ESS Workshop “Bunch Shape Measurements” Bunch Length Detector Based on X-Ray Produced Photoelectrons Peter Ostroumov February 11, 2013

2 Content  BSM based on secondary electrons for CW beams  Motivation for X-BLD  Properties of X-rays produced due to the interaction of proton/ion beams with material  Bunch length detector based on X-ray produced photoelectrons –Principle of operation –Photocathode –Registration of photoelectrons  Experimental results, bunch shape of ion beams  Possible modifications of the detector for application in high-power accelerators

3 Bunch Shape Monitor Based on Secondary Electrons  Very well established technology for proton, H-minus, ion linacs  Requires a wire biased to -10 kV to produce and accelerate secondary electrons  Limited beam pulse length (~50 to 150  sec) and beam repetition rate to avoid destruction of the wire (for high-intensity beams)  Proton beam space charge and magnetic field can impact on electron trajectory – requires more studies for high current beams  About 12 years ago at Argonne we developed BSM for CW low intensity ion beams

BSM for CW Beams 4  Beam fundamental frequency = MHz, deflector frequency =97MHz  CCD Camera

BSM for Low Intensity CW Beams 5

Electron Beam Image, no RF 6

CW RFQ Off-line Commissioning, July-August

300 keV/u CW RFQ, Beam Characterization in July 2012  125 µA helium beam accelerated in CW RFQ  Bunch frequency = MHz  RF deflector frequency = MHz  Red –simulations, Blue- measurements with BSM 8

9 Motivation for X-ray Based BLD  Stripping of heavy-ion beams in high-power driver accelerators –Time focus of the bunched beam on the stripper results in the lowest longitudinal emittance growth –Heavy-ion beam density is high and measurement of the bunch time profile must be non-interceptive –A liquid Li stripper is planned in future heavy-ion linacs –Beam-stripper interaction produces significant amount of X-rays  X-BLD for application in high-power proton, H-minus accelerators –Does not require wire or target permanently inserted into the beam. A gas flow can be used or a solid “needle” can be dropped across the beam –Bunch time profile can be monitored on-line which is not possible using conventional methods

X-Ray Production Mechanisms by Bombarding Target with Ion Beam  2 mechanisms –Characteristic x-rays. –Brehmsstrahlung or "braking radiation"  Characteristic x-rays. –Electrons from higher states drop down to fill the vacancy, emitting X- ray photons with precise energies determined by the electron energy levels  Bremsstrahlung (braking) radiation –Charged particles (electrons, ions, protons) are suddenly decelerated upon collision with the metal target –Protons and alpha particles produce negligible bremsstrahlung radiation 10

11 Properties of X-rays from Target Potentially, picosecond-level resolution can be acheived Intensity of K-shell X-rays is 2 order of magnitude higher of all other X-rays For test purpose at low-intensity ion linac we use copper target For high-intensity beams: use, for example, xenon gas flow

10 MeV/u Beams on Copper Target  X-ray spectra excited by 10-MeV/u projectiles passing through a 965- mg/cm 2 -thick Cu foil, measured with a Si(Li) spectrometer.  The dashed lines indicate the positions of the Cu Kα and Kβ (single vacancy) diagram lines. 12

13 X-ray Based Bunch Length Detector: Principle of Operation  Main goal was to detect X-rays, produce photoelectrons and measure bunch time structure

14 X-ray Based BLD  General view of the X-BLD installed at the ATLAS facility. 1 – target translator, 2 – target, 3 – photocathode assembly, 4 – RF deflector, 5 – detection unit (chevron MCP), 6 – CCD-camera.

X-BLD Installed in ATLAS Beamline 15 CCD camera Phosphor screen Target, photocathode & Deflector

16 Photocathode Properties  The best response of the photocathode is with 1 kV X-Ray Back-surface secondary electron quantum yield for a 1020-Angstrem  CsI transmission photocathode

Photocathode 

Photocathode Substrate  Slit dimensions are 10x1 mm 2 18

Photocathode Assembly 19

20 Photocathode – Properties of Photoelectrons Energy distribution of the photoelectrons: Photoelectron energy distribution for CsI photocathode excited by X radiation with an energy of 277 eV. Photoelectron energy distribution for CsI photocathode excited by X radiation with an energy of 8080 eV. the both distributions are almost the same with a peak at 0.5 eV nearly the same width at half maximum. 80% of the photoelectrons are below 2 eV.

21 Trajectory of 10 keV Photoelectrons PhotocathodeElectron trajectories Collimator Deflector (focusing) plates Potential contours b) a) Beam X-ray source

22 Commissioning of the Detector  The slit downstream of the photocathode was too wide (1 mm) –Reduced to 0.15 mm  Stray magnetic field steers 10 kV electrons –Shielding against magnetic field  RF frequency of the deflector is 97 MHz and is driven by the accelerator master oscillator  Primary ion beam must be tuned to avoid losses on the walls –Can produce X-rays on the walls which increases the background  Chevron MCP –Very good amplification, up to 10 8 –Problem: Narrow dynamic range, Dynamic range depends from amplification. Large dynamic range MCPs are available but expensive ~$10K

23 Electron Beam Image on the Phosphor with no RF Applied Focused electron beam profile: Resolution is ~5 pixels = 0.4 mm  We installed a grounded slit (0.15 mm) downstream of the photocathode

24 Oxygen Beam (~1.6 MeV/u, 9 MeV/u, 0.2 to 0.5  A) Detector location ECR PII Booster

Experience with X-BLD  Main mission of the XBLD development was successfully accomplished: –We clearly see X-ray produced photoelectrons –Successfully used for bunch time profile measurements –Excellent properties of the secondary electrons from photocathode –No issues with electron beam optics  Beam current at ATLAS is very low, less than 0.5 µA that we were not able to use “point” source of X-rays  The resolution of our detector was limited by low energy beam size on the target and it is quite low: ~170 psec for 9 MeV/u beam 25

Possible improvements of the XBLD 26

Time resolution  X-ray source size must be small, below 1 mm  X-rays must be collimated to reduce “visible” source size  Transverse size of the X-ray emitter contributes to resolution 27 Beam X-ray source Photocathode

28 Electron detector Use a Spectrometer to Select X-rays Produced by Filling of K-shell Vacancies (8-10 keV)  Make sure that X-ray production takes less than 1 femtosecond

29 High-Power Proton Accelerators 0.15 x 10 mm slit grounded X-ray collimator Proton beam Gas jet, macroparticles, needles (pulsed) Nozzle design to produce a gas jet X-ray Spectrometer

Sub-Picosecond X-ray Streak Camera  Developed for detection of electron bunch length using synchrotron radiation 30

Next Steps  X-BLD with wire target and proton beam –Will greatly increase resolution, perhaps to 5-10 psec level even for low energy (3-5 MeV) beams –ANL hardware components will be available for future experiments  X-ray spectroscopy  Gas target  Falling needles or jet of macroparticles  Higher proton energies 31

32 Summary  We have developed and tested an X-ray based Bunch Length Detector (X-BLD) for application in ion and proton accelerators  The sensitivity of X-BLD is high enough to measure bunch length even for ion beams with quite low intensities such as 0.15  A  Temporal resolution of an X-BLD can be improved for high intensity ion beams by incorporation of an X-ray spectrometer into the device.  An electron beam detection based on MCP-phosphor system has low dynamic range and requires improvement. Secondary Electron Multipliers are proven to be better option.  High power proton and ion accelerators: –X-BLD can be applied for on-line monitoring of the bunch length. Requires either pulsed gas flow or dropping of micro-particles or “needles” across the beam –No thermal issues with the target