High Gradients in Dielectric Loaded Wakefield Structures Manoel Conde High Energy Physics Division Argonne National Laboratory AAC 08 – Santa Cruz, CA.

High Gradients in Dielectric Loaded Wakefield Structures

Manoel Conde

High Energy Physics Division

Argonne National Laboratory

AAC 08 – Santa Cruz, CA

AWA & wakefield generation High gradient excitation Upgrades and future goals Wakefield excitation in SLAC / KEK X-band structure

Outline

Accelerator R&D at Argonne National LaboratoryHigh Energy Physics DivisionAWA Group

Wei Gai Sergey Antipov Manoel Conde Felipe Franchini Feng Gao

Chunguang Jing Richard Konecny Wanming Liu Marwan Rihaoui John Power Zikri Yusof

External Users / Collaborators:

Euclid TechLabs U.Chicago NIU Fermilab

U.Md. IIT APS Yale / Omega-P

Research at the AWA Facility

:

Advanced Accelerating Structures

Current effort has lead to comprehensive knowledge on construction and testing of dielectric based accelerating structures.

High Power Electron Beam (~ GW) and RF Power Generation

Operating a unique facility to study high current electron beam generation and propagation for efficient beam driven schemes, and high power RF generation.

Fundamental Beam Physics and Advanced Diagnostics

High brightness beam generation and propagation, phase space measurements, emittance exchange schemes.

Electron Beam Driven Dielectric Wakefield Accelerator

A high current relativistic electron beam passing through a dielectric structure can generate high gradient field, high power microwaves instantaneously. A new way to power accelerating structures by transporting the power in the electron beam.

Justifications for looking at dielectric structures:

Comparable accelerating properties as metal structures.

More material options; possibly higher gradients.

Simpler geometry, simpler construction and HOM damping.

Applications: Collinear wakefield acceleration Two-beam acceleration

Drive Beam

Dielectrics

Deceleration Acceleration

Accelerated Beam

Wakefields in Dielectric Structures (a short Gaussian beam)

Drive beam is king!

Energy Charge Bunch length Emittance

b a

Q

Cu 21

N

r

)cos(2exp)(2

2kz

a

QzW

n

zZ

Wakefield Amplitude Dependent on a

1

10

100

1000

10000

100000

0.01 0.1 1 10

Inner Radius a (mm)E

z(M

V/m

/10n

C)

AWA Drive Beamline

Drive Gun

Linac & Beam Optics Quads

Wakefield Structure

ExperimentalChambers 4.5 m

GVGV

YAG1 YAG2

Spectrometer

YAG5Dump/Faraday Cup

Slits

YAG4YAG3ICT1

ICT2 BPM

Single bunch operation– Q = 1-100 nC (reached 150 nC)– 15 MeV, 2 mm bunch length (rms), emittance < 200 mm mrad (at 100 nC)– High Current: ~10 kA

Bunch train operation– 4 bunches x 25 nC or 16 bunches x 5 nC (present) – 16 - 64 bunches x 50 - 100 nC 10 - 50 ns long (future)

The Argonne Wakefield Accelerator (AWA)

~ 1 meter

rf-gun

magnetic lenses

8 MeV 15 MeV

Laser Inλ =248nm

Linac

Quads Spectrometer

Faradaycup

YAG2

YAG1

Direction of beam propagation

Experimental Area

Experimental Setup for High Gradient Tests

WF signal

RF field probe (- 60 dB)

43 nC

time (ns)

0 2 4 6

-100

0

100

Monitor for breakdown

Infer Gradients from MAFIA

Q

Cu

Goal: Test breakdown thresholds of dielectric structures under short RF pulses.

Dielectric Loaded Structures Tested

SW Structure #1 C10-102 #2 C10-23 #3 C5.5-28 #4 Q3.8-25.4

Material Cordierite Cordierite Cordierite QuartzDielectric constant 4.76 4.76 4.76 3.75Freq. of TM01n 14.1 GHz 14.1 GHz 9.4 GHz 8.6 GHzInner radius 5 mm 5 mm 2.75 mm 1.9 mmOuter radius 7.49 mm 7.49 mm 7.49 mm 7.49 mmLength 102 mm 23 mm 28 mm 25.4 mmWakefield Gradient 0.45 MV/m/nC 0.5 MV/m/nC 0.91 MV/m/nC 1.33 MV/m/nC

Wakefield Measurements: Structure #1 (C10-102)

10 12 14 16 18

-200

0

200

4 6 8 10 12 14

-200

0

200

12 14 16 18 20

-200

0

200

4 6 8 10 12

-200

0

200

6 8 10 12 14

-200

0

200

time (ns)

0 2 4 6 8 10

-200

0

200

4.0 nC

9.2 nC

11.8 nC

16.8 nC

20.5 nC

25.6 nC

mix

er

ou

tpu

t (m

V)

bunch charge (nC)

0 10 20 30 40 50

pe

ak

mix

er

ou

tpu

t (m

V)

-600

-400

-200

0

200

400

600

46 nC → 21 MV/m

MAFIA Simulation of Structure #1 (C10-102)

Snapshots of wakefield amplitude

Wakefield Measurements: Structure #2 (C10-23)

Measurement

Simulation

Measurement

Simulation

TM013 (14.1GHz)

TM014

(16.2GHz)TM012 (13GHz)

HEM111 (12.4GHz)

HEM111 (12.2GHz) TM012

(13GHz)

TM013 (14.3GHz) TM014

(16GHz)

HEM112 (14.7GHz)

HEM112 (14.7GHz)

Freq (GHz)

Freq (GHz)

86 nC → 43 MV/m

Measured and simulated Er probe signals

Wakefield Measurements: Structure #3 (C5.5-28)

TM013

86 nC → 78 MV/m

E-field pattern

Wz (V/m)Wz > 1MV/m @ 1nC for 10GHz Structure

28mm

2.5mm

7.5mm

MAFIA Simulation of Structure #3 (C5.5-28)

Wakefield Measurements: Structure #4 (Q3.8-25.4)

75 nC → 100 MV/m

HEM111

TM012 TM013

TM014

Structure #1 21 MV/m Structure #2 43 MV/m Structure #3 78 MV/m Structure #4 100 MV/m

Gradients Reached:

Where we go from here:

Once the upgrade is complete, the goal is to achieve:Higher gradient excitation: ~ 0.5 GV/m in structures ( 3 mm apertures)Higher RF power extraction: ~ 1 GW (10 ns)

To improve the drive beam, we are currently upgrading the AWA facility:A new L-band 1.3 GHz RF station to increase the drive beam energy from 15 to 25 MeV.Fabrication of Cesium Telluride high quantum efficiency photocathodes to produce long, high charge bunch trains.A second RF gun to restore two-beam-accelerator capability.

Future AWA Facility (25 MW + 25 MW = 50 MW)

*all distances in cm

D.U.T.

Drive Gun(12 MW)

Linac 1(10.5 MW)

0 225 cm 351.6 cm 581.6 cm29.1 cm 455 cm 650 cm

Linac 2(10.5 MW) D.U.T.7.5 MeV 15.75 MeV 25 MeV

Witness Gun(12 MW)

0 29.1

8 MeV

Single Bunch: 50 - 100 nCBunch Train: 16 – 64, total charge 1 – 2.5 µC

Some goals of the AWA program: Verify that structures can withstand gradients higher than 100

MV/m. Generation of ~ GW level RF power (25 MeV, 130 Amps, 10

ns, 3 GW beam power). Demonstrate high gradient, broad bandwidth, low cost

structures (power extractors and accelerators). Typical parameters:

– f = 10 – 30 GHz, Vg = 5 – 20%– Required power: 0.5 – 5 GW peak– Example (for dielectric based structure):

• 21 GHz, a = 3 mm, b = 4 mm, ε=12, Vg=0.112• For Ez=200 MV/m, it requires P= 1 GW• 16 ns RF pulse, structure length : 30 cm, fill time = 8.4 ns• Beam loading (fundamental mode):0.75 MV/m/nC

Wakefield Observation on SLAC / KEK X-Band Standing Wave Structure

• Single bunch charge: up to 80nC.

Measurements:• Time structure and spectrum of the generated wakefield;• Linearity of the voltage-charge curve.

Dimensions for copper only 3-cell standing wave structure experiment

t

b_c

onv

b_e

nd

b1

b2

a_cp

l

a_p

ipe

5×Rb

Rpipe

5×el

lips_

r5×D

2b_end 23.241249

2b1 22.968

2b2 23.072

2b3 22.965

2b_cpl 22.970

2b_conv 22.86 0.9inch

2a 11.295

2a_pipe 12.7 0.5inch

4×a

a_cpl 5.2715

D 13.116

Rpipe 3.00

Rb 1.00

t 4.5980688

ellips_r 3.398573

Dimensions for 20 deg. C V.A.Dolgashev, 2 March 07

b3

b_c

pl

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

-0.2

-0.1

0

0.1

0.2

0.3

5 6 7 8 9 10 11 12 13 14 150

20

40

60

80

100

120

11.427GHz5500

9.408GHz4600

8.361GHz3600

7.783GHzQ ≈ 6700

t -μs

volta

ge -

Vol

t

f – GHz

volta

ge s

pect

rum

13.711GHz1700

13.825GHz3900

q = 34.6nC

Measured voltage signal

5 6 7 8 9 10 11 12 13 14 150

20

40

60

80

100

120

11.427GHz

9.408GHz

8.361GHz7.783GHz

f – GHz

volta

ge s

pect

rum

13.711GHz

13.825GHz

5 6 7 8 9 10 11 12 13 14 150

5

10

15x 10

4

f – GHz

spectrum of the wakefield in the vacuum break for ICT (simulated)

The wakefield in the ceramic vacuum break for ICT may contribute to some of the peaks

vacuum break

5 6 7 8 9 10 11 12 13 14 150

20

40

60

80

100

12011.427GHz

f – GHz

volta

ge s

pect

rum

The 11.4GHz component is filtered out

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-0.04

-0.02

0

0.02

0.04

t -μs

volta

ge -

Vol

t

q = 34.6nC

IFFT

Q ≈ 5500Vmax

Vmin

vol

tage

pea

k va

lue

charge q - nC

For 11.427GHz component:

0 10 20 30 40 50 60 70 80 900

0.01

0.02

0.03

0.04

0.05

0.06

0.07