.id . I OFI ORNLP · . · - prido '. . . 1337 32841 () 1.1:25 114 11.6 MICROCOPY RESOLUTION TEST CHART NATIONAL BUREAU OF STANDARDS - 1963 ORNU-P-1339 CONF-650935-15 MASTER *WWUN 2 4 1965 . modele trou de voord work. Metohet dhe Outted - LEGAL NOTICE - ol de Covadon" mchudnu of the lalondon contained in die report, or that dhe me Batue, nor the contington, nor pornon acting a ball of the Counterton: A. Makes may warranty or ropucatadan, apred os lapllad, wd ropec: . dhe ucco odhod, or process declaudus die report may not letrare sployee of ma contractor, to do adot chat The moport mumo rod una seconx al Gover may labilities ma roapect to do wec ol, or for denne un ol ay baloration, appendu, odhod, or procu diacloud ua Wo report suck saploma or contractor of the Communalow, or employee of muda contractor meparsa davominatoa, or pre des access to, may taboration purwabwa amployee or contract mbova, "person to b rucy, completeneum, or water ol ay talan udan, apparatu, wo do Coalustoa, or ho aplogut with such contacter. ployee or contractor of the coaminotou, or printly owned nota; ar 8. Anna Ao wewe wa SEVERE MICROINSTABILITY-DRIVEN LOSSES IN AN ENERGETIC PLASMA* J. L. Dunlap, H. Postma, G. R. Haste, and L. H. Reber* Oak Ridge National Laboratory Oak Ridge, Tennessee, U.S.A. 1. Introduction , the DCX-1, 1 cailed descriptions of the basic experimental apparatus/have appeared earlier [1, 2, 3]. Briefly, a steady state plasma with a volume of 2-10 liters and a very energetic proton component is formed by dissociation of part of a molecular hydrogen ion beam (usually 600 keV H') that makes a single pass through a 2:1 magnetic mirror trap. The standard molecular ion trajectory (Figure 1) is confined to the median plane with beam turnaround (point of closest approach to the magnetic axis) at the equilibrium orbit radius for protons of initial energy. The usual central field value is 10 kG, which gives an equilibrium orbit radius of 3.25 in. for 300 keV protons. The radius of the fast proton plasma is restricted to 8-9 in., and the distance between mirrors is 304 in. Differential pumping and the liberal use of titanium evaporated onto liquid nitrogen cooled surfaces provide base operating pres- sures of 1-2 X 10^9 torr (ion gauge, uncorrected) and charge-exchange lifetimes for low density plasmas (1-2 x 107 cm 3) as long as 2 min. Both gas dissociation and Lorentz dissociation have been employed in .. these experiments. Trapping by both of these mechanisms is distributed along the molecular ion trajectory and protons are introduced with a spread in canon- ical angular momentum. The distribution of initially trapped protons extends out to the location of the radial limiter (8-9 in.). It is therefore quite unlike the simple ring characteristic of earlier experiments (1) in which an are discharge was used for dissociation. PATENT CLEARANCE OBTAINED. RELEASE TO Research sponsored by the U. S. Atomic Energy Commission THE PUBLIC IS APPROVED. PROCEDURES ARE ON FILE IN THE RECEIVING SECTION under contract with the Union Carbide Corporation. Lorenüz dissociation at low pressures has given poter cially accessible fast proton densities (those expected in the absence of instability-driven losses and scattering to larger volume) as large as 2 x 10 cm, more than an order of magnitude gicater than those available with gas dissociation. For this reason, Lorentz dissociation has been emphasized in the recent experiments. The presence of microinstabilities in this plasma has long been recog- nized. Radio frequency signals at w > war and dispersion of the proton energy distributions for plasmas established by gas dissociation have been reported previously (23), as have the correlations of these rl signals with plasma poten- tial fluctuations and ejections of electrons through the mirrors [4, 5]. How- ever, it was not until the apparatus was altered to utilize Lorentz dissocia- tion of H** (Ha generated from Hy in a water vapor cell) [6] that the density limitation set by microinstability losses became apparent [7]. Since then, these losses have been the principal subject of our investigations. A detailed report on microinstability spectra and losses of plasmas established by gas dissociation of 600 keV Ha and by Lorentz dissociation of 600 keV H** has been submitted for publication elsewhere (8). This material, which emphasizes the standard injection trajectory, is summarized in Section 2. Section 3 extends the observations to include plasmas established by Lorentz dissociation of Hi and of Ha extracted directly from the duo-plasmatron source. Section 4 describes attempts to stabilize against instability losses by increasing the disorder of trapped proton orbits. 2. Microinstability Spectra and Losses with Gas Dissociation and With ** Lorentz Dissociation of HD In this section we review earlier work, to be published in detail else- where [8], in which emphasis was given to operation in the standard field geometry with the standard molecular beam trajectory. The descriptions of the microinstability spectra apply to the later work as well, with but few excep- tions which will be noted. The experimental techniques described in connection with the loss measurements were also employed in the later work. . me: .. i ... 2.1 Microinstability Spectra ie. . asa.. irris...: Since instability losses are observed, we first want to make it clear that flute instabilities are not present and thus that flutes cannot be blamed for the ion losses. At pressures < 1 x 108 torr there is a low frequency disturbance, 3 to 200 kc/sec., most prominent on electrostatic probes and also tie , ;..;.'İking cbserved on "ow plasma collectors located outside the mirrors. The disturb- ance occurs in the absence of instability losses, and the level of these losses varies in response to microinstabilities through the level of low frequency disturbance remains about constant. The low frequency disturbance is probably a cold plasma phenomenon, for an interpretation as flutes involving hot ion is ruled out by observations given elsewhere [8]. The plasma is, however, afflicted by at least two microinstabilities. One, which we have termed the azimuthal instability and will shortly subdivide into two distinct modes, produces radio frequency signals in the neighborhood of the proton cyclotron harmonics and dominates the fluctuations of B, and or electrostatic potential (considering only the high frequency fluctuations of o). It is this instability that drives proton losses. The other, which we have termed the axial instability, dominates fluctuations of B, causes rapid ejec- tions of electrons and corresponding fluctuations of the plasma "dc" potential, and has a more complex frequency spectrum (though w still > w.). There are two modes of azimuthal instability. One, which for con- venience we term the gyrofrequency mode, has been long recognized. It has the lower density threshold (as low as 3 x 105 energetic protons cm 3; see Section 4.1). It probably involves mostly protons on near-equilibrium orbits, since the lowest radio-frequency signals generated are about 14.6 Mc/sec, the gyrofrequency on equilibrium orbit. The second mode, which we term the dimin- ished frequency mode, generates rf signals which are 1-2 Mc/sec below the gyrofrequency mode signals. Reducing the radius of the plasma has no effect on the lower frequency limit of the gyrofrequency-mode signals, but increases the limit for the diminished frequency mode. Protons involved in this mode are therefore sampling the magnetic field at the periphery of the plasma. ** ** The diminished frequency mode is rarely observed with gas dissociation but is a usual feature of plasmas established by Lorentz dissociation of H2 . The threshold is associated with a critical central proton density of 5-10 x 107 cm , a value which by variation of Ha current has occurred for Tox in the range 2 5 TDK S 30 sec. Some plasma properties associated with long Tok values seem to be required for this mode, for with gas dissociation at higher pressures, densities as high as 1-2 x 108 cm3 are achieved and the mode is not present. We have been unable to make unique mode assignments for these instabili- ties. Of the theoretically recognized instabilities, the possibilities are the drift-cyclotron instability with kyl to 19, 10), the negative mass insta- bility (11), and the maser (12) and Harris (13) instability with ki, to. es At the normal operating densities (107 - 108 cm 3) It appears that all of these are allowed. However the typical resonance condition at threshold for the lon- electron interactions (14, 15, 16), Woe = Ways requires n = 2.7 x 108 cm*3. Since the hot ion density at threshold is as low as 3 x 205 cm3, we are in- clined to attribute the threshold to the negative mass instabil.ity (the thresh- old for it has been calculated only for beams but is thought to be low) or the åriit-cyclotron instability (the electron temperature 18 iuw, and for cold electrons the growth rate of the drift-cyclotron instability does not vanish wien wpe < was but only decreases rapidly).. 2.2 Microinstability Losses The losses are evidenced by the failure of charge-exchange accountability and by direct observation of losses correlated with rf signals from the micro- instabilities. The charge-exchange measurements were made with a longitudinal array of fast atom detectors (foil-covered Faraday cups and/or secondary emission detec- tors) mounted outside the periphery of the plasma. The 2, e integral of the steady state loss current provides the total charge-exchange loss rate. The time (Tox) for this current to decay to i/e of the steady state value is the hot ion lifetime for loss by charge exchange (17, 18]. The integral of the decaying current over 2, 0, and t gives the total number of hot ions trapped in steady state (17). The axial distribution of the 0, t integrals gives a measure of the axial extent of the plasma and therefore a measure of instability-driven axial expansion. The central fast proton density is calculated using the value of the e, t integral of the current decay registered by the fast atom detector centered on the median plane. All density figures quoted in this paper are for this central density. With Lorentz dissociation of HD, the failure of charge-exchange ac- countability is first observed at densities of 5–10 x 107 cm , a value which by variation of injected current has occurred for Tox in the range 2 S TOKS 30 sec. Under the worst conditions (maximum injected current and longest Tok) instability losses account for 90 percent of the proton input current (19). The total number of stored protons has a definite limit of 1-2 x 1012, as magnitude much as an oråer/ less than the potentially accessible number. The central density saturates at 1-2 x 108 cm3, more than an order of magnitude less than the potentially accessible density because in addition to the losses there is significant dilution by instability-driven axial expansion. ** · Direct losses of hot ions occur both radially and axially, with radial losses dominating by about an order of magnitude. The threshold for observation of these losses is the same as that for departure from charge-exchange account- ability, 5-10 x 107 cm3. Radial losses were detected by probes (a foil-covered cup and/or a gridded probe) acting as radial limiters in the median plane. The radial signals are due to precessing particles (8), and therefore must be due to fact protons and not to charge-exchange neutrals. Figure 2 clearly shows the correlation of radial losses with rf signals from the microinstabilities. Eoth modes of the azimuthal instability drive radial losses. The loss currents that accompany une dizinished frequency mode are considerably larger but the mode does not occur often, and on a time-averaged basis the losses driven by the cyrofrequency mode are larger by about an order of magnitude. Axial losses were detected by a foi.l-covered cup located outside one of the mirrors, inclined at an angle with respect to the median plane. Axial losses are dominated by the diminished frequency mode, and, as indicated earlier, are minor in comparison with radial losses. With gas dissociation at low pressures, instability-driven scattering and the lower proton injection current result in maximum densities less than the threshold for observation of losses with Lorentz dissociation, and insta- bility losses are negligible. At high pressures (TDK 3 0.1 sec.) the scattering is much reduced and the maximum densities exceed this threshold, but instabil- ity losses are not present. 3. Trapping by Lorentz Dissociation of H; and of Directly Extracted H We have recently observed proton trapping by Lorentz dissociation of HS from a duo-plasnutron ion source and of Ha extracted directly from such a source. The general features of the microinstability spectra observed with these plasmas are the same as those described in Section 2.1. The limits imposed by instability losses are essentially those described in Section 2.2. 3.1 Lorentz Dissociation of HS Two experiments used the Ho output from an ion source optimized for Ha production. In one (20), the central field value was the usual 10 kg, and HD was injected at 400 keV in order to obtain the same beam trajectory as is normally used with 600 keV H injection (beam turnaround at 3.25 in.). In the other, a central field value of 12.3 kg was used to allow injection of 600 keV H3 on a trajectory with beam turnaround near the equilibrium orbit radius for 200 keV H'. The Lorentz trapping fractions in these experiments were 1-2 x 10°5. In the first experiment the gyrofrequency mode was consider- ably suppressed relative to the diminished frequency mode, and the instability lcsses were dominated by the latter. Some suppression of the syrofrequency mode is usually coserved in experiments that increase plasma disorder, as in the first of these experiments and in the experiments described in Section 4. The experiment cited, however, has been the only one in which the suppression was effective cough to allow the diminished frequency mode to dominate. A similar 600 keV injection experiment was performed using the Hg output from a different source, one optimized for H3 production. This source was not favcrabie for the production of excited vibrational states. The Lorentz trap- pirc fraction was not greater than 1? x 108 and the plasmas were not of high enough density to be instability limited. : 3.2 Lorentz Dissociation of Directly Extracted HA Our earlier measurements of the Lorentz trapping fraction for 600 keV Ha wtracted directly from the duo-plasmatron ion source fixed the upper limit of this fraction at 1 x 10°7, quite negligible in comparison with the 1-2 x 10°4 obtained with Ha* injection [6]. The source was recently changed to one per- mitting larger He currents, and much higher Lorentz trapping fractions, as grcat as 1 x 20°5, are now obtained. The "proton input rate is high enough to produce instability-limited plasmas. The trapping fraction depends somewhat on the conditions of source operation. Happily, those for higher extracted He current (essentially higher arc and source magnet currents) result in larger Lorentz trapping fractions. For example, 1 mA H2 current yields a trapping fraction of 7 X 10°6, and 10 mA H* yields 1 x 10°5. The details of source geometry, and oi' operating potentials and currents were cataloged in an effort to isolate those features which produced the changes in Lorentz trapping fraction, for both Ha and HS (Section 3.1) in- jection. Unfortunately, the catalogue is lengthy and most items vary from source to source, so it does not answer this question. ........... . .. 4. Effects of Deliberately Introduced Disorder ... As reported elsewhere [8], in earlier work the only definite indications of stabilizing effects on the azimuthal instability were associated with in- creasing disorder of proton orbits. The effects included control of the onset of instabilities by deliberate manipulation of the molecular beam trajectory (during both buildup and steady state), spontaneous suppression of instability during steady state operation at low pressures (probe sly caused by instability- driven scattering), and rapid disappearance of instability once injection ceases (again probably the result ci instability scattering). Since the azimuthal instability is responsible for most of the 1::stability, losses, several t. !ques for deliberately introducing disorder were explored in hopes of relwing tioc lensity limitations imposed by these 1088es. The dci!ls are given in the following paragraphs. 1.1 Disorderin by Additional Precession Operation with the molecular beam in the median plane but shirted from the standard trajectory (turnaround tangent to the equilibrium orbit radius for protons) resuits in increased precessional motion of initially trapped protons. Several experiments have shown that the added processional motion produces a higher instability threshold and relaxes the instability limitations on maximum density. The threshold was determined by evaluating the central density at which rf signals at was first appeared during the plasma buildup interval following injection of the Ha beam, and by evaluating the steady state central density barely sufficient to give these signals. The threshclds for nese two methods were essentially equai and were reproducible over the variations of pressure and of Ha current. The range of trajectories permitted by the steering Plagnet, which controls the beam entrance angle, was extended by varying the strength of the contain- ment field. The results of one run are shown in Figure 3. The minimum thresh- old, 5 x 105 cm3, was obtained with the standerd trajectory, and progressive variation from this trajectory resulted ir. progressively higher thresholds. In this run, the threshold increased a factor of 40 with beam turnaround at the magnetic axis. Other runs, presumably with a molecular beam of poorer focus, resulted in a minimum threshold of 3 x 105 cm and a threshold of 1 x 108 cm3 with beam turnaround at the magnetic axis. Despite the magnitude of the threshold change, increasing the proces- sional disorder when operating at densities above the larger threshoid has only a small effect on the average level of the steady state rf signals. In shifting from standard injection to through-axis injection, this level is reduced by only about 50 percent. The technique of trajectory variation is also only moderately successful in suppressing instability-driven losses. The degree of suppression gradually increases with displacement of the trajectory from the standard one. Figure 4 shows charge-exchange data obtained while injecting 10 mA of directly ex- tracted Ha. The data are for the two extreme trajectories, beam turnaround aü trc cuil. .0r7" u8 and at tre tilgnetic axis. Injection through C%10 residik.. :pping rates or both gas and Lorentz d!88cc.ation, Walch accounts :or the displacement of the curves in the region o: charge- &%C2.accountabiat::. In the region dominated by instahillty losses, the aivat CC of *6.6 Sp.ccc trajectory becomes clear. With this trajectory the 16::. densas about 2 x 108 cm , twice that obtained with the standard ürüfeciory. 21.1ting density 18 ru-axed by a similar amount when in- jectbrein. The aivantage of tre displaced trajectory 18 largely due to roducci croc ora!00 Wisch proclces a decay time that 18 longer at a given pressure. 1.2 Priven iveal Oscillations some experiments have involved driving axial. oscillations by externally isposed , ficids near the oscillailon frequency for small amplitude axial displacerent of prctons from the equilibrium oroit. Charge-exchange currents experiments. Croit studies have suggested an explanation. When the perpendicuies cncroy is kept co..stant and oscillation amplitude 18 increased by increase of parallci enerw, as is the case with the externally applied drive, in this macnetic rield the axial oscillation frequency varies out little with ampli- tude. "ne oscillations apparently stay nearly resonant with the driving field and, since the energy gained per cycle increases with oscillation amplitude, the protons are rapidly driven into the driving plates, which were at 2 - 4 53 in. There were reductions că rf sigral amplitude, but these were surely at least partially the result of reduced density. de.3 Axial Oscillations with Additional coils A brief series of experiments made use of an additional coil pair mounted inside the main mirror coils. The additional pair was energized in a cusp fashion to produce radial field components that generated axial forces. Thouch again there were reductions of rf signal amplitude, the charge-exchange current from plasmas dominated by Instability 1088es could only be decreased by energizing the added coils. The proton orbits in this field are too complex to permit a complete description, but it is believed that the proton oscillation amplitudes were increased to the point that there were significant losses by collisions onto the/coils, which constituted an obstruction at r - 5 in. aná 2 = 2 7/8 in. added -. ~ ~ ~ 5. che poucos established in these experiments is severely mited by mécrcingtab11ity-driven losses of last protons. There are indications of +6011!2ing e:Sects ascociatec !th disordered proton orbits. Taree tech- ..:: For increasing this disorder ware expiored. Only one, increasing the preceszécnul 10. by c..Og the 1.jection trajectory, was beneficial. Inis tenrique raises tác density thres.cold for instability by large l'i rs (1:0 to 200), si relaxes the density 14itatłon by only a factor of 2. 6. B owledom.e?.ts de particularly wish to express our appreciation to A. H. Snell ior his corti..ng support; to the other members of the XX- group, R. S. Eduards, L. is. wengili, 2. G. Reinharüi, W. J. Schill, and E. R. Wells, for their contri- butions to this work; to T. 2. Fowler and C. 2. lielsen for numerous Wformative discus810..6; to P. 3. Beli who had developed some of the diagnostic techniques asa c. 2. Parker for computer studies renated to this work. . - a REFERENCES a. * Research, sponsored by the U. S. Atomic Energy Commission under contract with the Union Carbide Corporation. + 1 [2] 13: [4] (5) Deceased. ELTETT, C. F., BELL, P. R., LUCE, J. S., SHIPLEY, E. D., and SDON, A., Proc. Cicond U. N. Conf. on Peaceful Uses of Atomic Energy, 31 (1958) 298/307. BAILETA', . F., DUNILAP, J. L., EDWARDS, R. S., HASTE, G. R., RAY, J. A., REINHARDT, R. G., SCHILL, W. J., WARNER, F. M., and WELLS, E. R., vuclear Fusion 1 (1.961) 264/272. DUNLAP, J. L., BARNETT, C. F., DANDL, R. A. and POSTMA, H., Nuclear Fusion: 1962 Supplement, Part 1 (1962) 233/237. FASTE, G. R., and BARNETT, C. F., J. Appl. Phys. 4 (1962) 1397/1399. DUNLAP, J. L., HASTE, G. R., POSTMA, H., and WELLS, E. R., Bull. Am. Phys. Soc. 8-2 (1963) 171. POSTMA, H., HASTE, G. R., and DUNLAP, J. L., Nuclear Fusion 3 (1963) 128/129. Our first evidence for instability losses, the failure of charge- exchange accountability with H** injection at low pressures, was described at the Conference on Mirror Configurations, Fontenary-aux- Roses, France, July, 1963. DUNLAP, J. L., "Plasma Properties o Insta- bilities in DCX-1". 16] [7] [8] DUNLAP, J. L., HASTE, G. R., NIELSEN, C. E., POSTMA, H., and REBER, L. H., "Microinstability Limitations of the DCX-1 Energetic Plasma", submitted to Physics of Fluids. A brief summary of the nature and magnitude of the instability losses has also been prepared; DUNLAP, J. L., HASTE, G. R., POSTMA, H., and REBER, L. H., "Energetic Plasma Losses Due to Micro- instabilities", submitted to Physical Review Letters. (9) MIKHATLOVSKII, A. B., and TIMOFEEV, A. V., Soviet Phys. JETP 17 (1963) 626/627. (10) BURT, P., and HARRIS, E. G., Phys. Fluids 4 (1961) 1412/1416; HARRIS, E. G., Culham Laboratory Report CLM-R32 (1963). (11) KOLOMENSKII, A. A., and LEBEDEV, A. N., Atomnya Energiya I (1959) 549; NIELSEIN, C. E., SESSLER, A. M., and SYMON, K. R., Int. Conf. on High Energy Accelerators and Instrumentation, CERN, Geneva (2959) 239/252. Et (1.2) ROSEIBIUPA, M. N., and POST, R. F., Phys. Fluids 8 (1965) 547/550. [13] HARRIS, E. G., J. Nucl Energy C-2 (1961) 138/145. (14) GUEST, G. E., ar.d DORY, R. A., Thermonuclear Div. Semiann. Progr. Rept. Oct. 31, 1964, ORNL-3760, 70/72. (ORNL reports are available from Office of Technical Services, U. S. Department of Commerce, Washington 25, D. C.) (15) HALL, L. S., HECKROTTE, W., and KAMMASH, T., UCRL-7677 (to be published). (26) SCPER, G. K., and HARRIS, E. G. (to be published, Phys. Fluids). These interpretations of the charge-exchange measurements require that the instability losses be insignificant during the decay interval. Fortunately this condition 18 met, for once injection ceases, this plasma rapidly becomes stable and essentially all stored fast protons are lost by charge exchange. [18] Dispersion of the trapped proton energy distribution makes TDK only approximately equal to the mean lifetime, but for this plasma the approximation is an accurate one. (19) Explanations for the failure of charge-exchange accountability not 107] an axial expansion severe enough to result in significant charge-exchange losses beyond the axial extent of the fast atom detectors, or attenuation of charge-exchange currents due to the energy cut-off of the foil-covered detectors) were excluded by ancillary experiments that are described in Reference (8). [20] HASTE, G. R., DUNLAP, J. L., POSTMA, H., and REBER, L. H., to be published Nuclear Fusion 5-2 (1965). 11 FIGURE CAPTIONS 1. (ORNL-LR-DWG 65-1505R) Midplane section of the DCX-1 apparatus showing the standard He bean trajectory and location of various probe.. 2. (PHOTO 68439) Simultaneous recordings of magnetic probe signals from the azimuthal instability (ORF) and proton 1088 current to a radial detector (460 HA HA , Tox = 6 sec.). The upper 4 cm of the oscillogram 18 the time- resolved frequency spectrum of the rf signal in the neighborhood of war with marker frequencies at 12, 13, and 14 Mc / sec. The intermediate trace 18 the oscilloscope display of the rf signal. The bottom trace is the radial 1088 signal. The diminished frequency mode occurs only once (at 6 cm) and there the amplitude of the radial 1088 signal 18 greater than 3 cm. There is an almost continuous radial loss associated with the syrofrequency mode. 3. (ORNL-DWG 65-5267) Proton density threshold for Way signals as a function of molecular beam trajectory showing increase of threshold with departure from standard trajectory. For points at R > 2.5 in. the containment field was fixed at the normal value (3.25 in. proton equilibrium radius) and the trajectory was varied by the steering magnet. For points at R < 2.5 in. the trajectory was varied by changing the magnetic field. At R ~ 0, the equilibrium radius had decreased to 2.6 in. 4. (ORNL-DWG 65-5270) Charge-exchange loss current as a function of decay time for two molecular ion trajectories, corresponding to the extreme ordinate values of Figure 3. The dotted extrapolations represent the response expected without instability losses. Data obtained by injection of 10 mA of directly extracted Ha. 12 H2H BEAM ORNL-D.1Û TITANIU: 1 EVALG OR I A PROBE LINER REGION ROTATING BEAM TARGET PROTON EQUILIBRIUM ORBIT NEUTRAL PARTICLE DETECTORS PROBES PHOTO 02:39 . .. - - - .. . ... pero non . 12 MC- 13 MC - 14 MC- . -- - - . .. ORF . caring RADIAL LOSS 9 RADIAL LOSS me oyda 1 . منه: د هغه نامت نما مسنننننننننننننعنهن سے موسم سے منه بسا - مه mama samima. .. .... commenting this maintai inen.com n + 6 0.5 sec 0.5 sec ORNL-DWG 65-5267 5x10? م ة و CENTRAL FAST PROTON DENSITY AT THRESHOLD (cm-3). Ö ci tttt! 105 L 2 3 4 H* TURNAROUND RADIUS (in.) Threshold for wci Siçnals as a Function of Mole- cular Beam Trajectory. (Daia DCX-4 4-30-65, 5x10'torr Gauge Pressure, Helium Leak, B in- creased for Points of R2.5 in.) ORNL-DWG 65-5270 ... . .. ' - -...- - C- .. L.... i ob NH4N CHARGE -- CXCHANGE LOSS CURRENT (arbitr y uniis) .. - -•. SYMBOL TURNAROUND EQUILIBRIUM I RADIUS ORBIT RADIUS 3.25 in. 3.25 in. . 2.6 in. 10-2 10- 10° 10'. CHARGE-EXCHANGE DECAY TIME (sec) Charge - Exchange Loss Current as a Function of Decay Time for Two Molecuicr Ion Trajectories. (Data DCX-1 5-6-65, 600 kev ha', 10 ma Injected Beam, Helium GCS Lecói .- ' ... come unco.com no commento mai . 1 : 7. END DATE FILMED 11/ 9 /65 .