the PM . asmenn. nr . medicinos que r.one......... . . . 2 TOFI P 3 ORNL UNCLASSIFIED C Lichte Peidet die harte , internet : weten mitindo w mtandao on viimeksi this in die meistesini registre 1 - LEGAL NOTICE This report was prepared as an account of Government sponsored work. Neither the United States, nor the Commission, nor any person acting on behalf of the Commission: A. Makes any warranty or representa- tion, expressed or implied, with respect to - the accuracy, completeness, or usefulness of the information contained in this report, or that the use of any information, appa- ratus, method, or process disclosed in this report may not infringe privately owned : rights; or B. Assumes any liabilities with respect to the use of, or for damages resulting from the use of any information, apparatus, method, or process disclosed in this report. As used in the above, “person acting on behalf of the Commission” includes any em- ployee or contractor of the Commission, or employee of such contractor, to the extent that such employee or contractor of the Commission, or employee of such contractor prepares, disseminates, or provides access to, any information pursuant to his employ- ment or contract with the Commission, or his employment with such contractor. inwohnt were others who has worked in his - 7 this download Ni-moisas.n .c. DTIE •: *". '...,.-* I n ntelis rierole oma. . .... te mire turning i den. W e nn ich Sal mini : ?10,7.. : MICROCARD ISSUANCE DATE 8/10 1964 I- ORNG-8-3 EXPERIMENTS ON A LARGE VOLUME ELECTRON-CYCLOTRON HEATED PLASMA* CONF-564- MASTER .. . . . . . . . . . . by W. B. Ard, M, C. Becker, R. A. Dandi, H. 0. Eason, A. C. England, G. M. Haas Washington 25, D. C. Department of Commerce Office of Technical Services Available from the Microfilm Price $_ , Facsimile Price $ / 260 S Oak Ridge National Laboratory Oak Ridge, Tennessee, U.S.A. INTRODUCTION + S As part of the thermonuclear effort at Oak Ridge National Laboratory, large volume high-B plasmas have recently been produced by microwave power at the electron-cyclotron frequency. - The original work was done with 13-cm microwave radition in a magnetic mirror and was reported at Salzburg." Since that time, higher frequency and higher power radiation has been used in a succession of experiments.2-4 The current experiments involve heating . a plasme in the EPA Facility which has a 3:1 magnetic mirror ratio. Up to 50-kw CW of 3-cm microwave power are applied to the plasma in a large volume cavity. The plasma resulting from this latest experiment in the EPA Facility has an electron temperature of ~ 120 kev and a density of ~ 5 x 1011/com when heated with 25 kw of 10.6-gc radiation. LEGAL NOTICE The report mo prepared una cont of Coronat m ore wort. Metter Who word mie, nor de Canninhou, w prm acting a ll of Count: A. Met my minuty or representation, grond or iOAK, Muratpect to the icy. rxy, compass, or mae old hornton coutum ua Wo mort, or that are ne of any taborwilon, imunow, wood, or proceso de lourd la to more way net latring minimiy om or B. Are way labaides au repect to the wol, or lor was nung from the No my warunun, man , whod, or more deler ta i report. As and in the worno, "perna estona a half of the couum " mchandra may me plore or rowcwr of the Connten, or omnmo nf auch miraclor, to uke oma uut sud tarne or wir wrot L Conniola. Of oplog. a mi murns top promrm, honume, or mente bo, ury hloration now to MI ME Nomor corte! in the Coualnam, or No Imporal much I lor. * Research sponsored by the U. S. Atanic Energy Canmission under contract with the Union Carbide Corporation. DISCUSSION While this is primarily a paper describing the plasma diagnostics, some mention should be made of the probable heating mechanisms. It is well known that "free" electrons in a magnetic field have a very large cross- section for the absorption of radiation at the cyclotron frequency. This phenomena has been studied extensively in both plasmas and semiconductors. The strong coupling between an oscillating electric field and gyrating elec- trons suggest that electromagnetic energy could be efficiently converted to electron energy in a plasma. Experiments reported previously showed that even in nominiform magnetic fields only modest amounts of microwave power were needed to produce an appreciable-number of high energy electrons. The exact mechanism by which the electrons interact so strongly with the radiation field in nonuniform magnetic fields is not understood at pres- ent. However, calculations have been made which show that at least some electrons gain energy very rapidly. While these calculations are based on the orbits of single electrons, they are not inconsistent with a possible qualitative model for the process. If a fraction of the electrons produced in the plasma by ionization of background ga.8 is quickly given appreciable energy perpendicular to the magnetic field, these electrons will be contained in the field, 1f the field has a magnetic mirror configuration. The average time required for these electrons to scatter 18 increased due to their higher energy. The result 18 that a plasma will be built up from that fraction of electrons that do gain eiergy. This fraction does not bave to be large since the 1068 rate of energetic electrons out the mirrors 18 relatively low. ..... - eine '1.- : .' so. . This process for producing a hot-electron plasma 18 fundamentally dif- i .::-. ferent from such processes as magnetic compression, ohmic heating, and mag- .. . n. netic pumping. In these processes the plasma 18 raised from sane lower i ni temperature to a higher temperature. However, the heating rate of the in- dividual electrons by the microwave radletion 18 so large compared with the rate at which the plasma is built up that instead of considering the plasma to be heated by the radiation, one would consider that the plasma is built up by the accumulation of energetic electrons. To the extent to which this model is valid, it is the plasma density rather than temperature that in- creases with the application of the radiation. This approach is also dif- ferent from the process of heating plasma with the ion-cyclotron waves. In this process a wave 18 launched in the plasma which then thermalizes due to the effect of the ion-cyclotron resonance on the propagation properties of the plasma. The microwave radiation interacts directly with individual electrons. -4- THE EPA FACILITY The electron-cyclotron plasma discussed here is formed in a facility called EPA. Continuous-wave microwave power at levels up to 50 kw is sup- plied by three type SAX-418 four-cavity klystron amplifiers operating at 10.6 gc. These tubes are water-cooled, electromagnetically focused types having a saturated gain of ~ 50 db when tuned for high efficiency. Each tube requires a 20-kv, 3-smp d.c. beam supply which is remotely located. The three klystrons are driven by a camon oscillator located in the operating area. Output power 18 controlled by adjusting the drive level. An extensive system of Interlocks protecte the klystrons against damage due to waveguide arcs, cooling-system failure, magnet failure, waveguide- pressurization failure, internal arcs, and improper operating procedure. Each klystron feeds an array of two diametrically opposite pairs of waveguide ports in the cavity wall. Theee pairs of ports are spaced at 60° intervals to provide uniform Illumination of the plasma. Each pair is sym- metrically spaced in the axial direction with respect to the midplane, The total feed array 18 thus composed of twelve waveguide ports having both azimuthal and axial symmetry. The only interconnection between klystron outputs 18 mutual coupling through the cavity. This effect is quite small. since the plasma 18 a very lossy medium at electron-cyclotron resonance. Due to the relatively high cost of microwave power per watt, impedance matching, 1.e., containing the microwave power in the cavity, 18 quite im- portant from an economic viewpoint even though the klystrong and waveguide components are capable of operating with severe output mismatch. Power division and matching for the output of each klystron are accomplished by ..... three sidewall-hybrid Junctions of the quadrature type, 5 mor die Typical operating procedure for the performance of plasma experiments er in this facility ic as follows: With no microwave power applied, the cur- som rent in the magnetic mirror coils is adjusted to the proper value for electron-cyclotron resonance at 10.6 gc along the desired constant-B con- tours within the cavity. Figure 1 18 a schematic illustration of the cav- ity, microwave-feed, vacuum tank, magnetic field complex. Best operation is normally attained with field current adjusted to place the resonance zones near the entry to the mirror-coil throatr Deuterium gus is then fed into the central region of the cavity at a rate sufficient to raise the gauge pressure in this region to 2 x 10-5 torr (base pressure of vacuum system s 10-7 torr). This gas feed rate generally results in a stable "over-fed" plasma condition as indicated by measurements of plasma parameters. Microwave power is then applied, increasing very slowly from zero to the desired input level. The plasma 18 initially formed when the deuterium gas ionizes at a very low input power level. When this occurs, the pressure in the ends of the vacuum chamber promptly rises due to "plasma- pumping" through the mirrors. A corresponding increase in the gas feed rate 1.- mainteins the center pressure constant after the plasma forms. Plasma electron-noise "temperature" and stored energy are monitored continuously for stable operation as microwave power is increased and, if necessary, the gas feed rate is adjusted slightly for stability. Stable operation is ináicated by smooth variation and absence of violent fluctua- tions in the observed parameters. When the desired microwave Input power level is reached, the gas fced rate 18 reduced gradually to maximize the -6 indicated plasma parameters under stable conditions. Both parameters usu- ally peak at the same gas feed rate corresponding to a central gauge pres- sure reading of N 1 x 10-5 torr. Plasma decay measurements as a function of time are made by simply switching off the drive power to the klystron semplifiers. Once the proper gas feed rate has been determined for optimum stable operation with a given combination of mirror field and microwave power, the feed rate can be left fixed for a series of decay measurements by control of microwave power alone. It should be mentioned that the normal x-ray dose rate at the face of the tank while operating at a nominal input of 25 kW 18 of the order of 300 R/hr. However, if the gas feed rate is reduced so as to make the plasma violently unstable, this radiation level can easily rise by an order of magnitude, The vacuum tank is surrounded by a lead x-ray shield, 12 feet high end of 2-inch nominal thickness. The wall thickness 18 3 inches adja- cent to the operating area to provide additional protection for personnel involved in experiments. BREMSSTRAHLUNG MEASUREMENT Measurements have been made of the bremsstrahlung from the hot-electron plasma operated with 25 kw of power. From the theory developed by Heitler, a calculation shows that the number of photons emitted per sec per comes above the photon energy Ez 18 given by * - 3.38 x 20-45 z® 1- FE/TE (1) This equation is true only for a maxwellian distribution. In the above equation, T 18 the electron temperature, 14 the density of nuclei, Z the atomic number of the nuclei, and n. the electron density. The values of E, En, and T are expressed in kev. For Ej = T, the integral in Eq. (1) has the value 0.2194. The initial attempts to measure plasma-bremsstrahlung were complicated by collimation problems arising fron nondiscrimination against wall x-rays due to energetic electrons leaving the plasma. A satisfactory collimation geometry was finally evolved as sketched in Figure 2. In this design extra care was taken to eliminate scattering from the collimator walls and x-ray transmission through the sides of the collimator. In addition, high inten- sity q ray sources were placed in the EPA Facility to locate the volume scanned by the collimator. The bremsstrahlung was observed by a 3 1n. x 3 in. NaI(TI) crystal which was mounted 24 feet away from the axis of the machine and viewed ~ 40 cm of plasma through thin Al windows. The fractional solid angle -8. subtended by the crystal through the collimator was ~ 6.5 x 10-7. Figure 3 shows a one minute count of the plasma bremsstrahlung at 25 kw with Da gas feed at a magnetic field of 2000 amps. Also shown 18 a spectron taken after turnoff. The suza of counts made after 10 turnoffs between 1 msec and 1 sec after turnoff 18 also shown. The two spectra have the same shape. To achieve this, it was ne essary for the analyzer to look into a "black" hole on the opposite side of the cavity. Dis hole consisted of an 0.010 in. Al window at the end of a tube which in turn was surrounded by lead to shield the surface seen by the collimator from the intense soft x rays produced at the mirrors. It was previously noticed that the spectra before turnoff were distorted in the vicinity of ~ 100 kev due to the scattering of the soft x rays by the black wall. After turnoff this intense x ray source disappeared within a few millibec and this effect was not observed in the spectrum. The above mentioned "black" hole largely eliminated this effect. Comparison of this spectrum with Eq. (1) indicates a temperature of ~ 120 kev. The density as determined from the solid angle, counting rate above 120 kev, and the total plasma volume viewed by the collimator 18 4-7 x 1011 e/cm". Defining B = BrenkT/B?, where Bo 18 the undisturbed (vacuum) magnetic field on the exis at the midplane, and nkt is the energy density of the plasma, then B 0.4 for this plasma. One difficulty with the measurement is bremsstrahl.ung production from plasma ions with 2 > 1, in particular, copper which 18 sputtered off the walls. Although No CuII light 18 seen from the cavity, this cannot be ruled out as a source of error. In addition, scattering in the collimator itself 18 probably present and probably influences the spectrum shape. . -. .. .9. . crna . . . PINHOLE CAMERA X-RAY PROIOGRAPHS OF THE PLASMA IN THE EPA FACILITY . . . . . . . With the use of a pinhole camera, located with its pinhole near the EPA cavity wall, excellent photographs were taken of the ECP. The pinhole camera consisted of a thick lead box with a gold plug into which a 1/44" diameter pinhole was drilled. The distance from the EPA axis to the pinhole divided by the distance from the pinhole to the film plane, was approximately 8. This provided a measured ~ 5-cm resolving power at the axis of the machine. - - - - The x rays had to pass through two 0.009-inch aluminum windows, one on the cavity (to contain the microwaves) and one vacuum window. The photographs were taken in sets of four pictures with variable expo- sure times on Eastman Kodak Type AA film. Manufacturer's recommendations were strictly followed in the development and fixing. Using the measured sensito- metric properties of the film, and the fact that the reciprocity law 18 known to hold for x-ray film, the measured film density was converted to inci- dent x-ray intensity (energy per unit time per unit area). - - Figure 4 shows a contour plot of the measured relative intensity of x rays , - on a typical film. Superimposi i on the plot 18 a scale on a plane passing through the axis of the EPA machine normal to the the camera axis. From this picture the radius to the half-maximo contour appears to be ~23 cm and the total axial length (to the half-maximim points) is estimated to be ~ 30 cm. The axial extent cannot be measured because of interfering film exposure due to x rays eminating from the cavity wall where it "necks down" at the mirror throat. These dimensions give a flaima volume of ~ 50 liters for a microwave power input of 25 kw and a magnetic field coil current of 2000 amps. -10- The spectrai intensity of bremsntrahlung radiation from the plasma 18 given by 2² an I(E) - (Const) e-B/T sec-dcm-> (2) The intensity of x radiation in the 1/4" dianeter area at the center of the film can be calculated from the solid angle, the volume of p.lasma visible to this area through the pinhole, the measured plasma density of 5 x 1011/com>, and plasma "temperature" of ~ 120 kev. This intensity 18 proportional to za n. _ Tê. Computing also the effect of the two 0.009-inch Al windows, the amount of energy is sufficient to cause the observed film density, when com- pared to the measured spectral sensitivities of similar x-ray films. However, as determined in reference 9, the film is most sensitive to low energy photons, e.g., the film absorbs most energy near the absorption edges of silver (~ 25.5 kev) and Bramine (13.5 kev) so that the film image 18 not represertative of the higher energy photons characteristic of the 120-kev "temperature" as measured by HaI(Tl) crystal-phototube combination." The observed x-ray intensity incident on the film has been transformed back into a radiance coefficient (energy per unit time, unit volume and unit solid angle) in the plasma by an inversion procedure resembling the Abel in- version schemes used in certain optical problems. 10 The results have shown that while the plasma shape as determined by this technique 18 changed some- what from the apparent film image shape, the radius determined by this tech- nique 18 changed by less than 20% from that measured from the film, 1.e., the picture ca the film can be used to determine the approximate dimensions and -11- hence an inversion procedure 18 not required to determine the plasma volume in this case. Inasmuch as the x-ray photographs may not give a representa- tive picture of the 120-kev plasma, for the reasons mentioned above, it 18 felt that this technique has reached the logical limit of its precision for us. Also, since the calculated radiance coefficient is proportional to z2 n n. 17, it is not clear that the radiance coefficient 18 truly representative of the radial electron density distribution. The radiance coefficient radius determined by this technique 18 in approx- imate agreeinent with the plasma radius measured by other techniques, e.g., Ball probe and do/at measurements. The plasma volume measured here is also in ap- proximate agreement with that calculated from a correlation of the bremsstrah- lung measurement and the dv/dt measurement. -12- CONCLUSION The two experiments described above 1llustrate the kinds of diagnostics that have been used to determine the density and "temperature" (or meun energy) of the electron-cyclotron plasmas produced at Oak Ridge". Correlation of these measurements with other measurements, not described aere, have resulted in a consistent picture of the main features of the plasma. The icon temperature is still undetermined. Efforts are bcing made to Pind methods of both measuring it as well as increasing it. The diagnostics described in this paper have demonstrated the reality of a high-B large-volume plasma which is stable in a simple .magnetic mirror. . Further efforts are being made to determine the theoretical aspects of the heating mechanism, the observed gross stability, the energy balance, and the conservation of particles. -13- REFERENCES AND FOOTNOTES lear Fusio I (1) M. C. Beclier, et al, Nuclear Fusion: 1962 Suplement, Part I, 345, (1962). (2) R. J. Kerr, Muciear Fusion, 3, p 197 (1963). (3) W. B. Ard, et al, Phys. Rev. Letters, 10, p 87 (1963). W. B. Ard, et al, 6th International Conference on Ionization Phenomena in Gases (In Press). (5) 1. J. Riblet, Proc. Inst. Redio Engrs., Feb 1952, pp 180-184. (6) W. Heitler, The Quantum Theory of Radiation, The Clarendon Press, Oxford, England, 3rd Ed., p 252 (1960). (7) Radiography in Modern Industry, Supplement #2, Eastman Kodak Company, X-Ray Division (1959). (8) The reciprocity law 18 an example of a general photochemical law that states that the same effect 18 produced for It = constant where I 18 the intensity of radiation and t 18 the time of exposure. (9) 1. E. Seemann, Rev. Sci. Inst., 21, p 314 (1950). (10) One of many references 18: Kiell Bockasten, J. Opt. Sci. Am., 51, P 943 (1961). . -14- FIGURES (1) EPA Microwave Heating Experiment (Schematic) (2) Schematic Drawing of EPA X-Ray Collinator (3) Bremsstrahlung Spectrum From EPA (4) Contour Plot of Relative Intensity on X-Ray Pinhole Photograph ORNL-DWG-63-7563 ORNL-DWG-64-3840 ORNI-DWG-63-7283 ORNL-DWG-64-315 woud ORUL DWG 63-7563 CAVITY VACUUM TANK WAVEGUIDE INPUTS CONSTANT \ 50 KW CW-10.6 GC B CONTOUR = D GAS INLET D. PLASMA PUMPING TUBES onli n e. - PLASMA TARGETS MAGNETIC FLUX LINES TO VACUUM PUMPS VACUUM BARRIER VACUUM PUMPS TO VACUUM PUMPS MIRROR COIL E-PA MICROWAVE HEATING EXPERIMENT (SCHEMATIC) Figure 1, ORNL-DWG-63-7563 - - - - - . . 3" ID ALUMINUM WINDOW (.040 THK.) - MICROWAVE CAVITY ALUMINUM WINDOW (.OO THK.) T3" LEAD 1/4" HOLE PHOTOMULTIPLIER TUBE- - - A -2" LEAD | CVA" HOLE 41-0" LLEAD 4%2* ID x 9" LONG 0-4" L 19'08* HV SUPPLY AMPLIFIER MULTICHANNEL ANALYZER SCHEMATIC DRAWING OF EPA X-RAY COLLIMATOR Figure 2, ORNI-DWG-64-3840 with PULSUS CHUIDU Vususiya UVWsch mimi n a mom "um is ur.: Vinde CO minyum ... 1 MINUTE COUNT dokumentu SOKAKUUHUONA per muamm m mm 1150 mg . . UUUUUUUUUUUwJUOtiuuuuwi susu COUNT AFTER TURNOFF DELAY 1 MSEC. AFTER TURNOFF GATE OFF 1000 MSEC. AFTER TURNOFF 10 TURNOFFS 25 KW MICROWAVE POWER 2000 AMPS FIELD CURRENT DEUTERIUM PLASMA 1 MEV FULL SCALE Figure 3, ORNL-DWG-63-7283 0.95 . . . -. .- . .- RADIAL DISPLACEME 0.95 20 10 0 10 20 AXIAL DISPLACEMENT (CM) X-RAY PIN HOLE CAMERA PHOTOGRAPH CON TOUR PLOT OF RELATIVE INTENSITY DEC. 17, 1963 25 KW 2000 AMPS Figure 4, ORNI-DWG-54-315 END