Kein Folientitel

1
PSI-XFEL Project
Low Emittance Electron Gun for XFEL
Application Christopher Gough on behalf of the
XFEL Team www.psi.ch leg.web.psi.ch
christopher.gough_at_psi.ch
PESP 1-3 October 2008
2
PSI XFEL Project
3
Paul Scherrer Institute, Switzerland
Nanotech lab Linac test bunker
Proton cyclotron 590MeV 2mA
SLS 2.4GeV 400mA
After completion of the SLS, what next ?
4
An XFEL! 6GeV 0.1nm, 200pC, 125fs
en 0.2 mm mrad
total length 800 m
5
Project milestones
2007
2008
2009
2010
2011
2012
6
PSI XFEL Features

• 1. Low emittance electron emission
• field emission from tips
• fallback to photoemission from copper
• 2. Fast acceleration after emission to limit
  effect of space charge forces
• diode configuration with high applied pulsed
  voltage
• fallback to existing 2.5 cell 3GHz RF gun for
  linac development
• 3. Low initial current to reduce beam blow up by
  space charge effects
• ? 3-fold bunch compression scheme for 300x
  compression
• 4. Electron gun design goals
• ? 5.5A, 30ps, uniform longitudinal
• ? 0.5mm diameter, uniform transverse
• ? lt0.1nm-rad
• ? 1MV, gt150MV/m acceleration

7
Field Emitting Arrays (from Soichiro Tsujino and
Romain Ganter)
8
Gated and focus layer
High Gradient
100 250 MV/m
Focusing
Gate
Tip
Substrate
5 mm
1 mm
9
First double gate device in PSI
Extraction-gate
Second (focusing) gate
4x4 array
Aperture ? 2.3 ?m
dbl-oxidation 1.5 ?m-base /5 ?m-pitch emitters
10
Early result with silicon substrate
Goal 5.5A
Si wafer internal resistance limitation
11
Adsorbates and non-uniform emission
With DC mode , bistable behaviour due to
adsorbates
12
Pulser and First Results
13
500kV pulser in bunker - September 2007
14
500kV Pulser
15
Cross section with Phase 1 (500kV) beamline
Dust-free glove box
Tesla Coil
Ceramic insulator separating SF6 from UHV
Cathode-anode region in UHV chamber
16
Cross section of 500kV pulser
Mover adjusts A-K separation 0-30mm
Dust-free glovebox
50kV thyratron pulser
Mover for cathode loading
Movers for cathode X-Y alignment /-2.5mm
50kV 500kV Tesla coil in SF6 _at_ 5bar
17
500kV pulser opening procedure
18
Open the AnodeCathode gap to the maximum
19
Pull back FEA holder
20
The manipulator releases FEA holder
21
The manipulator takes FEA holder out
22
Pump out SF6 and open the tank
23
Secure the stalk
24
Release the stalk connection
25
Separate SF6 chambers
26
Release the vacuum
27
Open the vacuum tank
28
Take out the Cathode
29
Take out the Anode
30
Pulser - Phase 1 (500keV)
Vacuum Chamber (typically 2e-8mbar)
Cathode Electrode (-500kV)
Anode Electrode (0V)
31
Operating waveforms
Cathode Voltage -250kV peak 1mm NaI
scintillator 1m from cathode
Scintillator measures gamma rays over wide energy
range Further work needed to correlate electron
emission and X-rays Well polished electrodes have
zero X-ray emission
32
Early electrode test
Flat surface machined on face of previously
conditioned copper cathode Result Field
enhancement due to sharp edges not important.
33
Early electrode test
With no conditioning 31MV/m
34
Early electrode test
100µm
Sharp edges did not provoke sparking
Old crater with new machining to give sharp edges
- no spark
Transition from slope (left) to flat area. One
of the few spark locations near to sharp edge.
Source E. Kirk
35
Early electrode test
100µm
Some locations showed layering in copper
Detail of spark location
Spark location (left) and knife cut made after HV
test (right)
Source E. Kirk
36
Early electrode test
100µm
Some locations showed layering in copper
Detail of spark location
Spark along machining groove
Source E. Kirk
37
Early electrode test
100µm
Some locations showed layering in copper
Detail in middle of crater
Typical spark location - darkness on edges is
artifact only
Source E. Kirk
38
Two frequency, 2-cell Cavity for the 250MeV
Injector (from Jean-Yves Raguin)
39
Two frequency cavity
4.5 GHz
1.5 GHz
Gun2-3FullFiltWG3D
4.5 GHz
40
Two frequency cavity
TM 012-p
TM 010-p
4.5 GHz
1.5 GHz
Loaded Q 11,360 Coupl. Coef. ß 0.96 Peak
surf. field 46 MV/m (for on-axis peak field of
40 MV/m) Required power 3.5 MW Frequency
separation between 0 mode and operating mode 8
MHz
Unloaded Q 25,770 Required power 220 kW (for
on-axis peak field of 40 MV/m)
41
Duplex filter

• SOME INNER DIMENSIONS
• Inner conductor radius 18.12 mm
• Outer conductor radius 41.72 mm
• Total length 107.70 mm
• Cell lengths 30.20 mm and 33.30 mm

42
Pulsed Solenoid
43
Pulsed solenoid
Pulsed solenoid
44
Pulsed solenoid
Primary current flows for 100usec, then turns
off quickly
Secondary coil precision ring with non asymmetry
45
Pulsed solenoid
Advantages low power at 10Hz as stable as DC
operation zero leakage field on the cathode due
to eddy current shielding asymmetry limited by
mechanical tolerances (/-20um over 20mm ?)
46
High Gradient Tests with Pulser
47
SS electrodes 10 Nov07
Stainless steel cathode Stainless steel mesh for
anode
Mesh tested if local photon generation provokes
breakdown. Mesh has very low thermal
conductivity - glowing areas give quite precise
measure of emission location
48
SS electrodes 16 Nov07
Polished stainless steel cathode and anode
49
Ball bearing cathode, Cu anode 18 Nov07
Hardened ball bearing for cathode, copper plate
for anode Most damage on anode
50
Copper electrodes 20Nov07
Anode plate with a selection of holes 2 - 6mm
diameter Sparking not particularly related to
holes Possibly a preferred location for sparks
near the change of gradient on the
cathode Severe discoloring occurred during
washing for 30minutes in ultrasonic bath
(deionized water at 60C)
51
TiVAl electrodes 05Apr08
Polished extensively 130MV/m heavily damaged
by sparking then Remachined and polished 70MV/m
damaged by sparking then Polished
extensively 111MV/m
Variability Polishing or surface stress from
machining?
52
Typical waveforms
"Hard" spark - all energy into spark
Cathode negative voltage peak
Scintillator X-ray signal - often single photon
with variable energy
53
SS electrodes 19Apr08 400kV
Cathode Voltage
Delayed Xrays
Immediate Xrays
54
ZrC Tip 29Apr08
Gradient at first spark 10MV/m After several
sparks 30MV/m Tip becomes rounded from sparks
55
Typical breakdown test
Gradient at first spark 62MV/m - first spark
wrecks the electrodes Restarted at 30MV/m, but no
chance to recover
56
SS anode, 133MV/m _at_ 2mm gap
General appearance of polished surface Molten
globule from spark
1mm
57
SS anode, 133MV/m _at_ 2mm gap
Characteristic "vulcano" shapes caused during
electron beam melting Inhomogeneitites melt at
lower temperature than bulk metal and erupt as gas
20mm
58
SS cathode, 133MV/m _at_ 2mm gap
Voids and deep scratches Not a preferred site
for spark
20mm
59
SS cathode, 133MV/m _at_ 2mm gap
Melted SS drawn away from the surface - spike
about 14?m high Not destroyed by subsequent
pulses, so not dominating breakdown
20mm
60
Conclusions of tests
Up to 130MV/m, only surface polishing is
dominant geometry (within reason) material type
and purity sharp edges and scratches anode
thickness argon cleaning vacuum
pressure surface cleanliness are not
dominant. Parasitic emission measurement and
spark counting not so useful Improved polishing
means that no parasitic emission occurs until
breakdown - traditional "conditioning" is not
useable with the best surfaces.
61
Status and Future of Low Emittance Gun
62
Summary

• 1. Over 45 electrode pairs tested in pulser
• - considerable experience gained
• - reliable pepperpot measurements with
  photoemitted beam
• 2. Breakdown is presently limited by surface
  polishing only
• - 60MV/m is standard for most electrodes
• - stainless steel is standard material, low cost,
  easy to polish
• - exceptionally well polished electrodes reach
  130MV/m
• - when focused, 20uJ 266nm laser can provoke
  explosive emission
• 3. Improved surface treatment will give required
  gradient
• - EBeam melting, deposition and other surface
  treatment in progress
• 4. Development will continue for field emitting
  arrays, high gradient surfaces and the
  2-frequency cavity
• - in the mean time, the 250MeV test stand will
  use a conventional RF photogun

63
PSI-XFEL Project
Thank you for your attention
christopher.gough_at_psi.ch
PESP 1-3 October 2008