2002 ATF Newsletters

Jan | Feb - March | April - June | July - Sept | Oct - Dec

Dear Readers,

Another year of achievements passed by. The ATF's experimental program keeps on growing stronger and diversifying in various directions. Last newsletter reported the first interaction of electrons and guided laser beam in a plasma channel. Now we can report our first entry into plasma acceleration with some spectacular result: The first observation of the phase dependence of the focusing forces along the beam bunch, as reported below. The VISA experiment, a proof-of-principle experiment for the LCLS that provided a wealth of FEL physics results such as the shortest gain-length SASE, non-linear harmonic-generation and other results, is continuing with an ambitious new agenda.

We also keep on improving the ATF's performance. During the past few months the ATF went through a major shutdown for upgrade of electron beam and laser systems. This shutdown is now over (see Vitaly Yakimenko's brief report) with a significant step up in the ATF beam handling quality. The new new CO2 large bandwidth  preamplifier has arrived and is being installed. This is the last component necessary for the establishment of the terawatt picosecond CO2 laser, a device that is expected to usher the ATF into a new era of advanced acceleration and advanced source R&D. See also a report by Marcus Babzien about the generation of short pulses for the new CO2 system. Bob Malone reports on the computer control system upgrade, also nearing completion.

The 11th Advanced Accelerator Concepts Workshop will be held at nearby Stony Brook University in June 2004. A dedicated Local Organizing Committee, chaired by Marcus Babzien, is already hard at work preparing the workshop.

Best wishes for the holidays and a Happy New Year,

Ilan Ben-Zvi .  

Back to Top

We report the generation of plasma wakefields by a single relativistic electron bunch and study of the focusing component of the fields. The leading edge of the electron bunch excites a high-amplitude plasma wake inside the overdense plasma column and the acceleration and focusing wakefields are probed by the bunch tail. By monitoring the dependence of the acceleration and focusing forces upon the plasma density we approached the beam matching condition and achieved an energy gain of 0.6 MeV over the 17 mm plasma length, corresponding to an acceleration gradient of 35 MeV/m. We characterized the focusing variation along the plasma wake. We confirm the theoretical predictions for the phase offset between focusing and accelerating components of the wake.

The layout of the experiment is shown in figure 1.

The data taken by the electron beam spectrometer with, and without plasma, are shown in figure 2.

For closer study of the focusing/defocusing produced by the wake field we analyze figure 2 by taking energy slices. Energy slices are assigned a phase under assumption of a sinusoidal energy modulation with the maximum acceleration at 90 degrees. The analysis of sliced distributions reveals an interesting tendency. The profile of the slice taken close to the maximum and minimum energy extremes can be approximated well by a single Gaussian curve compatible in width to the initial (no plasma) image. Slices taken at the intermediate energy fit into the double-Gaussian approximation.

Figure 3 shows the longitudinal and radial fields as sine and cosine plots. Evidently, the radial field crosses the zero axis at the phase points where the longitudinal field reaches maximum or minimum.  For any intermediate amplitude of the longitudinal field we can identify two phases of the radial field different in polarity. This explains the observation of focused and defocused beamlets for each energy slice.

The experimental data points in figure 3 are reconstructed from the double-Gaussian approximations applied to the sliced energy distributions. The vertical scale is normalized divergence of the beam for either the focused component or the defocused component.

Our study shows that the electron focusing in the plasma wakefield is amplitude-correlated to the accelerating field. Furthermore, it provides evidence that the phase offset between the longitudinal and transverse components of the wakefield is pi/2, the same as in conventional RF accelerators. The conclusion is that plasma wakefield accelerators would allow certain phase range to provide both acceleration and transverse focusing of the particle beams.

Back to Top

Vacuum in the H-line reached middle -8 range permitting us to open valves to the rest of ATF beam lines on Wednesday, December 18. Most of the electrical connections were made and machine was ready for Radiation Fault Study. The ATF was informed by RCD about addition formal requirements on the morning of Radiation Fault  Study. We plan to conduct next attempt of the Radiation Fault  Study on December 30. Request for the running time for the scheduling purpose would be email after ATF is cleared to run (hopefully this year) and first beam study are carried out to confirm property of the beam that experimenters can use. 

One of the major tasks of this shutdown was a complete rebuild of the ATF's High-Energy beam line (H-line). This rebuild was a major undertaking. Its success is due to the untiring efforts of John Skaritka and a large team of designers and technicians from the NSLS. Very important was the hard work done by the riggers, masons and carpenters, under strict conformity to health and environmental issues. We are grateful for this strong support. Naturally, very hard and dedicated work was done by ATF staff and physics support staff on mechanical support, work control, ES&H and optical, electrical and control systems. The new line sports a most advanced design, including precision rail support, innovative alignment systems, remote control of quadrupole triplets for ultra-high precision beam-based alignment, all designed by John Skaritka. The electron beam diagnostics is also much improved as reported earlier.

We expect to enjoy improved beam brightness in all the ATF beam lines as a result of this upgrade. In addition, the upgrade cleared room to install a bunch compressor, intended for a new experiment: Electron Beam Pulse Compression Based Physics at the ATF, AE26 . This compressor will also play an important role in the VISA experiment. The compressor parts have been delivered to the ATF and are nearing installation.

Back to Top

With the installation of a chicane compressor designed and constructed at UCLA, the infrastructure at the BNL ATF is now capable of supporting advanced pulse compression and related SASE FEL experiments. These experiments are intended to address critical issues in beam compression, transport, and lasing short wavelength SASE systems such as the LCLS in an integrated system. Such an integrated approach may directly test the FEL community's experimental and computational tools for dealing with the plethora of collective effects in the SASE injector/FEL systems.

The pulse compression experiments will investigate the production of 0.2-nC pulses with 25-micron length, with emphasis placed on direct detection of coherent synchrotron radiation (edge radiation) near the entrance to the final bend magnet. After compression, linear transport through final bending systems has also been found (notably in the first round of VISA experiments), to have dramatic effects on the final beam phase space, and thus the SASE process.

Having studied ways to mitigate strong nonlinearities in the transport of ATF beamline 3, using an optimized linear lattice as well as sextupoles, we are now proposing a series of SASE FEL experiments using the existing VISA system. This set of experiments will emphasize the lasing action in chirped systems that are fully or partially compressed, which are of interest for creating ultra-short radiation pulses. Two distinct sets of experiments are planned:

1. An experiment using a fully compressed electron beam that achieves saturation in less than 3 meters out of the full 4 meter VISA undulator;

2. An experiment that uses the chirped and slightly compressed beams, that produces a chirped radiation pulse that can be shortened using gratings compression.

These experiments, which are undertaken by collaboration between UCLA, BNL, SLAC and the INFN, are described further at the joint web-site:

http://stout.physics.ucla.edu/visa/

Since the funding of both the compressor and VISA II in August 2002, significant progress has been made in both experiments.

Compressor:

— All the compressor magnets have been manufactured and tested at UCLA, and shipped to BNL.

— Final designs of supports and diagnostic sections are complete and under construction, leading to installation of the entire assembly in early 2003.

— The critical THz diagnostics based on coherent transition radiation and coherent edge radiation are now nearly finished in design, with construction beginning soon at UCLA and Univ. of Georgia (U. Happek’s group).

— More complete modeling of the bunch compression process using ELEGANT, TREDI, and a semi-analytical approach to coherent edge radiation are now being undertaken.

VISA II:

— The undulator vacuum has been verified, as well as the alignment of the undulator sections using optical fiducials.

— The optical diagnostics have been restored to working order, along with their alignment system.

— Wavelength spectrometer capabilities have been restored.

— The transport optics in beamline 3 have been studied with ELEGANT, with a more forgiving linear tune, and sextupole correction scheme identified.

— Start-to-end simulations of chirped beam FEL scenarios, using PARMELA/ELEGANT and GENESIS, are underway.

— First beam, using VISA I-like conditions, in the VISA undulator is expected in mid-January.

Back to Top

In our last newsletter we reported observing the highest energy gain (13%) ever obtained in an inverse free electron laser (IFEL) and a mysterious undermodulation occurring in our IFEL buncher, which would lead to less than ideal microbunch formation.  We have discovered that this undermodulation appears to be caused by the combination of an annular laser beam being delivered by the ATF laser and diffraction effects caused by an oversized laser beam propagating through the various optics used to focus the laser beam into the IFELs.  This causes the laser beam to be annular in shape at the location of the buncher, which means the electrons experience a decreased field when they pass through the center of the annulus as they propagate through the buncher.

Modification of the optics associated with the ATF Terawatt Amplifier should largely eliminate the problem of the annular beam being delivered to our optical transport line.  The best solution concerning the oversized laser beam propagating through our system is still being considered.  Options include repositioning some of our optics and/or using partially closed apertures to control the beam size.  Repositioning optics should not sacrifice laser power; however, it may introduce more aberrations on our beam.  Apertures will decrease the laser power somewhat, but we know from tests conducted on our optical transport system that this approach appears effective in preventing the annular laser beam at the buncher location.

As a result of a number of tests on our optical transport line, which were performed during the ATF shutdown period, we believe there are no other intrinsic problems with the individual optical components.  Thus, we should be ready to resume testing the fully-integrated STELLA experiment as soon as the shutdown in finished.

A preliminary comparison with the model was performed on the 13% energy gain spectrum.  Figure 1 compares the data energy spectrum with the model assuming a laser intensity inside the undulator of 0.5 TW/cm², a chicane phase delay of -30° away from the optimum phase corresponding to the best trapping of the accelerated microbunch, and optimum modulation by the buncher.  (The measured spectrometer resolution was used to generate the model plots.)  We see fairly good agreement in overall shape with the peaks of the spectra roughly agreeing with each other.  The model predicts slightly less acceleration and deceleration.

STELLA Report Figures 1 and 2.

Since the exact conditions for the data were not well known for this particular measurement, in particular the intensity and quality of the buncher laser field, it is risky to compare the model and data too closely at this point.  Various conditions were varied in the model, such the laser intensity in the tapered undulator and the amount of modulation in the buncher.  (Bear in mind, the model was not designed to simulate an annular laser beam in the buncher.)  However, these did not improve the agreement.

Thus, the agreement between the model and data is tantalizing close.  Hopefully we will be able to obtain even better agreement in our future runs.  The model predictions do show that every peak seen in a spectrum is important and means something.  The absolute energy calibration of the energy spectrum must also be as accurate as possible since 1-MeV shifts have significant impacts on the parameter values used in the model.

Back to Top

The X-ray spectrometer, designed for observation of nonlinear Compton scattering, consists of a Silicon/Molybdenum (Si/Mo) multilayer crystal, and a two-dimensional position sensitive detector.
Fig. 1 shows a schematic view of the spectrometer.

The multilayer crystal, which has 45-pair periodic structure with thickness of each layer 20A, is curved cylindrically with the curvature of 4.8m as shown in Fig. 2.

The curvature makes incident angles of X-rays with respect to the surface vary from 25mrad to 10mrad, corresponding to the X-ray energy satisfies the Bragg's law from 6keV to 15keV.
Thus energy spectrum of X-rays appears as distribution of reflection angle along the X-axis in Fig. 2.
On the other hand, reflection angle along the Y-axis directly represents emission angles of X-rays from the collision point. Therefore, energy and emission angle of X-rays are measured simultaneously by detecting reflected X-rays using a two-dimensional position sensitive detector.

We use a CCD based X-ray detector (MarCCD X-ray detector, Mar USA Inc.) for two-dimensional detection. It consists of a scintillation screen with 162mm diameter, a fiber optic taper which guides scintillation lights, and a CCD detector. The CCD has 2k x 2k pixels and each pixel provides a 16-bit readout. The pixel size corresponds to a 80microns x 80microns square on the scintillation screen.

 
Reflectivity of the prototype multilayer crystal is measured using X-rays of 12keV. Result is shown as a function of incident angle in Fig. 3. The 1st, 2nd, and 3rd order reflections are seen. By fitting the reflectivity curve, thickness of the Si layer and Mo layer will be obtained.
 

Back to Top

Upgrade of the ATF control system continues to make substantial progress and will soon be completed. The last major block of work (migration of ATF application programs) now stands at ~75% completion. Once these final pieces of software are converted and tested, the new system will be ready to begin deployment for production work. (A summary of software conversion tasks and the status of each is appended below.) Extensive use was made of the time available during the recent ATF shutdown; This block of quiet time (for the control system, not ATF in general!) proved valuable in testing software components.

This has been a major effort requiring, essentially, a wholesale change in every major hardware and software component of the control system:

VAX hardware --> Intel Pentium-based hardware

CAMAC serial crate controllers --> Ethernet crate controllers

CAMAC serial highway driver --> (not needed; functions handled in Ethernet crate controller)

CAMAC serial highway --> Fast Ethernet

Operating system: VMS --> Linux

Fortran,C --> C++

Vsystem / VMS --> Vsystem / Linux

As reported earlier, all the new hardware is in place and is operating in parallel with the existing VAX-based system.

While there are far too many technical details to report in this newsletter, here are new highlights which may interest ATF users:

* System backup and recovery: Although the new control system has reliability features such as RAID storage and redundant power supplies, there is always the concern as to how to rebuild the system should the software suffer irreparable damage. Such "bare metal" recovery can be very difficult since merely re-installing the operating system will not restore the many custom settings and patches needed to restart ATF operations. ATF has addressed this issue by purchasing a commercial product (LoneTar) which makes complete system backups and bare metal boot kits. The result is a complete backup kit which contains everything needed to restore from a bare metal state.

* System stability: The new control system has demonstrated excellent stability and continues to equal the VAX system in reliability.

* Throughput: As reported earlier, every key operating parameter (CPU speed, storage, etc.) has been improved by at least an order of magnitude. ATF operators and users will see noticeable improvement in system response. Also, there is room for growth: the present machine is configured for only 25% of its full capability.

Tests during this shutdown period uncovered some inefficiencies in the ATF network libraries which have been corrected by rewriting some key functions to exploit the multithreaded Linux environment. The resulting software is much better at handling network data and, for example, has no problem in supporting both ATF frame grabbers concurrently. This same technique will be used to control and monitor the beam-based alignment system which was installed during this shutdown.

* User interface: As mentioned at the last User Meeting, ATF users can expect the same look and feel of the control system since all of the original displays were ported without change. We are pleased to report this is still the case: the testing during the shutdown did not uncover any conditions where displays had to be changed to accommodate the new system.

----------

Summary of major software conversion work to date:

1) Low level libraries: ~15K lines code, 100% completed

A) Support for CAMAC hardware modules:

- Bi-Ra model 3224 24-bit output register

- DSP Technologies model 3016 16-channel DAC

- DSP Technologies model E500 stepping motor controller

- DSP Technologies model IO-612 24-bit parallel I/O register

- DSP Technologies model PR-612 TTL output register

- EG&G model DA 4-16 4-channel DAC

- Hytec Electronics model 1365 Ethernet crate controller

- Kinetic Systems model 3072 48-bit output register

- Kinetic Systems model 3074 24-bit relay output

- Kinetic Systems model 3075 16-bit relay output

- Kinetic Systems model 3095 24-bit discrete output

- Kinetic Systems model 3116 16-bit DAC

- Kinetic Systems model 3291 CAMAC diagnostics card

- Kinetic Systems model 3296 CAMAC diagnostics card

- Kinetic Systems model 3344 RS-232 communications card

- Kinetic Systems model 3388 GPIB communications card

- Kinetic Systems model 3472 48-bit input gate

- Kinetic Systems model 3516 32-channel scanning ADC

- Kinetic Systems model 3623 6-channel counter

- Kinetic Systems model 3640 up/down counter

- Kinetic Systems model 3655 timing generator

- Kinetic Systems model 3821 RAM memory

- LeCroy model 2249SG charge ADC

- LeCroy model 2249W charge ADC

- LeCroy model 2259B peak sensing ADC

- Jorway 45-S1 input register

B) ATF database support library:

- Database startup/shutdown handlers

- Database protection handlers for error trapping

- Put/get handlers for binary, double precision, integer, real, string and time channels

- Functions for channel value type conversions

- Channel search functions

- Asynchronous change notification handlers

- Diagnostics/tracing functions to

C) ATF General functions libraries:

- date/time functions

- asynchronous signal setup/handling

- I/O stream management

- fatal error handlers

- general diagnostics

D) ATF Network library:

- Socket creation/management

- Socket stream I/O management

- Buffer management

- Error handling

2) Operator graphic user interface: 100% completed

- ~800 displays containing ~24K control / display items

3) Databases: 100% completed

- ~10K channels

ATF Application Programs: ~55K lines code, 75% completed

- Bruker magnet power supplies

- Stepping motor controllers

- Ganged magnet power supplies ("super-elements")

- Pop-in monitor controls

- Laser shutter controls

- Darlington magnet power supplies

- RF system control and readbacks

- Video camera switching

- Correlation plots

- Scanning ADCs

- Frame grabbers

- Quick data logging service

- Database socket servers (support for Mathcad, C++, LabView, Expect)

- Emittance measurement

- Chipmunk radiation monitoring

- Data acquisition synchronized to facility rep-rate

- Cathode cleaning application

- Magnet degauss service

Back to Top

The testing of the CO2 short pulse upgrade is now beginning. Optics for delivering the YAG pulses to the CO2 room are in place, and have been aligned using a green HeNe laser. Because of the installation of the new CO2 large bandwidth  preamplifier, the wall between the CO2 main and amplifier rooms has been opened. This required some administrative controls to be developed before beam could be delivered to the area. All safety precautions and documentation are now ready, and high power beam is being delivered.

The new three-stage optical scheme for generating few ps 10 micron pulses with the YAG is shown in the linked sketch. The first stage utilizes the spent 1 micron YAG pulse from the second harmonic compressor in the YAG room. This energetic pulse of about 7 mJ in 15 ps switches the carbon disulfide Kerr cell to reduce the 10 micron oscillator pulse duration and energy by a factor of about 10,000. The final two stages utilize the low energy 3 ps green pulses from the YAG for controlling germanium semiconductor switches. The lower available green energy requires the spot size on the germanium switches to be smaller than previously, necessitating the first stage reduction in 10 micron pulse energy to prevent damage. With one semiconductor switch operating in transmission, and the other in reflection, we expect to have control of the 10 micron pulse duration down to about 3 ps.

The first tests involve measuring contrast of the system to the level required to cross correlate the pulse duration using the two semiconductor switches. Performing the cross-correlation is the simplest demonstration of proper switching. Later, the pulse can be sent to the preamplifier for amplification to the level needed to use the second harmonic generation autocorrelator.

Back to Top

With the emerging significance of a magnetized electron beam (i.e., angular momentum dominated beam) for the flat-beam source development for linear colliders and electron cooling projects, it becomes much more interesting in revealing the details of the magnetized beam, e.g., angular momentum, in its natural coordinate system, i.e. a polar coordinate system rather than in the usual Cartesian coordinate system. For that purpose, we used a tomography technique to reconstruct the phase spaces in Polar coordinate system and thus one can apply it to improve our understanding of a magnetized beam. In July and August, we did the preliminary beam test for a non-magnetized beam (30 pC charge). Figures 1a-1b (see link below) show the transverse phase space tomography in Cartesian system. Figures 1c-1e show the converted phase spaces in polar coordinate system. We can see from the figure 1e that the angular velocity is cancelled for such a non-magnetized beam, which agrees with PARMELA simulation. For the next step, we will carry out an experiment for a magnetized beam. Figure 2 shows the PARMELA simulation. It shows that a significant angular momentum is created when employing 560 Gauss magnetic field on a cathode. Ideally the angular velocity is proportional to the magnetic field and radius of the beam. But in Figure 2 it shows that angular velocity has an angular spread and also the beam magnetization is partially lost in larger radius. Following on this success, we will investigate the mechanics of the spread of the angular velocity and the magnetization loss by using the phase space tomography in polar coordinate system. This experiment is on the progress.

     Click here for figures.

  The past several months have been rather busy with ESH issues at ATF:

- With much patience & cooperation from the Physics ESH group (Mike Zarcone, et al), the transition from NSLS to Physics Dept continues;

- The H-line upgrade proposal was presented by Ken Batchelor to the BNL ESH committee

(LESHC) on December 3, and subsequently approved with conditions imposed by the LESHC;

- Painted lead bricks were used to replace the bare lead bricks that covered the old H-line components, in an effort to minimize the risk of future contamination and associated costs;

- The CAD interlock group successfully performed their first recertification of the ATF radiation interlock system recently, which is an important step for ATF to return to operational status;

- Also transferred and recertified were the radiation chipmunks;

- CO2 preamp installation work proceeds following safety analyses, with laser interlock modifications awaiting NSLS involvement;

- The RHIC/AGS users center continues to provide a great service by processing incoming ATF experimenters;

- The first Tier 1 inspection of ATF under the supervision of the Physics Dept was performed recently, with no major problems.

Shortly, radiation fault studies will be performed to verify the sufficiency of the new H-line shielding configuration. Upon successful completion of this, ATF will be allowed to resume linac operations

Back to Top

Last Modified: December 3, 2007
Please forward all questions about this site to: Vitaly Yakimenko

One of ten national laboratories overseen and primarily funded by the Office of Science of the U.S. Department of Energy (DOE), Brookhaven National Laboratory conducts research in the physical, biomedical, and environmental sciences, as well as in energy technologies and national security. Brookhaven Lab also builds and operates major scientific facilities available to university, industry and government researchers. Brookhaven is operated and managed for DOE’s Office of Science by Brookhaven Science Associates, a limited-liability company founded by Stony Brook University, the largest academic user of Laboratory facilities, and Battelle, a nonprofit, applied science and technology organization.

Privacy and Security Notice  | Contact Web Services for help