Alex Lumpkin and Manfred Wendt,
Fermilab,
Presented at the LCWS11
September 27, 2011
Granada, Spain

I.
Introduction
II.
Beam profiling with YAG:Ce* scintillation
• Scintillator resolution
• Depth-of-focus issue
III. Optical Transition Radiation (OTR)
• OTR basics
• OTR point-spread-function (PSF) aspects
• Microbunching instability and coherent OTR
IV. Future tests
V. Summary
*Yttrium Aluminum Garnet, Cerium doped
A. Lumpkin and M. Wendt LCWS September 26, 2011 2

Beamline and diagnostics support for EEX applications
Parameter Units Value
Energy
Energy spread
MeV keV
15
10 ‒ 15 transverse emittance mm mrad 2.6
±0.3
Bunch length
Bunch intensity psec nC
3.1
±0.3
~0.1
‒ 5
3.9 GHz TM
110
Cavity
A. Lumpkin and M. Wendt LCWS September 26, 2011 3

• The charged-particle beam transverse size and profiles are part of the basic characterizations needed in accelerators to determine beam quality, e.g. transverse emittance.
• A basic beam imaging system includes:
– conversion mechanism ( scintillator, optical or x-ray synchrotron radiation
(OSR or XSR), Cherenkov radiation (CR), optical transition radiation (OTR), undulator radiation (UR), and optical diffraction radiation (ODR).
– optical transport (lenses, mirrors, filters, polarizers).
– imaging sensor such as CCD,CID, CMOS camera, with or without intensifier and/or cooling
– video digitizer
– image processing software beam image processing
& controls software conversion screen imaging camera video digitizer optical transport
A. Lumpkin and M. Wendt LCWS September 26, 2011 4

• System related
– YAG:Ce powder and crystal screen spatial resolution.
– Camera resolution and depth of focus.
– OTR polarization effects and OTR point spread function.
– Camera calibration factor.
– Finite slit size (if applicable) .
• Accelerator / beam related
– Beta star term in spectrometers.
– Macropulse blurring effects on energy spread , beam size, and beam divergence in OTR images.
A. Lumpkin and M. Wendt LCWS September 26, 2011 5

• Uncorrelated terms are treated as a quadrature sum to actual image size Act (see Lyons’ textbook a ).
– Observed image size Obs
– YAG screen effects YAG
– Camera resolution Cam
– Finite slit width Slit
• In addition there can be macropulse effects and OTR polarization effects.
𝑂𝑏𝑠 2 = 𝐴𝑐𝑡 2 + 𝑌𝐴𝐺 2 + 𝐶𝑎𝑚 2 + 𝑆𝑙𝑖𝑡 2
𝐴𝑐𝑡 = 𝑂𝑏𝑠 2 − 𝑌𝐴𝐺 2 − 𝐶𝑎𝑚 2 − 𝑆𝑙𝑖𝑡 2 a Louis Lyons , Statistics for Nuclear and Particle Physicists (1986)
A. Lumpkin and M. Wendt LCWS September 26, 2011 6

Efficiency
Spatial resolution
YAG:Ce (Cerium doped) powder or single crystal
~100x
OTR screen, e.g. Al or aluminized Si
1x
Volume effect, grain size EM surface phenomena
Spectral content Narrow band (~20 nm)
Saturation, non-linearities at high beam intensities
Response time ~50 – 100 nsec
Screen geometry: normal / angular (45
Screen thickness, energy deposition, beam scattering
0 )
Broad band no
~10 fsec (skin depth) depth of focus, scattering, effective thickness, system simplicity, etc.
100 μm range minimum: 1 μm (fragile!) maximum: some 100 μm
Light scattering
Halo effects through scintillating volume
None
A. Lumpkin and M. Wendt LCWS September 26, 2011 7

• YAG:Ce screens, used at the A0 Photoinjector:
– The screens have nominally a 5 µm grain size and are coated at
50µm thickness on various metal substrates.
– Substrates are Al or SS and 1 mm thick.
– In the A0PI arrangement the scintillator was on the front surface of the substrate, and oriented at 45 0 to the beam direction.
– Powder screens are kindly provided by Klaus Floettmann (DESY).
• Observed Characteristics
– The response time is about 80 ns FWHM.
– There have been reports of saturation of the mechanism for incident electron beam areal charge densities ~10 fC /µm 2 .
• This effect can cause a charge dependence of the observed image size in addition to the low-charge, screen resolution limit.
A. Lumpkin and M. Wendt LCWS September 26, 2011 8

# of bunches X5 linear polarization Fit σ (pixel) Size ( μm)
YAG:Ce 1 none 5.67 ± 0.05
128.7
OTR 10 vertical none vertical
5.71 ± 0.04
129.6
5.49 ± 0.05
124.5
4.47 ± 0.09
101.0
• Both screen surfaces at 45 0 to the beam direction.
• Gaussian fits to the projected beam profiles of 10 images.
• Deduced YAG resolution term (page 6): 80 ± 20 μm
• YAG resolution, averaged using three measurement sets: 60 ± 20 μm
A. Lumpkin and M. Wendt LCWS September 26, 2011 9

2
• Scintillator screen resolution vs. thickness after applying corrections discussed on page 6.
350
Chromox,
Elettra,45 0
300
Powder Data
Crystal Data
250
Chromox,
APS/ANL,45 0
200
150
100
YAG:Ce,
A0PI,45 0
50
YAG:Tb,
BNL,0 0
YAG 5
μm grain size
0
YAG:Ce, single crystal
SCSS and Mainz,0 0
200 400
Scintillator Thickness (
 m)
600
A. Lumpkin and M. Wendt LCWS September 26, 2011
800
10

• System related
– YAG:Ce powder and crystal screen spatial resolution.
– Camera resolution and depth of focus.
– OTR polarization effects and OTR point spread function.
– Camera calibration factor.
– Finite slit size (if applicable) .
• Accelerator / beam related
– Beta star term in spectrometers.
– Macropulse blurring effects on energy spread , beam size, and beam divergence in OTR images.
A. Lumpkin and M. Wendt LCWS September 26, 2011 11

• 1 mm (X5) / 4 mm (X24) spaced slits, 50 μm wide
– Camera calibration ~30 μm / pixel.
• Depth-of-focus issues in extended field of view for 45 0 arrangement of the YAG:Ce scintillator screen
X5 X24
A. Lumpkin and M. Wendt LCWS September 26, 2011 12

• Application tool provides online emittance and
C-S parameter calculations to facilitate operations.
A. Lumpkin and M. Wendt LCWS September 26, 2011 courtesy
R. Thurman-Keup
13

• System related
– YAG:Ce powder and crystal screen spatial resolution.
– Camera resolution and depth of focus.
– OTR polarization effects and OTR point spread function.
– Camera calibration factor.
– Finite slit size (if applicable) .
• Accelerator / beam related
– Beta star term in spectrometers.
– Macropulse blurring effects on energy spread , beam size, and beam divergence in OTR images.
A. Lumpkin and M. Wendt LCWS September 26, 2011 14

• OTR can be used for beam
– profile / size
– energy,
– position
– relative intensity
– divergence
– bunch length
• Charged particle passing a media boundary (EM dipole).
OTR angular intensity distribution of a single charged particle
A. Lumpkin and M. Wendt LCWS September 26, 2011 15

• OTR single particle spectral-angular distribution:
– Ω spatial angle d
2
N
1 d
 d

– ω angular frequency
 e
2
 c
 2
1


 
2


 x
2

 x
2


2 y


2 y
2

E = 220 MeV
 x’, y’
= 0.2 mrad
–
N
1
– Θ x,y
# of photons radiation angle
– E, ħ, c, π constants
1.0
0.8
• Coherent spectral-angular distribution from a macropulse d d

2 d
N

 r 
, //
2 d d
2
 d
N
1

I
   
–
N # of photons from per unit frequency and solid angle (typ. 1 e -> 0.001 photons)
– r reflection coefficient
–
I interference function (double foil)
– F coherence function (can be non-linear)
0.6
0.4
0.2
0.0
-0.010
-0.005
0.000
Angle (radians)
0.005
0.010
A. Lumpkin and M. Wendt LCWS September 26, 2011 16

• New developed imaging station in collaboration with RadiaBeam, Inc.
A. Lumpkin and M. Wendt LCWS September 26, 2011 17

• Switchable assemblies to compare options.
– Still tweaking!
Option 1
Impedance
Screen
Option 2
Impedance
Screen mirror
YAG:Ce, plus Al on Si mirror
OTR, normal to beam, plus Al on Si mirror light screen
OTR,100 µm Al plus Al on Si mirror
OTR foil, 1µm plus 1µm foil at
45 0 beam
A. Lumpkin and M. Wendt LCWS September 26, 2011 18

• Optics focused on crystal location:
– gives superposition of focused OTR and defocused OTR source from mirror.
Single-Gaussian Fit
σ
1
= 16.2 ± 0.2 pix
Double-Gaussian Fit
σ
1
= 9.2 ± 0.1
σ
2
= 28.1
± 0.2
A. Lumpkin and M. Wendt LCWS September 26, 2011 19

• 1996: Lebedev evaluated OTR resolution.
• 1998: Castellano, et.al. points out the OTR PSF has a polarization feature.
– Calculated a 12 λ (FWHM) effect for the total width and 0.1 rad collection angle.
• 2007: FNAL & JLAB observed OTR / ODR polarization effects on the beam image size.
• 2007: Xiang, et.al. calculates PSF for OTR and ODR.
• 2010: OTR polarization on beam images reported by
FNAL A0 photoinjector staff at the BIW10.
• 2010: KEK OTR experiment uses a polarizer to analyze the details of the PSF (IPAC2010).
– In 2005 the experiment was performed without polarizer.
A. Lumpkin and M. Wendt LCWS September 26, 2011 20

• At the diffraction limit the image of a point source radiates a ring pattern defined by the OTR point spread function (PSF):
Source Lens Image f
2
  
PSF: convolution integral of

0
  m

2

2
J
1
 
]
2
Z
Point charge diffraction
OTR response a b
–
–
– 𝜃 = 𝑅 𝑖
/𝑎
  i 𝜃 𝑚 maximum acceptance angle
M

/ magnification factor
0.002
𝑅 𝑖 radius of the lens
0.0015
0.001
Cross section Example:
M=1, E=4GeV, λ=500nm courtesy C. Liu
5

10
4
2

0
10
5
1

10
5
0 1

10
5
Radius on image plane
2

10
5
A. Lumpkin and M. Wendt LCWS September 26, 2011 21

0.002
0.0015
0.002
0.0015
0.001
5

10
4
0.001
5

10
4
2
0

10
5
1

10
5
0 1

10
5
2

10
5
2
0

10
5
1

10
5
0 1

10
5
2

10
5
Radius on image plane
PSF dependence on beam energy
• Insensitive to the beam energy
Radius on image plane
PSF dependence on acceptance angle
• Sensitive to the acceptance angle
O
• Polarizer mitigates PSF imaging errors
– E.g. horizontal polarizer reduces vertical image size by 𝑦/ 𝑥 2 + 𝑦 2
0.002
– With a mask blocking rays of f 2 angle ≤Θ
1
(
   m

, the PSF will be:
  m

1

2


2
 
2
J
1
(
 
] 2
0.0015
0.001
5

10
4
2
0

10
5 courtesy C. Liu
1

10
5
Lens
0 1

10
5
Radius on image plane (m)
I
2

10
5
A. Lumpkin and M. Wendt LCWS September 26, 2011 22

• 14.3 MeV, M=1, λ=500 nm, θ max
=0.010, sigma =25 µm
• This version with convolutions implemented at FNAL.
x 10
-4 x 10
-4
-1 -1
Total PSF
0 0
Hpol PSF
1
-1
Vpol PSF
0
-1 x 10
-4
0 1 x 10
-4
1
-1
Y
0 1 x 10
-4
At Image plane
1
-1 0 1 x 10
-4
X
A. Lumpkin and M. Wendt LCWS September 26, 2011 courtesy
R. Thurman-Keup
23

-1 x 10
-4
0
1
-1
-1 x 10
-4
0
1
-1 0 1 x 10
-4
0 1 x 10
-4
-1
0
-150 x 10
-4
0
• 14.3 MeV, M = 1, λ = 500 nm, θ max
=0.010, σ = 25 µm
Convolution with Beam
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
Total
HorPol-HorProj
HorPol-VerProj
0.6
0.5
0.4
0.3
0.2
0.1
1
0.9
0.8
0.7
None
Total
HorPol-HorProj
HorPol-VerProj
-100 -50 0 x (
 m)
50
Hpol PSF
100 150
0
-150 -100 -50 0 x (
 m)
50 100
Original Sigma = 25 μm
Total PSF Sigma = 33.18 μm
HorPol-HorProj PSF Sigma = 38.01 μm
HorPol-VerProj PSF Sigma = 29.39 μm
150
1
-1 0 1 x 10
-4
A. Lumpkin and M. Wendt LCWS September 26, 2011 24

• 14.3 MeV, M = 1, λ = 500nm, θ max
= 0.010, σ = 10 & 50 µm
Convolution with Beam
1
0.9
0.8
0.7
0.6
None
Total
HorPol-HorProj
HorPol-VerProj
0.5
0.4
0.3
0.2
0.1
0
-50 -40 -30 -20 -10 0 x (
 m)
10 20 30 40
Original Sigma = 10 μm
Total PSF Sigma = 22.59 μm
HorPol-HorProj PSF Sigma = NA
HorPol-VerProj PSF Sigma = 16.63 μm
50
Convolution with Beam
0.6
0.5
0.4
0.3
0.2
0.1
0
-300
1
0.9
0.8
0.7
None
Total
HorPol-HorProj
HorPol-VerProj
-200 -100 0 x (
 m)
100 200 300
Original Sigma = 50 μm
Total PSF Sigma = 55.63 μm
HorPol-HorProj PSF Sigma = 58.49 μm
HorPol-VerProj PSF Sigma = 53.05 μ
A. Lumpkin and M. Wendt LCWS September 26, 2011 25

• OTR Perpendicular component has 15 % smaller profile.
– Beam measurements with a vertical stripe, optics generated.
Total Pol.: Left
Single-Gaussian Fit
σ
1
= 66.8 ± 0.3 μm
Vert. Pol.: Right
Single-Gaussian Fit
σ
1
= 55.1 ± 1.1 μm
10µm effect @ 55 µm
(Cal.: 5.3 µm/pixel )
A. Lumpkin and M. Wendt LCWS September 26, 2011 26

• KEK staff used vertical polarizer and small beam to observe PSF and suggested potential use of structure.
– Use PSF valley for profile measurements at the PSF limit.
*Legend reversed with respect to zero which included a constant background; b is the amplitude of the distribution; c is the distribution width; σ is the smoothing parameter dominantly defined by the beam size; and Δx is the horizontal offset of the distribution with respect to zero courtesy A. Aryshev
A. Lumpkin and M. Wendt LCWS September 26, 2011 27

• Estimation of OTR/COTR spectral effect for LCLS case.
8
6
4
2
0
0
16
14
12
10
OTR vs. COTR
OTR Rel. Intensity Model
CCD response
COTR with OTR gained up (3 keV)
COTR
(3 keV)
OTR~1/ λ 2
200
-UV-
CCD Resp.
400 600
Wavelength (nm)
800 1000
1.6
1.4
0.6
0.4
0.2
1200
0.0
1.2
1.0
0.8
A. Lumpkin and M. Wendt LCWS September 26, 2011 28

I
Y
• Reduction of COTR effects with 400x40 nm BPF, but need more sensitive camera than 40dB analog CCD to see remaining OTR.
COTR:ND0.5
COTR:400x40 nm
LSO: 400x40 nm
Y Y
X(ch) X(ch)
X(ch)
I I
A. Lumpkin and M. Wendt LCWS September 26, 2011 29

• Injector being installed with First beam expected in 2012.
40-MeV Injector electron gun booster cavities
3 rd harmonic cavity flat beam transform chicane deflecting mode cavity spectromet er magnet beam dump
1 st cryomodule test beamlines beam dump
Booster cavity 2 (from DESY and Saclay) installed in NML courtesy M. Church
A. Lumpkin and M. Wendt LCWS September 26, 2011
First cryomodule (from
DESY) installed at NML.
30

• Scintillator resolution terms should be characterized,
– Use normal incidence of beam preferred geometry to minimize depth-of-focus, effective radiation thickness in beam images.
• OTR polarization effects need to be elucidated
– Plan to optimize OTR PSF and optical resolution.
– Plan to use linear polarizers with OTR imaging for the perpendicular profile components at ASTA.
• Mitigate microbunching instability effects for profiling of bright beams.
– Plan to use 400x40 nm band pass filters and LYSO:Ce* crystals after bunch compression at ASTA to suppress expected diagnostics complications due to COTR.
• The future remains bright for imaging techniques!
*Lutetium Yttrium oxyorthosilicate (420 nm, violet), Cerium doped
A. Lumpkin and M. Wendt LCWS September 26, 2011 31