Upload
thinfilmsworkshop
View
476
Download
1
Tags:
Embed Size (px)
DESCRIPTION
http://www.surfacetreatments.it/thinfilms An Approach to Chemical Free Surface Processing for High Gradient SRF Cavities (Frederick Mako - 30') Speaker: Frederick Mako - FM Technologies | Duration: 30 min. Abstract Frederick Mako1, Ph.D., Bing Xiao1, Ph.D. and Larry Phillips2, Ph.D. Chemical treatment such as buffered chemical polishing (BCP) or electro polishing (EP) followed by high pressure rinsing (HPR) of niobium (Nb) superconducting RF (SRF) cavities is expensive and complex multistep process. Furthermore, the cavity RF surfaces after the treatment still have numerous bubbles and pits that result from welding. These quench-producing weld defects together with the particulate contamination, result in significant scatter of the multi-cell Nb SRF cavities performance characteristics. This scatter is the major problem in the current manufacturing of the Nb SRF cavities. FM Technologies proposes a new approach to chemical-free processing for multi-cell Nb SRF cavities using an internal electron beam (IEB). Specifically, FMT proposes to develop a new electron gun system that will perform electron beam melting over the entire interior surface of Nb SRF cavities to produce a smooth surface, free from voids, bubbles, and other imperfections. This may allow manufacturing of the Nb SRF cavities with a reduction in the above chemical treatment procedures and increase the cavities high gradient performance. FM Technologies will design, build, test the new IEB system and process samples/cavities and Thomas Jefferson Laboratory will measure RF performance of processed samples/cavities. Preliminary electron beam melting results will be presented. 1FM Technologies, Inc., Chantilly, VA, USA, 2Thomas Jefferson Laboratory, Newport News, VA, USA
Citation preview
An Approach to Chemical Free Surface Processing for High Gradient Superconducting
RF Cavities
Frederick Mako1, Ph.D., Bing Xiao1, Ph.D. and
Larry Phillips2, Ph.D., William Clemens2
1FM Technologies, Inc., Chantilly, VA, USA, 2Thomas Jefferson Laboratory, Newport News, VA, USA
Work Supported by US DOE under SBIR
Seven-Cell Nb SRF Cavity at Thomas Jefferson National Accelerator Facility.
Pictures Show Typical Defects Occurring Inside The Nb SRF Cavity Cells Around The Equator
EBW Overlaps And Remaining After The Chemical Treatment
Outstanding irregularity (step) near equatorEBW overlap of cell#7 from waveguide.
Many “bubbles” sporadically present inside the weld.
Two other cells have less pronounced feature.Four other cells have no recognizable feature.
Many apparent “deep pits” in heat affected zone.
Electron beam weld
1 mm
Edge of electron beam weld
1 mm
1 mm
1 mm
step
Study of Beam Processing for Cavities
Objectives:• Achieve a Smooth Surface with Minimal
Defects and Impurities• Achieve a Low Strain Surface to Reduce
Corrosion• Final Goal is to Attain Reproducible High Q
(>1010) and High Field (40MV/m) Cavities
Electron Beam Melted Nb Samples Using J-lab SCIAKY Welder
25 mm
Each single pass melt region is about 6 mm x 74 mm x 0.1-0.2mm deep. A 10 kHz circular to elliptical raster with 0.5 mm diameter 50 keV beam was used.
Beam current and translation rates varied from 20-250mA and 5-20 inches/minute.
28 plates of Nb with dimensions 3 mm thick x 25.4 mm wide x 88.9 mm long.
Magnification of Melt Zone
Melted Region
Un-Melted Region
HIROX digital microscope view of sample #6.
The bottom half of the image shows the smooth melted region that highlights the grain size of about 300-400 µm, while the upper half of the image shows the rough un-melted small grain region.
Grain Reference Orientation Deviation (GROD) Map for Nb Flat Sample #7
Both samples #6 and #7 show GROD in the range of 0 – 3° over a distance of 100 – 300 μm.
40mA and 10in. /min.
No chemical etching has occurred.
Grain Reference Orientation Deviation (GROD) Map for Nb Flat Sample #15
Samples #14 and #15 show less lattice distortion compared to samples #6 and #7 as measured by GROD from 0 – 2° with the majority of the distortion less than 1°over 100 – 300 μm.
40mA and 18in./min.
No chemical etching has occurred.
Grain Reference Orientation Deviation (GROD) Map for Nb Pipe Sample #16
The pipe sample #16 that utilized a double low current (24 mA & 18in./min.) e-beam pass also shows compared to samples #6 and #7with very little lattice distortion as indicated by its GROD with the majority of the distortion less than 1° over 100 – 300 μm.
No chemical etching has occurred.
Grain Reference Orientation Deviation (GROD) Map for Deep Drawn Half Cell
Courtesy of Dr. Roy Crooks of Black Laboratories
The section is taken from 3 mm from the equator.
The GROD map shows extreme lattice distortion up to an angular rotation of about 20 degrees.
No chemical etching has occurred.
Comparison of Grain Deviation for E-Beam Processed Nb vs. Deep Drawn Nb Cell
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 200
10
20
30
40
50
60
70
80
90
100
Deep Drawn
Sample 6
Sample 7
Sample 14
Sample 15
Double Pass
Maximum Grain Orientation Deviation Angle (deg.)
Frac
tion
of G
rain
s at
Max
imum
Dev
iatio
n A
ngle
(%)
Comparison of Grain Deviation for Various E-Beam Processed Nb Samples
1 2 30
10
20
30
40
50
60
70
80
90
100
Sample 6
Sample 7
Sample 14
Sample 15
Double Pass
Maximum Grain Orientation Deviation Angle (deg.)
Frac
tion
of G
rain
s at
Max
imum
Dev
iatio
n A
ngle
(%)
HIROX Imaging for Sample #15
Sample #15 shows very smooth HIROX and AFM images in the melt zone.
AFM of Sample #15
AFM RMS value of less than 3 nm, which is 1/20 of what can be accomplished by electro-polishing.
Magnification of Melt Zone
Melted Region
Un-Melted Region
Grain Boundary Step
HIROX digital microscope view of sample #6.
Grain boundary steps show up in higher heating rate samples.
AFM of Sample #6 at Grain Boundary
Chemical Free Half-Cell Processed in the J-lab E-beam SCIAKY Welder
Electron Gun
Half Cell
Finished E-Beam Processed Half-Cell
The beam parameters were: 40 mA, 0.5mm diameter beam, travelling at 18 inches per minute, the melting diameter is about 6 mm with a circular pattern at 10 kHz.
Summary of Beam Parameters for A Smooth Low Strain Surface
• Both HIROX & AFM suggest that a smoother surface is attained with lower heating rates
• EBSD (GROD) maps suggest a lower strain surface is attained at lower heating rates
• Desired Beam Parameters are: 50keV, 40mA, 46cm/min. in single pass or 50keV, 24mA,46cm/min. in double pass
Electron Gun and Beam Transport Design
• Two Strategies: Ballistic Focusing and Magnetized Beam
CRITERIA:– Can process from iris to equator and the circumference– Prevent Nb vapor arcing– Can tolerate beam induced thermal radiation & filament
heat load– Beam Parameters ~50kV, up to 200mA, spot ~1mm– Focal Length 30-100cm.
Ballistic Focusing Gun-For In or Out of Cavity Processing
High Voltage Insulator
Grid Provides 1st Focus, V=-50 to -51kV, Current Not Effected
Anode, V=0
Cathode & Filament,
V=-50kV
Magnetic & Radiative Heat Shield, Water Cooled,V=0
Dipole w Soft Iron Case provides 2nd Focus, V=0
Quadrapole Provides Rastering, Circle or Ellipse, V=0
46mm
Diameter
Ballistic Focusing Gun-Electron Beam Trajectory
Helmholtz Coils Provide R-Z Beam Scanning from Iris to Equator.
q Scanning Provided by Rotating either Gun or Cavity or Helmholtz coils on Axis.
Magnetized E-Beam Gun
Simple Gun Design Only Needs Cathode, Grid & Anode
Magnetized E-Beam Gun-Electron Beam Trajectory
Beam location is dependent on bucking coil position.
50 keV, 50 mA, 3 mm diameter.
Also, beam position can be adjusted by changing the bucking coil current.
Summary and Conclusions
• Beam Parameters Have Been Determined to Give a Smooth Low Strain Surface using a Conventional Rastered Beam
• Two Gun Designs Examined (Ballistic & Magnetized Beams) to Meet the Above Beam Conditions With & Without Rastering for Both Internal & External Gun Operation
• Maintaining a Constant Beam Power Density on the Cavity Surface Will Require Computer Controls
• Further Sample Studies are Required Before Gun Selection