Collisional polarization

The simplest and most straightforward idea for producing collisionally polarized molecules in a thermal cell consists of using collisions with the participation of particles which are polarized in the laboratory frame. It seems that the earliest one was the method based on collisions of atoms which have been optically pumped (optically oriented in their ground state) by the Kastler method see Section 1.1, Fig. 1.1. If the gas constitutes a mixture of a molecular and an atomic component, the conditions being specially created in such a way as to produce such optical orientation of the atoms, we must expect, from considerations of spin conservation in molecular reactions, that polarization of the molecular component must also emerge. 

For dielectric relaxation to be observed there has to be a change in the polarizability of the media under the influence of an applied field. This selection rule implies that dispersion will only be detected in polar materials. However, in certain cases contributions to the polarizability have been detected in non polar materials and are ascribed to collisional polarization effects. If two non polar molecules collide there is the possibility that distortion of the electron density nd atomic positions may result in the formation of a tiansient dipole or multipole. The induced polarization can be destroyed by further collision with other non activated molecules. The lifetime of these collisionally activated molecules may be many times the collision frequency and it then becomes possible to observe the reorientational motion of the induced dipoles, which act as though they were permanent dipoles. 

Figure Bl.7.7. Summary of the other collision based experiments possible with magnetic sector instruments (a) collision-mduced dissociation ionization (CIDI) records the CID mass spectrum of the neutral fragments accompanying imimolecular dissociation (b) charge stripping (CS) of the incident ion beam can be observed (c) charge reversal (CR) requires the ESA polarity to be opposite that of the magnet (d) neutiiralization-reionization (NR) probes the stability of transient neutrals fonned when ions are neutralized by collisions in the first collision cell. Neutrals surviving to be collisionally reionized in the second cell are recorded as recovery ions in the NR mass spectrum.

The molecular beam and laser teclmiques described in this section, especially in combination with theoretical treatments using accurate PESs and a quantum mechanical description of the collisional event, have revealed considerable detail about the dynamics of chemical reactions. Several aspects of reactive scattering are currently drawing special attention. The measurement of vector correlations, for example as described in section B2.3.3.5. continue to be of particular interest, especially the interplay between the product angular distribution and rotational polarization. 

Another approach to ET reactions originated in the work of Weiss and Libby, who suggested that activation energy for ET reactions in solution does not arise from the collisional-vibrational or electrostatic interaction of the solvent in the first and second solvation sphere, but rather from continuum solvent polarization fluctuation far out in solution. This approach, called continuum theory, was further developed by Kubo and Toyozawa, " Marcus, Platzmann and Erank, and by Levich and Dogo-nadze. ... 

The rotational dynamics of nitrogen and carbon dioxide were recorded by Akhmanov and Koroteev [7]. The transients look similar to the transients by Morgen et al. [8], recorded with time resolved Raman induced polarization spectroscopy [9]. A fs-DFWM experiment was performed by Frey et al. [10] on diatomics and linear polyatomics. To prevent collisional dephasing, they transferred the method into the expansion zone of a molecular beam. In succession, experiments on linear molecules and symmetric tops were performed on molecules like CHCI3 [11] and CgHf, [12], Transients of asymmetric tops like the near oblate pyrimidine, pyrazine and pyridine [13] and SO2 [11] were reported in the following years. 

Interaction-induced absorption (as the new features were called early on [353]) has stimulated considerable interest. For a long time, explanations were attempted in terms of weakly bound (02)2 polarization molecules (that is, van der Waals molecules), but some of the early investigators argued that unbound collisional pairs might be responsible for the observed absorption. More recently, a study of the temperature dependence of the induced intensities has provided evidence for the significance of collisional complexes. The idea of absorption by collisionally interacting, unbound molecular pairs was, however, not widely accepted for decades. 

Welsh suggested correctly that similar transitions take place even if the molecular pair is not bound. The energy of relative motion of the pair is a continuum. Its width is of the order of the thermal energy, Efree 3kT/2. Radiative transitions between free states occur (marked free-free in the figure) which are quite diffuse, reflecting the short lifetime of the supermolecule. In dense gases, such diffuse collision-induced transitions are often found at the various rotovibrational transition frequencies, or at sums or differences of these, even if these are dipole forbidden in the individual molecules. The dipole that interacts with the radiation field arises primarily by polarization of the collisional partner in the quadrupole field of one molecule the free-free and bound-bound transitions originate from the same basic induction mechanism. 

Of a special astronomical interest is the absorption due to pairs of H2 molecules which is an important opacity source in the atmospheres of various types of cool stars, such as late stars, low-mass stars, brown dwarfs, certain white dwarfs, population III stars, etc., and in the atmospheres of the outer planets. In short absorption of infrared or visible radiation by molecular complexes is important in dense, essentially neutral atmospheres composed of non-polar gases such as hydrogen. For a treatment of such atmospheres, the absorption of pairs like H-He, H2-He, H2-H2, etc., must be known. Furthermore, it has been pointed out that for technical applications, for example in gas-core nuclear rockets, a knowledge of induced spectra is required for estimates of heat transfer [307, 308]. The transport properties of gases at high temperatures depend on collisional induction. Collision-induced absorption may be an important loss mechanism in gas lasers. Non-linear interactions of a supermolecular nature become important at high laser powers, especially at high gas densities. 

An important induced dipole component of pairs involving molecules is multipolar induction. Specifically, the lowest-order multipole consistent with the symmetry of H2 is the electric quadrupole. Each H2 molecule may be thought of as being surrounded by an electric field of quadrupolar symmetry that rotates with the molecule.-In that field, a collisional partner X is polarized, thus giving rise to an induced dipole moment which in turn is capable of emitting and absorbing light. For like pairs, molecule 1 will induce a dipole in molecule 2 and 2 will induce one in 1. In... 

If at least one of the interacting particles is a molecule, further induction mechanisms arise. Molecules are surrounded by an electric field which may be viewed as a superposition of multipole fields. A collisional partner will be polarized in the multipole field and thus give rise to induced dipole components. In the case of symmetric diatoms like H2 or N2, the lowest-order multipole is a quadrupole and asymptotically, for R - 00, the quadrupole-induced dipole may be written as [288, 289]... 

While exchange- and dispersion-induced dipole components are of a quantum nature, the multipole-induced dipole components can be modeled by classical relationships, if the quantum effects are small. For many systems of practical interest, multipolar induction generates the dominant dipole components. The classical multipole induction approximation has been very successful, except for the weakly polarizable partners (e.g., He atoms) [193]. It models the dipole induced in the collisional partner by polarization in the molecular multipole fields. 

Both photon-assisted collisions and collision-induced absorption deal with transitions which occur because a dipole moment is induced in a collisional pair. The induction proceeds, for example, via the polarization of B in the electric multipole field of A. A variety of photon-assisted collisions exist for example, the above mentioned LICET or pair absorption process, or the induction of a transition which is forbidden in the isolated atom [427], All of these photon-assisted collision processes are characterized by long-range transition dipoles which vary with separation, R, as R n with n — 3 or 4, depending on the symmetry of the states involved. Collision-induced spectra, on the other hand, frequently arise from quadrupole (n = 4), octopole (n = 5) and hexadecapole (n = 6) induction, as we have seen. At near range, a modification of the inverse power law due to electron exchange is often quite noticeable. The importance of such overlap terms has been demonstrated for the forbidden oxygen —> lD emission induced by collision with rare gases [206] and... 

Collisional redistribution of radiation. A system A + B of two atoms /molecules may be excited by absorption of an off-resonant photon, in the far wing of the (collisionally) broadened resonance line of species A. One may then study the radiation that has been redistributed into the resonance line - a process that may be considered the inverse of pressure-broadened emission. Interesting polarization studies provide additional insights into the intermolecular interactions [118, 388]. 

Here, a. and a L are the polarizabilities of the diatom parallel and perpendicular to the internuclear separation, R12. The electrostatic theory accounts for the distortions of the local field by the proximity of a point dipole (the polarized collisional partner) and suggests that the anisotropy is given by ft Rn) 6intermolecular interactions). This is the so-called dipole-induced dipole (DID) model, which approximates the induced anisotropy of such diatoms often fairly well. It gives rise to pressure-induced depolarization of scattered light, and to depolarized, collision-induced Raman spectra in general. 

The results obtained so far with collisional energy-transfer spectroscopy are restricted to excited sodium atoms A = Na(32/,3/2) and quenching by a variety of simple polar and nonpolar molecules. The technique is applicable to any vaporizable molecule and will be available for a number of other atoms as well in due course with the progress of laser technology. The E-V-R transfer processes from and to sodium atoms have a number... 

The possibility of deactivation of vibrationally excited molecules by spontaneous radiation is always present for infrared-active vibrational modes, but this is usually much slower than collisional deactivation and plays no significant role (this is obviously not the case for infrared gas lasers). CO is a particular exception in possessing an infrared-active vibration of high frequency (2144 cm-1). The probability of spontaneous emission depends on the cube of the frequency, so that the radiative life decreases as the third power of the frequency, and is, of course, independent of both pressure and temperature the collisional life, in contrast, increases exponentially with the frequency. Reference to the vibrational relaxation times given in Table 2, where CO has the highest vibrational frequency and shortest radiative lifetime of the polar molecules listed, shows that most vibrational relaxation times are much shorter than the 3 x 104 /isec radiative lifetime of CO. For CO itself radiative deactivation only becomes important at lower temperatures, where collisional deactivation is very slow indeed, and the specific heat contribution of vibrational energy is infinitesimal. Radiative processes do play an important role in reactions in the upper atmosphere, where collision rates are extremely slow. 

Now, we proceed to the evaluation of the collisional integrals Pc, qc, and y by using Eq. (5.293) as the form of a collisional pair distribution function. Use the coordinates in Fig. 5.12, in which ez is chosen to be parallel to the relative velocity vn. 9 and 0 are the polar angles of k with respect to ez and the plane of ez and ex, respectively. ex, ey, and ez are the three mutually perpendicular unit vectors corresponding to each coordinate in Fig. 5.12. k, as mentioned in the previous section, is the unit normal on the collision point directed from the center of particle 1 to the center of particle 2. Thus, we have... 

Reactions within a van der Waals (vdW) complex of calcium with hydrogen halides (HC1 and HBr) lead to electronically excited calcium halides. These reactions have been quite extensively studied in full collisions of excited calcium beams (Brinckmann et al. 1980 Brinckmann and Telle 1977 Rettner and Zare 1981, 1982 Telle and Brinckmann 1990). The electronic excitation of the calcium atom results in a strong chemiluminescence under collisional conditions. The efficiency of this chemiluminescence depends upon the electronic state and the fine structure component, and the final product state is influenced by the preparation conditions of the collision. In the reaction Ca(4s4p1P1) + HC1, the direction of the polarization of the P orbital with respect to the collision relative velocity (pK or pff) has an effect on the branching ratio to the products CaCl, A2n or B2X+ (Rettner and Zare 1981, 1982). 

---