WiZeus > Biography | Education | Scientific | Archive (Demjanjuk, Suslensky, Harasymiw, etc.) >

INRS-Energie Internal Report #____
May 1986
W.W. Zuzak

EXCITATION TRANSFER IN HELIUM
in a
HOLLOW CATHODE DISCHARGE
via
LASER-INDUCED FLUORESCENCE

PREAMBLE
This report is a summary of the experimental work carried out between Sept. 1984 and Sept. 1985, as well as the theoretical work on the interpretation of the results carried out intermitantly up to the present time (while preparing fluorescence experiments for the Tokamak de Varennes). Throughout this report we shall often refer to the 441 pages of experimental notes and more than 70 pages of theoretical notes (e.g. "see page 243 expt. notes") so as to facilitate easy access to the original data.

We note that it was necessary to repair a leaking capacitor and to replace the flash lamp in the Phase-R DL2100C dye laser to be used for the fluorescence experiments to detect heavy metal impurities sputtered from the walls, limiters and/or neutralization plates in the Tokamak de Varennes (TdeV). The original intention was to test out the operation of the laser and the concept of fluorescence due to excitation transfer from the 31P to the 31D levels in a helium plasma excited in a hollow cathode discharge. However, because the preliminary results were so encouraging (and because the construction of the TdeV was significantly delayed), it was decided to do an in depth study of excitation-transfer amongst the n=3 singlet levels in helium which is the subject of this report.

In addition to the intrinsic scientific interest of this research, we wished to examine the feasibility of applying these types of measurements to the TdeV.

INTRODUCTION
Since the development of tunable dye lasers, laser-induced fluorescence has become a valuable diagnostic tool in the field of plasma physics. Typically, the technique involves using a pulse of laser radiation at a particular wavelength to excite ground state (or metastable) electrons bound to an atom into a higher energy level and to observe their fluorescence (preferably at a different wavelength) as they decay to a (different) lower level.

Many measurements of the concentrations and fluxes of impurity species such as Fe, Ni, Ti, Al, Mo, Zr, Si, C sputtered from the walls enclosing fusion-type plasmas have been reported. (See over 20 papers on LIF in Conference Proceedings on "Plasma Surface Interactions in Controlled Fusion Devices" in J. Nuc. Mater. 111 & 112 (1982) and J. Nuc. Mater. 128 & 129 (1984)). One of the more powerful and sophisticated systems developed at the Argonne National Laboratory for such measurements has been recently described by Young et al[3].

The method can also be applied to the detection of hydrogen[4] and helium atoms in particular, as well as a wide variety of other atoms, ions, molecules, radicals. For example, Kychakoff et al[5] have made two-dimensional measurements of OH molecular concentrations in turbulent flame fronts.

Because of its relevance to fusion plasmas, we have a particular interest in applying LIF to helium. In this case an extra step utilizing excitation-transfer between excited states is involved. Thus for example, when 501.6 nm laser radiation is used to raise an electron from the metastable 21S state of helium to the 31P state, there is a substantial probability that a collision with a free electron or another atom will transfer the excitation to a different electronic energy level (the 31D state, for example) before the electron can decay back down to the 21S state via spontaneous 501.6 nm radiation or to the 11S state via 53.7 nm radiation.  Such excitation-transfer to the 31D state is detected by monitoring the spontaneous 667.8 nm radiation to the 21P level.

Historically, measurements of excitation-transfer cross-sections in helium have been carried out in two distinct regions according to the concentrations of electrons and neutral atoms in the plasma. In the low pressure, high electron density region (where ne < 10-5 na), collisions with electrons are the dominant excitation-transfer mechanism. Burrell and Kunze[6] have measured such e-a* transfer cross-sections from the 31P and 41D levels at pressures from 0.01 to 0.05 Torr, ne = (0.1 - 4.0)x1012cm-3, Te = 2.8 - 4.7 eV. Tsuchida et al [7],[8] have reversed the procedure to measure electron density profiles using known electron excitation-transfer cross-sections.

In the high pressure, low electron density region where a-a* collisions are presumably the dominant excitation-transfer mechanism, Dubreuil and Catherinot[9] (P = 0.2 - 7.0 Torr, Ta = 325 K, ne = (0.5 - 8.0)x1010cm-3, Te = 3 - 20 eV) have confirmed the work of Wellenstein and Robertson[10] (P = 1 - 4 Torr, Ta = 320 K, ne = (1 - 27)x1011cm-3, Te = 3.5 - 6.4 eV).

In this report we have specifically chosen to work in the intermediate pressure region (P = 2.6 Torr, Ta = 400 K, ne = 3.3x1011cm-3, Te = 0.54 eV) where the fluorescence signals and excitation-transfer rates are maximized. As described below, the published excitation-transfer cross-sections for e-a* and a-a* collisions are insufficient to account for the observed rates. Indeed, we postulate that in addition to e-a* and a-a* collisions other processes such as the formation and subsequent destruction of excited helium molecules contribute to the large excitation-transfer rates between the 31P and 31D levels.

Our STANDARD dc helium plasma is produced in a hollow cathode discharge (HCD) operated at 2.6 Torr, I = 60 ma, ΔV = -456 volts. Under these conditions the 21P and 23P states as well the unstable 21S and 23S states are all reasonably well populated. This would, in principle, allow us to "pump" to any of the higher lying (n =/> 3) s, p, d states with appropriate laser radiation in the wavelength region from 260.0 nm to 728.1 nm. However, due to equipment and time limitations, we have limited ourselves to observing 17 spectral lines from 328.0 nm to 728.1 nm and only pump the n = 3 singlet levels with 501.6, 667.8 and 728.1 nm radiation.

The hollow cathode discharge and the measured steady state plasma parameters and population densities are presented in section 2.

The description of the laser, the fluorescence experiment and the fluoresence results are given in section 3.

The principle of the method used to relate the measured fluoresence signals to excitation-transfer rates and our attempts to obtain appropriate e-a* and a-a* excitation-transfer cross-sections from these rates is described in section 4.

[ ... ]
REFERENCES
[1] Conference Proceedings on "Plasma Surface Interactions in Controlled Fusion Devices" in J. Nuc. Mater. 111 & 112 (1982)
[2] ibid, J. Nuc. Mater. 128 & 129 (1984)
[3] C.E. Young, D.M. Gruen, M.J. Pellin, W.F. Calaway; Fusion Technology 6, 434 (1984)
[4] Hα Fluorescence
[5] G. Kychakoff, R.D. Howe, R.K. Hanson, M.C. Drake, R.W. Pitz, M. Lapp, C.M. Penney; Science 224, 382 (1984)
[6] C.F. Burrell and H.J. Kunze; Phys. Rev. A 18, 2081 (1978)
[7] K. Tsuchida, S. Mayake, K. Kadota and J. Fujita; Plasma Physics 25, 991 (1983)
[7a] N. Yasumaru, S. Oku, T. Fujimoto and K. Fukuda; J. Phys Soc. Japan 49, 696 (1980)
[8] K. Tsuchida, Japanese Journal of Applied Physics 23, 338 (1984)
[9] B. Dubrueil and A. Catherinot; Phys. Rev. A 21, 188 (1980)
[9a] B. Dubrueil and P. Prigent; J. Phys. B: At. Mol. Phys. 18, 4597 (1985)
[10] H.F. Wellenstein and W.W. Robertson; J. Chem. Phys. 56, 1072 (1972); 56, 1077 (1972); 56, 1414 (1972)
[ ... ]