High-Gradient Test of a 3 GHz Single-Cell Cavity 4th Annual X-band Structure Collaboration Meeting Silvia Verdú-Andrés (TERA, IFIC) U. Amaldi, M. Garlasché (TERA), R. Bonomi (TERA, Politecnico di Torino), A. Degiovanni, A. Garonna (TERA, EPFL), R. Wegner (TERA, CERN), C. Mellace (A.D.A.M.)

Hadron therapy: the basics Charged hadron beam that loses energy in matter 27 cm Tumour target 200 MeV - 1 nA protons protons 4800 MeV – 0.1 nA carbon ions carbon ions (radioresistant tumours) Courtesy of PSI Silvia Verdú-Andrés tail httt://global.mitsubishielectric.com/bu/particlebeam/index_b.html light ion (carbon) proton 2 Depth-dose distribution

Treating moving organs requires... Silvia Verdú-Andrés3  Fast Cycling machine (high repetition rate ~ Hz) Tumour MULTIPAINTING  Fast Active Energy Modulation (a couple of ms) Fast 3D correction of beam spot position in depth Single ‘spot’ pencil beam Lateral scanning with magnets: 2 ms/step 3D conformal treatment Depth scanning: ACTIVE ENERGY MODULATION

TERA’s proposal: cyclotron + high-freq. linac = cyclinac 120 MeV/u 400 MeV/u Silvia Verdú-Andrés Cell Coupled Linac RF frequency: 5.7 GHz 18 accelerating modules - Length of each module ~ 1.3 m High gradient : 40 MV/m (TERA+CLIC collaboration) Cell Coupled Linac Standing-wave structure RF frequency: 5.7 GHz 4 (CArbon BOoster for Therapy in Oncology) CABOTO

TERA’s proposal: cyclotron + high-freq. linac = cyclinac 120 MeV/u 400 MeV/u CArbon BOoster for Therapy in Oncology CABOTO = Silvia Verdú-Andrés Cell Coupled Linac RF frequency: 5.7 GHz 18 accelerating modules - Length of each module ~ 1.3 m High gradient : 40 MV/m (TERA+CLIC collaboration) Higher accelerating gradients  Smaller complex! 5 Cyclinac: E 0 ~ 40 MV/m  E Max ~ 200 MV/m BDR ~ bpp/m in 30 m

Silvia Verdú-Andrés 3 GHz high-gradient test: motivation and objectives 6

Silvia Verdú-Andrés Break Down Rate BDR per length  Operation limit for S-band cavities  Break Down Rate BDR per length Limit given by: E s or  surface field E s (Kilp.) or S c  modified Poynting vector S c + scaling law (X, K-band) 3 GHz high-gradient test: motivation and objectives 7

Silvia Verdú-Andrés Break Down Rate BDR per length  Operation limit for S-band cavities  Break Down Rate BDR per length Limit given by: or  surface field E s (Kilp.) or  modified Poynting vector S c + scaling law (X, K-band)  Scaling laws at S-band [E s, S c, pulse length, temperature, repetition frequency] 3 GHz high-gradient test: motivation and objectives 8

Silvia Verdú-Andrés Break Down Rate BDR per length  Operation limit for S-band cavities  Break Down Rate BDR per length Limit given by: or  surface field E s (Kilp.) or  modified Poynting vector S c + scaling law (X, K-band)  Scaling laws at S-band [E s, S c, pulse length, temperature, repetition frequency]  Applying found limit to future designs  ensure reliable operation  optimize RF structures (efficiency, length, cost) 3 GHz high-gradient test: motivation and objectives 9

 single-cell test cavity Silvia Verdú-Andrés Break Down Rate BDR per length  Operation limit for S-band cavities  Break Down Rate BDR per length Limit given by: or  surface field E s (Kilp.) or  modified Poynting vector S c + scaling law (X, K-band)  Scaling laws at S-band [E s, S c, pulse length, temperature, repetition frequency]  Applying found limit to future designs  ensure reliable operation  optimize RF structures (efficiency, length, cost) 3 GHz high-gradient test: motivation and objectives 10

max. E max. H max. S c RF & mechanical design Silvia Verdú-Andrés β= 0.38 (E kin =70 MeV) f 0 = 3000 MHz Q 0 = 9000 E max /E 0 = 6.5 Cooling ANSYS P ave = 350 W Mass Flow = 2.5 L/min  f/  T= -1.1 MHz / 20K nose 11

Silvia Verdú-Andrés Production Accelerating 3 GHz (two unsymmetrical half cells) RF coupling system (waveguide, short circuit) Connection to data acquisition (through CF flanges) Cooling system (two plates, in-out pipes)  OFE copper  0.02 mm tolerance  0.4  m roughness 12

Production Silvia Verdú-Andrés  Machining at Veca (Modena, Italy)  Cleaning at CERN (Geneva, Switzerland)  (vacuum) Brazing at Bodycote (Annecy, France) Done in less than one month! 13

Low power test Silvia Verdú-Andrés Short Circuit Test Cavity RF input f 0 = MHz Q 0 = 8650 (96% Q 0,sim )  = 0.92,  =-27 dB 14

Silvia Verdú-Andrés First high-power test: objectives  Debugging test set-up and cavity  First check of cavity behaviour under high-power  Finding improvements for precision test to evaluate scaling laws [BDR(E s, T p, f rep )] Only 1-2 weeks foreseen for test Purpose of first test: 1. First check of RF behaviour of test cavity 2. Debugging set-up and test of cavity 3. Finding improvements for precision test (machining, instrumentation, data acquisition) to evaluate scaling laws [BDR(Es, Tp, frep)] 15

Silvia Verdú-Andrés High power test: set-up Faraday Cup Peak Power Analyser Temperature sensors no Data Acquisition System no control system for stabilising frequency & amplitude no RF CTF3 16

At resonance: At resonance: adjusting frequency until P REF (f) < 10% P FW (f) Hot cavity 17 High power test: measurements Peak Power Meter P FW P REF T pulse = 5  s f rep = 50 Hz  no RF Pickup  relying on power measurements 0.93 E S = E S,calculated * 0.93 P cav = P FW -P RF E acc

Silvia Verdú-Andrés 18 Breakdown! 125 pulses, 1 BD  BDR= T Flat top = 3 μs High power test: breakdown evaluation Breakdown! Faraday Cup 18

High power test: first results # Sett FLAT [µs] P ABS +/-  [kW]E S  +/- 10  MV/m] S c [MW/mm2]BDR/L [bpp/m]Kilp / E / E / E / / / E / E / / E / E / E / E / E / E / E / E / E / E / E Silvia Verdú-Andrés

20 Gradient limitations for high frequency accelerators, S. Döbert, SLAC, Menlo Park, CA 94025, USA (2004) High power test: comparison to other tests Limit in copper to surface field by breakdown surface damage TERA Silvia Verdú-Andrés20

21 A New Local Field Quantity Describing the High Gradient Limit of Accelerating Structures, A. Grudiev, S. Calatroni, and W. Wuensch, Phys.Rev. Accel. Beams (2009) The modified Poynting vector as an RF constraint to high gradient performance The square root of S C has been scaled to t pulse =200 ns and BDR=10 -6 bbp/m first results Sqrt(S c ) є [ ] High power test: comparison to other tests 21Silvia Verdú-Andrés

High power test: scaling laws 22 Max.BDR: bpp/m Treatment Session: Hz Max. 1 BD per session X = 30 (other exp.) TERA X = 6 TERA 22

Silvia Verdú-Andrés High power test: scaling laws 23 Max.BDR: bpp/m Treatment Session: Hz Max. 1 BD per session X = 30 (other exp.) TERA X = 6 TERA Not enough statistics for scaling law’s evaluation 23

Surface inspection around nose region Activity in the cavity: ~ breakdowns Silvia Verdú-Andrés [90X Optical Microscope] max. E max. H max. S c 24

Silvia Verdú-Andrés Surface inspection around nose region Max. roughness: 0.4  m Copper Grains (tenths of  m) 25

Silvia Verdú-Andrés Surface inspection around nose region Several craters… and a lot of activity 26

Summary SYSTEMRESULTS / IMPROVEMENTS Vacuum systemvery good Cavity - low powervery good >96% Q 0, f 0 ok, β=0.92 (  <-27dB) Cavity - high powervery good E S > 350 MV/m, BDR < /m, T pulse = 2 μs Coolingfreq. shift too big => water flow too small  Excellent Pre-Test, cavity works, even first results Silvia Verdú-Andrés27

Summary SYSTEMRESULTS / IMPROVEMENTS Vacuum systemvery good Cavity - low powervery good >96% Q 0, f 0 ok, β=0.92 (  <-27dB) Cavity - high powervery good E S > 350 MV/m, BDR < /m, T pulse = 2 μs Coolingfreq. shift too big => water flow too small  Excellent Pre-Test, cavity works, even first results  Improvements for high-precision test  cooling / water flow control  data acquisition (P forward, P reflected, φ forward, φ reflected, V faraday cup, vacuum)  control system for stabilising frequency & amplitude Silvia Verdú-Andrés28

Next steps: a lot to do!  High-precision high-power test of the 3GHz test cavity  Design, construction and high-power test of another single-cell cavity operating at 5.7 GHz  to evaluate scaling laws  learn more about breakdown phenomena  Design has already begun; to be tested before the end of 2010! Silvia Verdú-Andrés29

Thanks to… o CLIC RF and Breakdown Group o CTF3 Group o CERN General Services o VECA, Bodycote, ADAM o Vodafone All the Cyclinac team for their support And all of you for the attention! TERA -Cyclinac team The research leading to these results has received funding from the Seventh Framework Programme [FP7/ ] under grant agreement n° Silvia Verdú-Andrés30

BACK-UP SLIDES Silvia Verdú-Andrés 31

Silvia Verdú-Andrés RF & mechanical design Requirements  Average power to cool (350 W)  Nº of parallel circuit (2)  Turbulent flow  Avoid erosion/corrosion  Reference temp. for coolant properties (37ºC)  High heat transfer coefficient Choices  dT in-out = 1ºC  D eq = 5.5 mm  Re =  v = 1.77 m/s  h = W/m2/K  Mass flow 2.5 l/min  Δf 0 / ΔT RF = MHz / 20 K 32

Halfcell design 33 Øcell [mm]64.50 coupling slot [mm]25.75 x 6.00 cavity profile0.02 mm tolerance Ra 0.4 materialOFE copper Silvia Verdú-Andrés

Cooling design 34 Silvia Verdú-Andrés OFE Copper / 316 L Two pipes coated and brazed to cooling plate Q water < 2.5 l/min

Waveguide design Silvia Verdú-Andrés35 Design evolution: Standard WR-284(OFE copper) with two LIL flanges for connection to RF source and short

t FLAT [µs]P ABS [kW]E S [MV/m]E 0 [MV/m]Sc [MW/mm2]BDR/L [bpp/m]Kilp E E E E E E E E E E E E E E E E Measur e MAX E S = 150 MV/m S C = 0.46 MW/mm 2 E S /E 0 = 6.5 P = 128 kW SFSF E K = 47 MV/m 36Silvia Verdú-Andrés Cavity Performance Old values

Surface inspection around nose region Activity in the cavity: ~ breakdowns Silvia Verdú-Andrés [90X Optical Microscope] 37

Silvia Verdú-Andrés38