6. Studies in Electron-Molecule Collisions
In collaboration with Robert R. Lucchese and Frank J. Lovas, NIST
Current and past support: Environmental Protection Agency and NSF-CTS

In collaboration with Professor Robert R. Lucchese, we are investigating application of ultrahigh resolution(24 pico-eV) final state resolution techniques that we have developed to chemical processes and energy transfer that occur in the collision of mono-energetic electrons with molecules. Initially, we are emphasizing electron collisions with fluorinated hydrocarbons that are relevant to semiconductor manufacturing processes, and the application of surface wave plasma abatement of perfluorocompounds and other global warming emissions from semiconductor manufacture. Experimentally, we are studying a wide variety of such processes from which we can then obtain a variety of cross sections. Theoretically, we are also developing and validating numerical methods, for studying the same processes. With these computational methods, we are then able to estimate absolute! cross sections and predict the corresponding cross sections for other molecules that are not amenable to experimental study.Initially, we are studying the interaction between monoenergetic electrons and fluorocarbon compounds expanded in a supersonic molecular beam. The primary spectroscopic instrumentation for characterizing final molecular state distributions are a pulsed-nozzle Fourier-transform microwave spectrometer and our newly developed terahertz spectrometer, both capable of 6 kHz resolution or better. This corresponds to a final state resolution of 24 pico-eV, a resolution more than a million times better than the best electron loss techniques previously available for such studies. We consider this capability to be of of some significance, as many experiments that were previously not considered viable become readily accessible to this approach. The final states of interest include both the vibrational excited states of the target molecule, which can be distinguished by their differing rotational parameters, and any fragment molecule obtained by dissociation and/or ionization of the target molecule at resolutions 6 to 8 orders of magnitude better than currently available electron energy loss techniques. We have already demonstrated that cross sections for these various processes can be studied as a function of electron impact energy over the range of impact energies of 5 eV to 1000 eV. The state specific spectroscopic studies of the these interactions provide the basic data for the modeling of vibrational excitation and dissociative processes. Theoretically, we are also developing the appropriate tools for studying these processes. We are making three extensions of previously developed computational tools for studying the electron-molecule scattering:
i) Firstly, we are implementing a contracted partial-wave expansion that will allow the single-center method to achieve convergence in large non-symmetric molecules. ii) Secondly we are implementing an off-shell T matrix approach for computing vibrational excitation. iii) Finally we are implementing coupled electronic channels so that we can consider dissociation processes which occur through electronic excited states of the target molecules. As part of the electronic excitation problem we are considering appropriate polarization potentials which can approximately include the effectsof neglected electronic states. Initially, our combined experimental and theoretical investigations are focused on molecules that are important in semiconductor manufacturing processes. Specifically, we are concentrating on the sequence CFxH4-x (x = 1,2,3) which allow for a systematic study of how the F atoms affect the electron scattering properties of the molecules. In addition, we are also considering monoenergetic electron collisions with CF4, C2F6, C3F8, and c-C4F8. The current introduction of 300 mm semiconductor tools require sophisticated modeling to optimize such manufacturing processes rather than empirical approaches that have been used in the past. Furthermore, the current recognition (Kyoto Conference 1997) that PFC and HFC semiconductor emissions are the cause of virtually irreversible global warming over tens of thousands of years will not only require optimization for production efficiency but also processes that are environmental compatible. The information generated during such studies provide electron-molecule collision parameters that have hitherto have not been available for modeling of semiconductor etch, chamber clean, and plasma based abatement devices that are currently being developed for effective abatement of such PFC and HFC emissions. Our investigations thus have direct relevance to semiconductor manufacturing and related environmental issues.

[9] I. Shimamura and K. Takayanagi, Electron-Molecule Collisions (Plenum, N.Y, 1984).

[10] i) M. Kushner, Workshop on the Treatment of Gaseous Emissions via Plasma Technology (NIST, Gaithersburg, Maryland, 1995). ii) K. H. Becker, "Status and Availability of Collision Data for Plasma Remediation," ibid (NIST, Gaithersburg,MD, 1995).

[11] D. E. Milligan and M. E. Jacox, J. Chem. Phys. 48, 2265-2271 (1968).

[12] F. A. Gianturco and A. Jain Phys. Rep. 143, 347 (1986).

[13] F. A. Gianturco, R. R. Lucchese, N. Sanna, and A. Talamo, "A Generalized Single Center Approach for Treating Electron Scattering from Polyatomic Molecules," in ElectronCollisions with Molecules, Clusters, and Surfaces, edited by H. Ehrhardt and L. A. Morgan, (Plenum Press, New York, 1994), p. 71.

[14] N. F. Lane, Rev. Mod. Phys. 52, 29 (1980).