I OFI $ D ORNL P 1005 : ? . . i EEEFEFEE 1.4 1.16 MICROCOPY RESOLUTION TEST CHART NATIONAL BUREAU OF STANDARDS - 1963 m. Orxa-p.joos XOTIE-SP -.nr . OFFICIAL ON THE POSSIBILMY OF EXCTATION - HEATING OF TONS TO HIGH TEMPERATURE S : . J. Rand McNally, Jr. FEB 1 1 1965 Oak Ridge National Laboratory Oak Ridge, Tennessee -- 7 This paper was submitted for publication in the opon literature at loauis months prior to the issuance dato of this Micro- card. Since the 1.8.A.E.C. has no ovi- donco that it has been published, the pa- per to being distributed in Microcard form as a prepruit. The observation of carbon fons up to 100 times motter" than the electrons in very long, steady state, high current, magnetically-confined, carbon and gas- fed carbon' arcssuggested the presence of a new type of energy puming mechanism which we call excitetion-heating. The pumping cycle under consideration 189 cat(as) + e + C2**(spo) + e - 6.5 ev, ion excitation c2+*(3p0) + c2**(3p0) + cat(+8) + cat(45) + 13.0 ev, ion heating Chot + écola Cwarm + ewarm) 1on cool.ing with radiative losses and energy feed to the electrons provided by the arc power. . The metastables, c2**(pº), would constitute an excess population in ordinary equilibrium because of the modest electron temperature of about 42,000 OK (5.4 ev) and the statistical weight of 9 for the spo 1. vels. compared to only 1 for the ground Is level, 1.e., cz**/c2* = 1.5. Each exothermic collision between metastables steps up the lon energy by 6.5 év - - - on the average. Hence, many collisions would be required to "pump" the ions to the observed ion temperature of 5 x 106 °K (~ 500 ev) in a 16-ft. long arc. It 18 be- lieved that the initial lon energy necessary to overcome the Coulomb barrier to the exothermic reaction arises from the precessional motion (due to E X H rotational drift about the arc axis) plus the axial electrical field. The ions attain an energy of ten to twenty volts from these electric fields; hence, a few near head-on collisions can overcome the potential barrier (~ 40 ev) and thus ignite the exci- tation heating mode. The lons then escalate to a high energy Li keeping with the rate of electronic excitation of metastable states, the exothermic reaction rate, the power drain to very cold electrons, and the mean lifetime of ions in the arc. the *Research sponsored by/U. S. Atomic Energy Commission under contract with the Union Carbide Corporation. RELEASED FOR ANNOUNCEMENT IN NUCLEAR SCIENCE A2:1.2ACTS The conventional cooling rate for lons of charge Z and of velocity, v colliding with slower electrons of density, 1n., is - her em 2** MAX MIN1 where the maximum impact parameter, Dmax, 18 usually taken as the Debye length and Domen is usually taken as This conventioiial cooling rate ( - 100 ev/ usec for c2") is however in error in the electron-ion case inasmuch as it does not take into accout the reduced collision time of close collisions due to the Coulomb acceleration. This 18 an especially trportant correction in the case of sluggish ions such as carbon, even at 500 ev. The corrected expression for the rate of cooling of the average ion is approximately .. . .. - - . - . . 4 . 2 - - SCH e * {2149)2 = the other sites interesante nie! Dit is * SKY...not on wher i th w Wey o me *. If we take as parameters for the carbon arc:. 2 = 2, v.in 2ze2 . 107cm/con, n. = 1024cm3, 11. = 5 x 1010 cm3 (for a Maxwellian distribution at T. = 42,000 °K) the cooling rate 18 only - 12 ev/usec for Day By but only - 6 ev/msec for bm A Druyvesteyn distribution of electron energies would .. reduce dw/at still further (to about – 4 ev/usec). The actual energy transfer .. rate is thus an extremely sensitive parameter. Choosing dw/at as - 6 ev/usec, we then require that the rate of collisional heating of an ion via the exothermic reactions be: s n* Ocol2 (13) > 6 x 108 or OC11 V> 1.5 x 108 cm®/sec for n* ~ 3 x 1023 cm*3 Thus, for vino ~ 1.5 X 107 cm/sec, coy > 10*15 cme, a fairly reasonable figure. * In addition, we require that the rate of excitation of an ion to a metastable state be: en Oexe V-(6.5)> 6 x 108 or osv. > 3 X 10^8 cm®/sec, where f is the frection of electrons above 6.5 ev( 1 ~ 0.3). The heating of ions by Coulomb collisions with faster electrons 18 about +0.5 ev/Msec. The mean lifetime of lons in a 16 ft. long arc 18 estimated at about 0.5 msec based on both Doppler drift of lons (~ 106cm/cond and particle conservation suggesting a net heating rate of ions of about 1 ev/usec. Although accurate evaluations or the individual processes are not possible at this time, the quite reasonable requirements on the pertinent physical parameters suggest the feasibility of this new mode.of heating lons by inelastic collision processes. It should be noted that the lithium arc does not exhibit such energetic Lons in agreement with expectations of the model. 1. J. R. McNally, Jr. and M. R. Skidmore, Applied Optics 3, 699(1963). 2. J. R. McNally; Jr. and M. R. Skidmore, ORNL-3652(1964) (available from ; author). 3. An auxiliary reaction 18 ca**(3p) + c2** (3p0)→ C2+(2p0). + cat(18) + 0.3 ev, which may enhance or quench the pumping cycle depending on whether or not a radiationless transition ( 4p0.-) occurs during the collision. 4. The reaction cat + Ar → 0* + Ar* +8.6 év has a broad maximm of 2 x 10-15 cm at about 1.kev. (see J. B. Hasted and R. A. Smith, Proc. Roy Soc. (London) A235, 354 (1955)]. bo'!... P : IK END - - - - - . Com . DATE FILMED 2 / 11 / 66 .... ... .. .. ..