High Current Density and High Brightness H - Sources for Accelerators

1. High Current Density and High Brightness H- Sources for Accelerators Vadim Dudnikov Brookhaven Technology Group, Inc. FNAL, December 2005

2. ACKNOWLEDGMENTS I am very grateful to the ISIS Team for choosing Charge Exchange Injection and Penning SPS for ISIS operation and for successful demonstration of it’s high performance in real accelerator operation.

3. Penning SPS in the ISIS RFQ

4. Abstract • Operation Experience of Compact Surface-Plasma Sources (CSPS) under operation in different laboratories around the world, will be considered. • Features of CSPS are small volume, small gaps between electrodes, high plasma density and high emission current density and high brightness, high pulsed gas efficiency and low electron current. • In many versions of CSPS were reached very long operation time. • Features of CSPS important for long time operation will be considered.

5. Contents • Introduction. • Historical remarks. • Negative ion production in surface- plasma interaction. • Cesium catalysis. • Surface Plasma Sources- SPS. • Charge-exchange cooling. Electron suppression. • Beam extraction, formation, transportation. • Space charge neutralization. Instability damping. • SPS design. Gas pulser, cesium control, cooling. • SPS life time. SPS in accelerators. • Further development. • Summary. • Acknowledgment.

6. Horst Klein (20 ICFA Workshop summary). “The ion sources, and especially the H- sources, are still somewhat a black magic. Therefore intense theoretical and experimental work has to be performed in different labs to achieve the new requirements. In Europe the Negative Ion Source network, supported by the European Union, with its 8 partners will help to reach the goal. But also such a meeting as we have had in Femilab is very helpful and intensifies the worldwide collaboration. Concerning the different types of ion sources, I think the most promising candidates for H- are the Penning ion source and the volume source (Large Volume SPS). The ECR source may be a hope for the future”. Intuition and hand experience are important components for H- sources development.

7. H- beam brightness in different SPS ( R.Welton, SNS). Beam brightness and pulse current of operational ion sources (points) and new facility requirements (rectangles)Magnetron sources: 1-DESY, 2-BNL, 3-ANL, 4-FNAL. Multicusp RF sources: 5-DESY, 6-SSC; Penning sources: 7-RAL and 8-INR;. Multicusp surface conversion sources: 9-KEK and 10 –LANL Multicusp filament sources: 11-TRIUMF and 12-Jyvaskyla.

8. Ion Source requirements for new accelerators projects ( from R. Scrivens review) Ion Source parameters required for selected high power project. 1rms, normalized, in mm mrad

9. HUASHUNG ZHANG, ION SOURCES,Springer, 1999. p.326 • Based on the achievements of positive ion sources, H- ion sources have been developed in two ways: • 1) Negative ions are extracted from the plasma of positive ion sources. Before the 1970's, the H- current was limited to less than 5 mA. This is because in a general high temperature plasma (Te ~> 10 eV) the H- formation cross section(~10-18 cm2 ) is 3 to 4 orders less than the H- destruction cross sections (~2 to 7x10-14 cm2). • In 1962, Krohn [7] discovered that the yield of sputtered negative ions increased by one order while Cs+ ions impacted the metal target. • Unfortunately, this result was not immediately used to develop a NIS up to 1970. An H- surface plasma source (SPS) was invented by introducing cesium into the hydrogen discharge plasma at 1971. • It quickly led to increasing the H- current to several Amperes. Also the cesium sputter NISs were rapidly developed. • Since discovering, at the end of the 1970’s, that the dissociative attachment cross section of highly vibrationally excited H2-molecules in a low-temperature plasma is higher by 104-105 than the groundstate[8,9], high-intensity volume H- ion sources have been developed. • At the end of the 1980's, H- volume ion sources combined with cesium has evolved with domination of surface- plasma generation of negative ions.

10. Adsorption of alkaline metals significantly increases the secondary emission of negative ions • In 1961, by Ahmet Ayukhanov (Tashkent Electronics Institute) was observed that the adsorption of alkaline metals significantly increases the secondary emission of negative ions. A little later the investigations of this effect were presented by Krohn (Argonne Nat. Lab.). However, even with the presence of cesium on the surface the intensity of beams of negative ions obtained by the secondary emission did not exceed the microampere level. • These results became the basis of secondary-emission sputtering negative ion sources with a microampere level intensity for tandem accelerators. • A. Ayukhanov, PhD. Thesis, Secondary emission of negative ions with bombardment by alkali positive ions. 1961. • U. A. Arifov, and A. Kh. Ayukhanov, Izvestiya AN Uzbek. SSR, Ser. Fiz. Mat. Nauk. No. 6, 34 (1961). • in book U. A. Arifov, Interactions of Atomic Particles with a Solid (Nauka, Moscow, 1986). • V. E. Krohn, J. Appl. Phys. 33, 3523 (1962).

11. Budker Institute of Nuclear Physicswww.inp.nsk.su

12. History of Surface Plasma Sources Development (J.Peters, RSI, v.71, 2000) Cesium Catalysis: “Enhancement of negative ion production by admixture into discharge a substance with a low ionization potential, such as cesium”.

13. Intensity of Negative Ion Beams: 1971-discovery of Cesium Catalysis.

14. H-/D- LV SPS for Tokomac Neutral Beam Injectors ~$0A, ~1 MeV, 1000s,… ~1 Billion $

15. History of Charge Exchange Injection(Rees, ISIS , ICFA Workshop) 1. 1951 Alvarez, LBL (H-) ; 1956 Moon, Birmingham Un. (H+2) 2. 1962-66 Budker, Dimov, Dudnikov, Novosibirsk ; first achievements;discovery of e-p instability.IPM 3. 1968-70 Ron Martin, ANL ; 50 MeV injection at ZGS 4. 1972 Jim Simpson, ANL ; 50-200 MeV, 30 Hz booster 5. 1975-76 Ron Martin et al, ANL ; 6 1012 ppp 6. 1977 Rauchas et al, ANL ; IPNS 50-500 MeV, 30 Hz 7. 1978 Hojvat et al, FNAL ; 0.2-8 GeV, 15 Hz booster 8. 1982 Barton et al, BNL ; 0.2-29 GeV, AGS 9. 1984 First very high intensity rings ; PSR and ISIS 10. 1980,85,88 IHEP, KEK booster, DESY III (HERA) 11. 1985-90 EHF, AHF and KAON design studies. SSC 12. 1992 AGS 1.2 GeV booster injector 13. 1990's ESS, JHF and SNS 4-5 MW sources

16. INP Novosibirsk, 1965, bunched beam Other INP PSR 1967: coasting beam instability suppressed by increasing beam current; fast accumulation of secondary plasma is essential for stabilization; 1.8x1012 in 6 m first observation of an e- driven instability? coherent betatron oscillations & beam loss with bunched proton beam; threshold~1-1.5x1010, circumference 2.5 m, stabilized by feedback (G. Budker, G. Dimov, V. Dudnikov, 1965). F. Zimmermann V. Dudnikov, PAC2001, PAC2005

17. Cs PATENTV. Dudnikov, The Method for Negative Ion Production, SU Author Certificate, C1.H013/04, No. 411542, Application filed at 10 March, 1972, granted 21 Sept,1973, published Bul. No 2, 15 Jan.1974. “Enhancement of negative ion production by admixture into discharge a substance with a low ionization potential, such as cesium”.

18. SPS was developed in cooperation of BDD, G.Dimov, V.Dudnikov,and Yu.Belchenko

19. SPS for Accelerators was developed in cooperation with G. Derevyankin

20. History of Volume Sources Development (J.Peters, RSI, v.71, 2000) Blue frame is separate Surface Plasma dominated H- formation Development of Volume Sources is finished by conversion into Large Volume SPS.

21. Marthe BacalFourth IAEA Technical Meeting on “Negative Ion Based Neutral Beam Injectors”9 May 2005 • “What ion source for volume production ?? • New ion sources were proposed for making use of the volume production mechanism. The magnetic multipole, used in 1976 in our first experiments (Nicolopoulou et al, J. Phys. 1977) was modified by the addition of a magnetic filter. This seemed to solve the problem of H- destruction by fast electrons, since they were eliminated from the extraction chamber. • However, this solution was only partial, for two reasons : • * the negative ions may not be formed in the extraction chamber, but in the driver, near the filaments ; • * the magnetic multipole is very efficient to dissociate molecules, but H atoms destroy H- and H2(v) ! • When cesium was introduced in the magnetically filtered multipole , it appeared as a suitable source for producing atoms and positive ions for surface production. Obviously, this device is not suitable for volume production !! It is really a good Large Volume Surface Plasma Source, not a Volume Source.”

22. General Diagram of the Surface-Plasma Mechanism for Production of Negative Ions in a Gas Discharge Surface plasma generation of H- on anode often is a dominant process of H- formation in discharges without Cs, as well with Cs

23. Schematic Diagrams of Surface Plasma Sources (a) planotron (magnetron) flat cathode (b) planotron geometrical focusing (cylindrical and spherical) (c) Penning discharge SPS (Dudnikov type SPS) (d) semiplanotron (e) hollow cathode discharge SPS with independent emitter (f) large volume SPS with filament discharge and based emitter (g) large volume SPS with anode negative ion production (h) large volume SPS with RF plasma production and emitter 1- anode 6- hollow cathode 2- cold cathode emitter 7- filaments 3- extractor with 8- multicusp magnetic magnetic system wall 4- ion beam 9- RF coil 5- biased emitter 10- magnetic filter

24. Probability of H- emission as function of work function (cesium coverage)

25. Schematic of negative ion formation on the surfaceMichail Kishinevsky, Sov. Phys. Tech. Phys, 45 (1975)

26. Coefficient of negative ionization as function of work function and particle speed

27. Enhancing surface ionization and beam formation in volume-type H- ion sourcesR.F.Welton, M.P.Stockli, M.Forrette, C.Williams, R.Keller, R.W.Thomae, EPAC 2002, Paris. • “Cleary, once again Cs must reside on the surface for the vast majority of its lifetime in the source and therefore surface ionization must account for the observed enhancement of H- yield. • In these cases, the term ‘volume ion source’ is misleading since, most of the H- results from surface, rather than volume ionization processes. Therefore, ion source design, careful consideration should be granted the interior surfaces of the source”. • Correct classification of ion sources is important, because it should determine a direction of devices optimization: to optimize a volume production, or surface-plasma production. Incorrect speculation of main mechanism of negative ion generation was reason of long time delay in improving of beam parameters.

28. First version of Planotron (Plain Magnetron) SPS, INP, 1972, Beam current up to 230 mA, 1.5x10 mm2 , J=1.5 A/cm2 with Cs

29. H- energy spectra from planotron The ion spectra from a planotron usually have two peaks separated by a valley. The location of the first peak coincides with the energy eUex imparted to the negative ions by the extraction voltage. The ion energy of the second peak is higher than that of the first peak by an amount close to eUd. The oscillograms in the upper part of illustrate the change in the spectra, as a result of increasing the discharge voltage Ud from 120 V ( l ) to 210 V (4) by reducing the cesium supply. The oscillograms (1-4) in the lower part of Figureillustrated how the spectra vary as a result of increasing the hydrogen supply to the discharge chamber

30. Cross sections of Planotron (Magnetron) SPS of second generation: 3.7 A/cm2 with Cs (0.75 A/cm2 without Cs)

31. H- current density from planotron with Cs (3.7A/cm2) and without Cs (0.75 A/cm2), INP, Novosibirsk, 1972

32. Schematic of semiplanotron SPS 1- emission aperture; 2- anode; 3- cathode; 4- cathode insulator; 5- discharge channal; 6- extractor; 7- magnet with magnetic insertions.

33. Beam Current vs an Arc Current for Different Slit Geometry in the Semiplanotron Dependences of the Н- ion beam current on the discharge current have the N-shaped form with three sections: linear growth at small discharge currents, saturation or a falling section at medium currents, and linear, but slow growth at the high currents.

34. Cross section through LANL version ofSPS WITH Penning Discharge. Beamlet images at pepper-pot scinti1lator (noiseless discharge). Emission slit 0.5x10 mm2. Vertical: Y Plane Horizontal: X Plane

35. Schematic of ISIS version of Penning discharge SPS

36. Cathode and Plasma Plate of ISIS Penning SPS after long time operation

37. H- Energy Spectra from Penning SPS

38. Review of Scientifi Instruments, March 2002, Volume 73, Issue 3, pp. 1157-1160Investigation of the mechanism of current density increase in volume sources of hydrogen negative ions at cesium adding • V.P. Goretsky, A.V.Ryabtsev, I.A. Soloshenko, A.F. Tarasenko, A.I. Shchedrin Institute of Physics of National Academy of Sciences of Ukraine, 46 prospect Nauki, Kiev 03650, Ukraine • In the present article the influence of adding cesium into the volume and on the surface of an ion source on its emission characteristics is studied both theoretically and experimentally. It is shown that cesium in the volume at conditions of a real ion source brings in a significant contribution to kinetic processes, but weakly influences the current of H– ions extracted from the source. It is shown both theoretically and experimentally that an observed increase of the current of H– ions with cesium added is due to the conversion of fast particles at the anode surface. • Thus, on the basis of experimental results and calculations it can be stated that cesium in a volume of the source under study can not lead to the increase of current H- ions. Observed growth of this current with cesium introduction is due to conversion of hydrogen atom at discharge anode surface, covered by cesium. In other words, cesium adding results in the transformation of the source of H- ions of volume type to the source of surface-plasma type. • Yu.Belchenko,G.Dimov, V.Dudnikov, Nucl.Fusion, 14, 113 (1974)

39. Operation of Dudnikov type Penning source with LaB6 cathodesK.N. Leung, G.J. DeVries, K.W. Ehlers, L.T. Jackson, J.W.Stearns, and M.D. Williams (LBL)M.G. McHarg, D.P. Ball, and W.T. Lewis (AFWL)P.W. Allison (LAML) The Dudnikov type Penning source has been operated successfully with low work function LaB6 cathodes in a cesium-free discharge. It is found that the extracted H– current density is comparable to that of the cesium-mode operation and H– current density of 350 mA/cm2 have been obtained for an arc current of 55 A. Discharge current as high as 100 A has also been achieved for short pulse durations. The H– yield is closely related to the source geometry and the applied magnetic field. Experimental results demonstrate that the majority of the H– ions extracted are formed by volume processes in this type of source operation. Review of Scientific Instruments -- February 1987 -- Volume 58, Issue 2, pp. 235-239

40. H- Detachment by Collisions with Various Particles and Resonance Charge-Exchange Cooling Resonance charge -exchange cooling

41. Cesium escaping from a pulsed discharge in SPS there is a strong suppression of the gas and cesium flow from the emission slit by the high density plasma of the discharge.

42. Gas trapping by discharge in CSPS • qo-gas flux without discharge • qp- gas flux with discharge • Id- discharge current

43. H- Beam Intensity of SPS Years Beam intensity vs discharge current for first version of semiplanotron 1976 Evolution of H- beam intensity in ISIS

44. Emittance, Brightness, Ion Temperature δ y Emission slit l Emittance Normalized emittance x Δx Normalized brightness Δα Half spreads of energy of the transverse motion of ions Reduced to the plasma emission slit Characteristics of quality of the beam formation:

45. Discharge Stability and Noise n,1016 cm-3 noiseless Diagram of discharge stability in coordinates of magnetic field B and gas density n no discharge n* noisy Bmin B, kG μ = eν/m (ν2 + ω2) μ noiseless The effective transverse electron mobility μ vs effective scattering frequency ν and cyclotron frequency ω ν / ω

46. Noise of discharge voltage Dependence of discharge noise of magnetic field

47. Discharge Noise Suppression by Admixture of Nitrogen P.Allison, V. Smith, et. al. LANL no N2 QN2 = 0.46 sccm

48. Design of SPS with Penning Discharge

49. 1 -current feedthrough; 2- housing; 3-clamping screw; 4-coil; 5- magnet core; 6-shield; 7-screw; 8-copper insert; 9-yoke; 10-rubber washer- returning springs; 11-ferromagnetic plate- armature; 12-viton stop; 13-viton seal; 14-sealing ring; 15-aperture; 16-base; 17-nut. Fast, compact gas valve, 0.1ms, 0.8 kHz

50. Photograph of a fast, compact gas valve