The measured cross section ratios are used in an analysis of temperature dependent photoabsorption spectra to obtain the photoabsorption cross section ratios σ a(0)/σ a295 K =0.41±0.04 and σ a(ν2)/σ a295 K =2.5±0.6 for the vibrationless and ν2=1 states of OCS, respectively. Isotopically selective, two-step photodissociation through the OCS(ν1) vibrational state was used to separate 33S or 34S isotopes with enrichment factors of β=2.2.

Notes: The rates for gas-phase hydrolysis of SOF2 (thionylfluoride) and SOF4 (thionyl tetrafluoride) have been measured at a temperature of 298 K. The second order rate constant for SOF2 hydrolysis in SF6 buffer gas was found to have the value (1.2±0.3)×10−23 cm3/s which agrees with previous estimates of Sauers et al., but is three orders of magnitude lower than the value obtained by Rüegsegger et al. at 340 K. The rate constant for SOF4 hydrolysis has not previously been measured and its value in both SF6 and N2 buffer gases was found here to be (1.0±0.3)×10−21 cm3/s.

Notes: An analytical potential-energy surface based on the Varandas–Murrell potential for equilibrium NH3 has been formulated using the results of scaled CI/6-31G* calculations to adjust the potential parameters to give the correct energies and geometries for equilibrium NH3, the inversion, and the NH2+H and NH+H2 dissociation channels. Microcanonical unimolecular decay coefficients have been calculated for both channels over the energy range 5.25–7.0 eV from classical trajectories. The overall dissociation mechanism is found to consist of two parallel first-order decay processes. Although the reaction thresholds for both channels are nearly identical on our potential-energy surface, it is found that dissociation to NH2+H is the major decomposition pathway at all energies. The computed product translational energy distributions for NH2+H are peaked at energies near zero, as expected for dissociation processes which have no barrier to the back reaction. In contrast, the corresponding distributions for the NH+H2 product are found to be shifted toward higher energies due to the presence of a 15 kcal/mol back-reaction barrier. Examination of the mechanistic details of individual trajectories shows that dissociation to NH+H2 occurs via a concerted elimination. Deuterium isotope effects are reported for the dissociation of ND3, NH2D, and NHD2.

Notes: The interaction of 3D, 1D, 3F, and 1D states of Sc+ with H2 has been explored through the generation of ab initio potential energy surfaces for two different interaction geometries. We find that all components of the ground 3D and lowest 1D react in an endothermic process to form ScH+ (ΔH0=48.6 kcal/mol for the lowest product state from 3D Sc+) from a collinear interaction, and that four of the five components of each state are likely to proceed to this product from any interaction geometry. The remaining component of 1D Sc+ will form ScH+2 in an activated process (ΔH0=−4.4 kcal/mol; E° =12.8 kcal/mol). Inclusion of the effects of spin-orbit interactions reveals that the last 3D component may also proceed adiabatically to ScH+2 (ΔH0=+1.8 kcal/mol; E° =19.0 kcal/mol) within a very narrow energy window. At least eight H2 adduct states are also formed in the interaction with excited singlet and triplet Sc+ states. We explore in detail the means by which Sc+ activates the H2 bond, and relate our results to other metal–H2 systems.

Notes: Quasiclassical trajectory calculations have been performed to study intramolecular vibrational energy redistribution (IVR) from CH overtone states in benzene. The rate and extent of this redistribution is sensitive to details of the potential energy surface. A particularly important potential energy surface property is attenuation of the HCC bending frequency upon CH stretch excitation. This property gives rise to non-irreversible IVR when the CH stretch is highly excited. An analog to quantum beats is observed in phase averaged quasiclassical trajectories.

Notes: The butynal molecule (CH3–C≡C–CHO), possessing an intermediate S1–T1 level structure, has been investigated in a pulsed supersonic expansion of Xe or Ar. Employing a ∼5 ns dye laser pulse (Δν˜coherence) (approximately-equal-to)0.006 cm−1, Δν˜spectral =0.04–0.2 cm−1) the vibrational state dependence of the coherent and incoherent decay of the first excited singlet state has been explored in the excess vibrational energy range Eexcess =0–1400 cm−1 where the density of vibronic states is changing from ρvib ∼80 to ∼1200 states/cm−1. The analysis of the "biexponential'' decay behavior, which is characteristic in intermediate molecules, shows in case of butynal that the fast component is consistent with the intramolecular dephasing of S1–T1 mixed molecular eigenstates, and that the slow component represents the incoherent decay of these eigenstates given by their lifetimes. Based on this picture and in conjunction with molecular information obtained from a previous spectroscopic study on butynal [J.Chem. Phys. 84, 3014 (1986)], a simulation of the biexponential decay has been successfully performed.

Notes: Decay rate measurements of internal energy selected ions were performed with a novel technique employing highly selective two-photon laser ionization in a reflectron time-of-flight mass spectrometer. Resonantly enhanced two-photon ionization leads to vibrational state selected C6H+6 cations. These are further excited with a second laser pulse of variable frequency beyond the dissociation threshold for the four decay channels of lowest energy. The excitation results in a metastable decay of C6H+6 with typical decay rate constants of some 106 s−1. Decay rate constants for two C-loss channels and for the first time also for the two H-loss channels are measured in the energy range from 5.1 to 5.5 eV. It is shown that H-loss and C-loss dissociation occurs from a common electronic state and is competing. From the total decay rate constants and the measured branching ratios individual decay rate constants for all decay channels are obtained and found to be in good agreement with RRKM calculations. This confirms the statistical character of the C6H+6 ion dissociation.

Notes: The lifetime of the excited 2 1Ag state of cis,cis-octatetraene (the eight carbon oligomer of cis-polyacetylene) in a n-octane host crystal has been measured as a function of temperature in the region 8〈T〈116 K. The lifetime is relatively long at 8 K (130 ns), drops to a lower value at approximately 30 K, and then begins to decrease rapidly with increasing temperature in the neighborhood of 100 K. The highly resolved optical spectra show unambiguously that at 30 K the decrease in lifetime is associated with thermal isomerization to cis,trans-octatetrane in the excited 2 1Ag state. These data are combined with other spectroscopic and kinetic data to obtain estimates for the energies of all three double bond isomers in ground and excited states and the activation energies for double bond isomerization in the excited state.

Notes: We present a method whereby the excited triplet state of diphenylmethylene can be populated in a spin-polarized manner. The method features sensitization by benzophenone triplet excitons. We ascertain the spin-polarized nature of the excited triplet by observing magnetic field effects on the fluorescence. The observed effects provide information concerning the spin dipolar and intersystem crossing nature of excited diphenylmethylene.

Notes: A microwave spectroscopic method has been developed to study elementary reactions in real time through in situ observation of rotational spectra of reaction intermediates such as free radicals with lifetime as short as 1 ms. This method was applied to the O(3P)+ethylene reaction in order to assess the roles of (a) vinoxy+H and (b) CH3+CHO channels in the initial process. The reaction was initiated by irradiating an N2O/C2H4 mixture containing a trace amount of mercury with the 253.7 nm mercury resonance line, and the time evolution of vinoxy, HCO, and H2CO was followed by measuring their microwave absorption intensities as functions of time. The branching ratio was thus determined to be 0.4±0.1 and 0.5±0.1 for (a) and (b), respectively, at the sample pressure of 30 mTorr. The present result agrees with those obtained by Hunziker et al. [J. Photochem. 17, 377 (1981)] using much higher pressures of samples, but is not compatible with the observation of Buss et al. [J. Photochem. 17, 389 (1981)] that (a) is dominant in collision-free conditions.

Notes: The dependence of the chemiluminescence channels on incident spin-orbit state has been determined by utilizing optical pumping state selection for the reactions of metastable Ba(6s5d 3DJ) with Cl2, Br2, CCl4, N2O, and NO2. The cross sections for the neighboring metastable 1D level were related to those for the 3DJ multiplet by optical pumping on an intercombination line. For the halogen-containing reagents, the spin-orbit dependence of the reactivity was substantial for BaX B–X emission and was in the order J=3〉J=2〉J=1. The differences in reactivity were somewhat less for the C–X chemiluminescence channel in the Cl2 and Br2 reactions. Essentially no spin-orbit dependence was observed in the N2O and NO2 reactions.

Notes: Thermal electron attachment to NO2 in He, Ar, and N2 as buffer gases has been studied at various temperatures between 257 and 354 K with the pressure range of a few to 900 Torr. At buffer-gas pressures from a few to several tens Torr, the effective two-body attachment rate constant (keff) initially increases and then shows a saturation at all temperatures for three buffer gases. Further increase of the pressure up to near 1 atm gives a gradual increase of keff for Ar and N2 systems, whereas no such an increase is seen for He at three different temperatures. These behaviors are explained by a mechanism of a combination of the two-step three-body attachment and the attachment to van der Waals molecules containing NO2. A simulation analysis supports this explanation. A remarkable finding is that in all systems studied keff's become larger at lower temperatures. An Arrhenius plot for the initial two-body attachment rate constant gives an activation energy of −0.17 eV. It has been suggested that an excited triplet state of NO−2 ion may be the species formed by the initial electron attachment.

Notes: A wide variety of reactive phenomena in gaseous swarms is studied using an extension of momentum-transfer flight theory. Empirical formulas, such as the generalized Einstein relations and the Wannier energy relation are derived, with modifications due to reactive effects made explicit. The emphasis throughout is on simple mathematics and succinct physical reasoning.

Notes: Hydrogen peroxide has been optically excited at a wavelength of 266 nm and the OH photofragment completely characterized by Doppler and polarization spectroscopy using the laser-induced fluorescence technique. The entire internal state distribution (vibration, rotation, spin, and Λ components), translational energy, angular distribution, rotational alignment, and vector correlations between rotational and translation motions of OH products is measured. The hydroxyl radicals are formed in the X 2Π3/2,1/2 ground state with 90% of the available energy (248 kJ/mol) being released as OH recoil translation. The angular distribution is nearly a sin2 θ distribution about the electric vector of the photolysis laser. The internal motion of OH is vibrationally cold (no vibrationally excited OH was found) while the rotational excitation in v‘=0 can be described by a Boltzmann distribution with a temperature parameter of Trot=(1530±150) K. The two spin states are found to be populated nearly statistically, in contrast to the Λ components which show an increasing inversion with increasing OH rotation. The observed profiles of recoil Doppler broadened spectral lines are strongly dependent on the nature of the transition, the excitation–detection geometry, and the relative polarizations of the dissociating and analyzing laser light. However, the line intensities show only a minor dependence on geometry and polarization indicating a low alignment of OH photofragments (A(2)0≤0.1). For the first time the vector correlation between product rotational and translational motions was analyzed and evaluated in terms of the three bipolar moments β20(20), β00(22), and β20 (22). The bipolar moment β20(20) corresponds to the conventionally defined anisotropy parameter β=2β20(20) =−0.71. The angular distribution peaks in the direction perpendicular to the electric vector of the dissociating laser light, indicating the predominant electronic excited state in H2O2 being of 1A symmetry. The moment β00(22) increaseswith JOH showing a bias towards vOH and JOH being parallel to one another. The moment β20(22) is a measure of the mutual correlation of the fragment translational and rotational vectors and the transition dipole vector μ in the parent molecule. The positive value of this moment [β20(22)=0.11] indicates that the expectation value of 〈J2y〉 should be very small when μ is parallel to the z axis and vOH perpendicular to μ (x axis). 〈J2x〉 originates in the torsional motion in the H2O2 parent molecule, while 〈J2z〉 reflects the bending vibration of nearly planar H2O2 where the H atoms are in the trans position.

Notes: A master-equation formalism is applied to the problem of overtone-induced isomerization of CH3NC to CH3CN. The results are compared to the experiments of Reddy and Berry, who measured the yield of isomerization as a function of pressure after excitation to the fourth and fifth overtones of the CH stretching mode. The master-equation model predicts the yield and the curvature in the yield−1 vs pressure plots observed in the experiments. For the lower overtone (5←0) the results are consistent with a simple strong-collider model. However, even under strong-collider conditions the yield is very sensitive to the parameters in the master equation. For the upper overtone (6←0) the data do not fit a strong collider model and multistep deactivation dominates. We are able to determine from the data the average energy transferred in a collision by assuming a particular form for the energy-transfer function. In addition, the effect of changing the shape of the energy-transfer function is investigated.

Notes: A similarity transference procedure for the calculation of molecular parameters is proposed. The theoretical relationships between the direct transference method normally used and the proposed procedure are discussed. The importance of adequately defining the similarity models used in the transference calculations is emphasized. As an example, similarity models are constructed using the experimental and the STO-3G and 4-31G molecular orbital values of the atomic polar tensors of the HCN, C2H2, CH3CN, C4H2, CH3CCH, and C2N2 molecules. Partial least squares calculations based on these similarity models and using the STO-3G and 4-31G values of the atomic polar tensors of HC3N result in estimates of the experimental tensors of cyanoacetylene which have about one-half or less the root mean square error of the molecular orbital values. Also the partial least squares detection of potentially unreliable estimates of polar tensor elements is illustrated.

Notes: Expectation values of various operators with respect to nonrelativistic, self-consistent-field wave functions of good quality for 46 diatomic molecules are computed to examine the differences between the relativistic kinetic energy 〈Hr〉 and the quasirelativistic kinetic energy 〈Tnr〉+〈Hmv〉 in which 〈Tnr〉 is the nonrelativistic kinetic energy and 〈Hmv〉 is the mass–velocity correction. Then 〈Hrc〉=〈Hr〉−〈Tnr〉=〈H mv〉+〈δE〉 is the full relativistic correction to the kinetic energy. 〈Hrc〉 can differ appreciably from 〈Hmv〉 for molecules containing at least one atom with a moderately large atomic number Z. These differences are greatly amplified when the relativistic corrections to dissociation energies are considered; the mass–velocity contribution to the binding energy is found to be inaccurate even for moderate values of Z. Great care is necessary to ensure that the molecular and atomic calculations are of comparable accuracy. A qualitative argument is provided to explain why 〈Hmv〉 can provide a reasonable approximation to 〈Hrc〉 for small enough Z despite the fact that the two operators are inequivalent for αp≥1 where α is the fine structure constant and p is the momentum. Finally the asymptotic behavior of the pertinent integrands is used to show why the numerical evaluation, in momentum space, of 〈Hrc〉 is easier than that of 〈Hmv〉.

Notes: Ab initio self-consistent field and configuration interaction (SCF-CI) calculations with a minimal basis set are carried out on a few lower π→π* states of free base porphin and free base chlorin. Excitation energies and oscillator strength of these states are calculated and are compared with observed values. For the two lowest states of porphin and chlorin, the calculated excitation energies and oscillator strength are in good agreement with the observed ones. In particular, a characteristic of chlorin spectra is reproduced by the calculation. Namely, the oscillator strength of the lowest transition is several times larger than that of porphin but the oscillator strength of the second transition is similar to that of porphin. Such differences of the oscillator strength between chlorin and porphin are explained by an interchange of the order of two highest occupied orbitals and that of the two lowest unoccupied orbitals. Those differences are caused by a change of orbital shapes of chlorin due to the addition of two hydrogen atoms to the porphin skeleton.

Notes: High accuracy quantum chemical calculations (CASSCF and contracted CI) have been performed on the transition metal hydrogen complexes in the Mg2FeH6 and Mg2NiH4 crystals. The effects from the surrounding crystal are simulated by a Madelung field. If the conventional charge of +2 is assumed for the magnesium ions the transition metal hydrogen complexes receive a charge of −4. Good agreement with experiment is then obtained for the bond distance and totally symmetric vibrational frequency for the iron complex. The calculated Fe–H distance in the octahedral FeH4−6 complex is 1.56 A(ring), which is identical to the value obtained from neutron diffraction data on the Mg2FeH6 crystal. For the nickel compound also the unconventional case of Mg+ has to be considered, yielding a 2− transition metal hydrogen complex. The structure for the NiH4 complex is not as well determined as the iron complex. Experimental data has been interpreted in terms of a planar structure but a tetrahedral structure is not ruled out. Our calculations show that the conformation of the complex depends on the charge it is assumed to have. A 2− complex leads to a square planar structure and a 4− complex leads to a tetrahedral structure.

Notes: The harmonic vibrational frequencies and infrared intensities of the H2Si=CH2 molecule have been predicted by ab initio molecular quantum mechanics. Self-consistent-field (SCF) and configuration interaction (CI) methods have been used in conjunction with a double zeta plus polarization (DZ+P) basis. Comparison with pertinent experimental data is made.

Notes: Recently developed approximation methods for quantum mechanical problems which treat the spatial dimension D as an expansion parameter offer approximations to energy levels at arbitrary D. Rather than simply being a detour to the D=3 case, there is physical interest in nonphysical values of D due to degeneracies between states in different dimensions. For example, such degeneracies make it possible to calculate some excited states of two-electron atoms in three dimensions from the ground state energy at nonphysical values of D. Such relationships can be exploited in a simple derivation of the hydrogen atom spectrum in arbitrary D, using only the solution at D=1 and a combination of inter- and intradimensional symmetry arguments. Applications to the Yukawa potential and an anharmonic oscillator are also presented. A large class of interdimensional degeneracies is found for two-electron atoms. Approximate degeneracies are also identified for these atoms which allow highly excited D=3 states to be treated as perturbed low-lying states in another dimension. The approximate degeneracies also serve to generalize the treatment of the hydrogen atom spectrum in a way appropriate to two-electron atoms.

Notes: The recently introduced method of partial averaging is developed into a general formalism for computing simple Cartesian path integrals. Examples of its application to both harmonic and anharmonic systems are given. For harmonic systems, where analytical results can be derived, both imaginary and complex time evolution is discussed. For two representative anharmonic systems, Monte Carlo path integral simulations of the imaginary time propagator (statistical density matrix) are presented. Connections with other Cartesian path integral techniques are stressed.

Notes: A detailed investigation is made into the use of adiabatic approximations for describing excited stretching and bending vibrations of the water molecule. The goal is to determine precisely how effective this approach can be in a fully quantum mechanical triatomic calculation which incorporates anharmonicities to all orders in each of the modes. Great care is taken to avoid introducing unnecessary limitations or approximations: (i) Curvilinear coordinates are used rather than the Cartesian coordinates which form the starting point for normal mode calculations; (ii) the exact quantum kinetic energy operator in these coordinates is used as the basis for both the adiabatic and full three-dimensional calculations; (iii) a Sorbie–Murrell-type potential energy surface is used, giving a reasonable representation of the ground electronic surface for large excursions from the equilibrium configuration. In addition to the bond and bond-angle variables of earlier local mode investigations, a slightly different set of fully curvilinear coordinates is also investigated. These coordinates are shown to provide a more nearly separable description in both the exact and adiabatic treatments of this specific problem. The conventional adiabatic approach, in which the slower bending mode experiences an effective force due to averaging over the faster stretching modes, is reaffirmed to be accurate for excited stretching states. For states with any appreciable bending excitation, however, it turns out that the adiabatic calculations quickly erode in reliability. In answer to this problem, the reverse adiabatic procedure (with the bend treated first) is also implemented here. While counterintuitive, this latter method is found to yield a significant improvement for the calculated bending overtones, as well as many of the combination bands. Thus, by thorough consideration of both the coordinates and order of averaging employed, the adiabatic method is shown to be very effective for either bending or stretching overtones in a realistic, fully anharmonic, triatomic vibrational problem. In addition, introduction of a new orthonormal set of basis functions for the bending angle overcomes some of the problems associated with use of the less flexible Legendre basis.

Notes: Using a basis set of 99 contracted Gaussian type orbitals and Meyer's coupled electron pair approximation (CEPA), several spectroscopic properties such as vibrational frequencies, rotational, and centrifugal distortion, and vibration–rotation coupling constants were calculated for different isotopomers of the azide anion N−3. The ν3 band (asymmetric stretch) of 14N−3 is predicted to be extremely intense.

Notes: Theoretical predictions for various neutral and charged forms of polyaniline are presented and discussed. The MNDO semiempirical method is used to predict geometries which serve as input for valence effective Hamiltonian (VEH) calculations of the electronic band structure. The VEH calculations provide predictions for the band gaps, ionization potentials, electron affinities, and redox potentials for the various forms of polyaniline. Where possible, comparison is made to experiment with generally favorable results. The discussion emphasizes the electrochemistry of polyaniline and the structural evolution of the polymer during electrochemical oxidation and reduction.

Notes: A novel, efficient, and accurate quantum method for the calculation of highly excited vibrational levels of triatomic molecules is presented. The method is particularly well suited for applications to "floppy'' molecules, having large amplitude motion, on potential surfaces which may have more than one local minimum. The discrete variable representation (DVR) for the angular, bend coordinate is combined with the distributed (real) Gaussian basis (DGB) for the expansion of other, radial coordinates. The DGB is tailored to the potential, covering only those regions where V(r)〈EMAX. The DVR permits a contraction of the primitive Gaussian basis to a small eigenfunction basis at each of the discretized values of the angular coordinate. It is shown for the floppy two-mode LiCN/LiNC system (fixed CN distance) that N lowest vibrational levels (N=131) can be converged to within 1 cm−1 (the lowest 117 to 0.1 cm−1) using only 3N basis functions. This appears to reduce the computational effort by a factor of 10–40 over standard methods. Moreover, only a very low order Gauss–Hermite quadrature, 3–5 points, is needed to evaluate each potential matrix element.

Notes: General and greatly simplified methods are presented for calculating terms in the exponent in Magnus and Magnus-like expansions to first order in a "small''perturbation and to infinite order in the overall Hamiltonian in the Schrödinger representation. These techniques are applied to four simple but important models and it is shown that in each case the Magnus exponent diverges for some range of the parameters of the model. This result casts serious doubt on the utility of the Magnus expansion in the Schrödinger representation. Several of the problems are also treated in the interaction representation giving results which converge throughout the useful range of parameters. There is no evidence that a similar question exists in this representation.

Notes: In the present work, the theoretical scheme applied to the study of hindered rotation of molecular impurity in rare gas matrices [J. Chem. Phys. 77, 1757 (1982)] is extended to librational motion. The librational motions are considered as the intermediate modes in the process of vibrational dephasing. The effect of libron–phonons coupling is formally calculated for a three dimensional lattice. The case of CO dispersed in Ar is examined within the frame of a simple cubic lattice. Despite the oversimplified model, the computed temperature dependence of the vibrational absorption band adequately reproduces the observed experimental trend.

Notes: A study is made of the surface–void, surface–surface, and surface–particle correlation functions that arise in expressions for transport properties associated with diffusion-controlled reactions and flow in porous media consisting of impenetrable spherical particles distributed randomly throughout void. The relationship between these three different correlation functions is noted for this model. We present an exact low-density expansion of the surface–particle correlation for such a distribution. All of the two-point correlation functions are computed, for the first time, from integral relations that we derived elsewhere, as a function of the distance between the two points for various sphere volume fractions up to 94% of the random close-packing value.

Notes: The heat capacity of NaOD from T=20 to 350 K is reported. An anomaly centered at T=153.2±0.1 K (ΔH=131.1±0.7 J mol−1, ΔS=0.864±0.005 J K−1 mol−1) corresponds to the onset of a low-temperature antiferroelectric phase. This transformation, which likely involves tunneling of the deuterons, does not appear to have an atmospheric pressure analog in NaOH above T=6 K.

Notes: The calculation of excess Gibbs' function and activity coefficients from simulation requires the precise evaluation of chemical potentials. Such high precision is usually not attainable using the conventional test-particle simulation methods. In this paper a special implementation of Widom's potential distribution theorem is developed and shown to give activity coefficients with high precision. Results for some model mixtures are presented. Comparison with results obtained via conventional simulation methods is made.

Notes: General expressions are derived for the pairwise fluctuations of thermodynamic properties in multicomponent systems, in terms of thermodynamic derivatives. The expressions are valid for an open system close to thermodynamic equilibrium in contact with an energy and mass reservoir. The system is defined by an arbitrary extensive constraint. Fluctuations involving the entropy S, the volume V, and the number of molecules Ni of each component, as well as their conjugate variables, are expressed as second-order derivatives of the energy U and Legendre-transformed functions. Fluctuations involving U itself are obtained in the entropy representation of the fundamental equation.

Notes: Hubbard–Onsager–Felderhof's (HOF) dielectric friction theory for rotational motion of ions is tested by using NMR orientational relaxation times recently measured for such polyatomic ions as ClO−4,SO2−4,PO3−4, etc. in D2O and CD3OD at infinite dilution. The HOF theory combined with the bulk solvent properties is qualitatively successful in explaining how rotational friction coefficients for ions with a high surface charge density depend on ion size and valence, temperature, and solvent. In the case of a low surface charge density, the continuum model shows its inherent limitations in the treatment of ion dynamics, rotational or translational. It turns out that the HOF theory serves an important theoretical framework for the interpretation of the semiempirical relation between orientational relaxation times of ions and the solvent viscosity divided by temperature.

Notes: In this study we simulate the temperature dependence of powder x-ray diffraction for ice Ih using a model ice sample and Metropolis Monte Carlo simulations. We permit the water molecules to interact through the revised central force potentials of Stillinger and Rahman [J. Chem. Phys. 68, 666 (1978)]. The model ice sample is a periodic unit cell possessing 192 water molecules which are arranged in their initial configuration to yield very small point dipole and quadrupole moments [J. Chem. Phys. 78, 5103 (1983) and J. Phys. Chem. 87, 4309 (1983)]. We report the x-ray powder simulations for input temperatures of 100, 200, and 260 K, comparing them with earlier work at 20 K. A full set of the corresponding structure factors is computed for 300 K. The results suggest a monotonic decrease in the logarithm of the diffraction intensities with increasing temperature and an abrupt change between 260–300 K. The drastically decreased values at 300 K are consistent with previous related work [J. Phys. Chem. 87, 4309 (1983)]. The results are compared with the literature on related x-ray measurements [Nature 188, 1144 (1960)]. Here we comment on the qualitatively good comparisons between experiment and simulation, we note quantitative differences between experiment and simulation, and we discuss the differences in light of limitations in both experiment and theory.

Notes: Vapor pressures of solid and liquid ozone have been measured over a temperature range 87 to below 66 K. The experiment was performed under flow conditions, the gas was analyzed by a precision mass spectrometer system. In the range of solid ozone two forms, supercooled and crystalline ozone, were found. A least square fit of the data for crystalline ozone resulted in the equation log P(Torr)=A+B/T where A=10.460 and B=−1021.6. The estimated uncertainty of the data is ±1.0%. A triple point temperature of 79.6±0.3 K was found where supercooled and crystalline ozone data intersect.

Notes: High resolution electron energy loss spectroscopy has been used to examine the effects of coadsorbed oxygen and carbon monoxide on the bonding of nitric oxide adsorbed on the Rh (111) surface. NO is converted from twofold bridge sites (1500–1600 cm−1) to linear sites (1840 cm−1) when annealed to 275 K in the presence of coadsorbed oxygen. Terminally bound linear NO does not dissociate, but desorbs molecularly, a finding consistent with the commonly observed oxygen inhibition of NO dissociation over noble metals. Coadsorbed CO does not significantly inhibit or promote the dissociation of a low coverage of nitric oxide compared to the clean Rh (111) surface. CO and NO appear to segregate and form islands of each species at 95 K if CO is adsorbed first. Annealing this adlayer above 250 K disperses the islands and mixes the adsorbate layer at temperatures below where NO dissociation and reaction with CO are significant. The Rh–CO bond is weakened in well-mixed CO+NO adlayers, indicating that the NO–CO repulsive interaction acts through the Rh substrate rather than directly between the adsorbed molecules.

Notes: Flow modification by dilute solutions of polymers in strong elongational flows is analyzed using a two-state model according to which the polymer contribution to the stress balance in the fluid is negligible as long as the polymers are in their coiled state (or are in the process of stretching) and becomes significant as they approach their steady configuration corresponding to an almost fully stretched state. The flow field is separated accordingly into weakly and strongly modified regions and, using mass and momentum conservation considerations, we obtain the flow profiles in these regions. Our prediction of a dip in the velocity and strain rate profiles agrees with experimental observations.

Notes: I point out that the results of a recent vapor pressure isotherm study of noble gas monolayers physisorbed on several lamellar halides by Millot, Larher, and Tessier can be explained in terms of an incipient triple point in some of the films and the usual liquid–gas critical point in others, whereas a critical point in all of the films is not in agreement with model calculations. Lamellar halides would thus provide a system in which the crossover, presumably via a critical endpoint, between the two cases can be studied experimentally.

Notes: Visible absorption, rheological measurements, polarizing light microscopy, and temperature–time dependent quasielastic light scattering was used to examine the dilute–semidilute properties of polydiacetylene solutions, specifically poly(4BCMU). A good (tetrahydrofuran) and poor (toluene) solvent were selected for examining solution properties over a broad range of solvent quality. In this particular instance, poly(4BCMU) dissolved in both the pure solvents and their respective mixtures. Based on the observation that dramatic color changes (yellow–orange–red) can occur with the hydrodynamic radius remaining invariant, it is quite likely that the color changes are due to a local modification in the conjugation length. That is, on a molecular level, a local stiffening of the segments occurs as the magnitude of the red absorption component increases. Interestingly, the stability of the solutions towards aggregation/phase separation also correlates well with the absorbance value of the red peak. Furthermore, careful thermal aging measurements confirm that as the aggregation process proceeds, the absorption spectrum remains unchanged, although the initially relatively broad spectral peaks do become significantly easier to resolve. As a result, an "intermediate'' peak (due to an intermediate conjugation length) in the orange region is resolvable. Finally, polarized light microscopy confirms that the aggregates formed from thermally aged solutions containing an orange/red component are highly anisotropic, indicative of high chain alignment. Similar results are obtained from evaporating off the solvent from a dilute yellow solution. The implications of this self-alignment process are discussed.

Notes: Direct photodissociation of molecules at structured metallic surfaces is considered, where the concepts of optimal and critical distances are introduced into the description of this phenomenon. Numerical results for the distortion of the line shape and the enhancement ratio are obtained for a shallow sinusoidal grating.