Berkeley Lab

ATAP News, October 2017


Researchers in our Berkeley Laser Accelerator Center and Accelerator Modeling Program, working with international colleagues, have validated a way to both circumvent a source of errors and improve performance when using an important class of computer models to predict the performance of laser-plasma mirrors. The LBNL-led US Magnet Development Program set a new record — by a dramatic margin — for the current density in magnets made with high-temperature superconductor. We’ve continued contributing to the future of photon science with deliverables for LCLS-II and ongoing research in anticipation of the ALS Upgrade, while our ion-beam expertise continues to enable improvements in magnetic resonance spectroscopy.


With the advent of petawatt class lasers, the very large laser intensities attainable on target should enable the production of intense high-order Doppler harmonics from relativistic laser-plasma mirror interactions. At present, modeling these harmonics with particle-in-cell (PIC) codes is extremely challenging, as it requires accurate description of tens to hundreds of harmonic orders on a broad range of angles.

In a new study published in the journal Physical Review E, BELLA Center graduate student Guillaume Blaclard, with colleagues from ATAP’s Accelerator Modeling Program and from CEA Saclay in France, describes effects that can potentially degrade the simulation results — and validates a new modeling method that could circumvent the problem and improve performance too.

A potential source of error

Standard finite-difference time-domain (FDTD) solvers for Maxwell’s equations — the solvers employed in most of the particle-in-cell (PIC) codes that are widely used in accelerator modeling — induce numerical dispersion when dealing with waves in a vacuum. This numerical artifact can in turn induce a spurious angular deviation into harmonic beams. (For another example of studies and mitigations of a different numerical artifact, see “Avoiding the Wave that Wasn’t There,” ATAP News, December 2016.)

The team studied this effect extensively and developed a simple model based on Snell-Descartes’ law that allowed them to finely predict the unphysical angular deviation of harmonics depending on the spatiotemporal resolution and the Maxwell solver used in the simulations.

The new model demonstrates that the mitigation of this numerical artifact with FDTD solvers mandates very high spatiotemporal resolution. This precludes realistic three-dimensional simulations even on the largest computers available at this time; the computing resources needed to arrive at a solution are unreasonable.

Maxwell solvers compared

Angularly resolved spectra for different Maxwell’s solvers in the frequency range of ωn/ω0 = 10 to 35, where ω0=2πc/λ0, λ0 is the laser wavelength, and c is the speed of light, for a spatial resolution of λ0/70 and a time resolution fixed by the Courant condition of the Maxwell field solver. The laser incidence is θi = 55◦. The white solid line is the curve expected from the refraction model. For two types of finite-difference solvers, as shown in (a) and (b), the numerical refraction index and angle of refraction are positive. For a pseudo-spectral solver with finite leapfrog time integration (PSTD), shown in (c), the numerical index and thus the refraction angle are negative. Panel (d) shows a dispersion-free spectrum obtained with the pseudo-spectral analytical time-domain (PSATD) solver.

A way around the problem

However, the new study also shows that nondispersive pseudospectral analytical time domain (PSATD) solvers can considerably reduce the spatiotemporal resolution required to mitigate this spurious deviation. Pseudospectral methods, which existed for decades, have enjoyed a renaissance in recent years as techniques for adapting them to highly parallel supercomputers were developed by ATAP researchers. PSATD solvers should enable accurate 3D modeling with a realistic amount of supercomputing time.

PSATD solvers have been recently benchmarked in 2D against Doppler harmonic generation experiments performed at CEA Saclay. These benchmarks demonstrated that PSATD solvers can reproduce experimental features with high fidelity and with much lower resource-to-solution than standard codes.

The benefits of these solvers have also been recently quantified for electrons emitted on plasma mirrors using 3D PIC simulations run at very large scale. These results showed that massively parallel PSTAD solvers require two orders of magnitude less resource-to-solution in 3D compared to standard codes.

To learn more…

G. Blaclard, H. Vincenti, R. Lehe, J.-L. Vay, “Pseudospectral Maxwell solvers for an accurate modeling of Doppler harmonic generation on plasma mirrors with Particle-In-Cell codes”, Phys. Rev. E 96, 033305 (11 September 2017),

A. Leblanc, S. Monchocé, H. Vincenti, S. Kahaly, J. -L. Vay, and F. Quéré, “Spatial properties of high-order harmonic beams from plasma mirrors: A ptychographic study”, Phys. Rev. Lett. 119, 155001 (13 October 2017),

H. Vincenti, J.-L. Vay, “Ultrahigh-order Maxwell solver with extreme scalability for electromagnetic PIC simulations of plasmas,” submitted for refereed publication; (26 July 2017).


The ATAP-led US Magnet Development Program (MDP) has set a current-density record for accelerator-style magnets made from high-temperature superconductor (HTS). The 3.3-tesla magnetic field quadrupled the HTS performance achievable just two years ago and was half again what could be achieved earlier this year.

Using a coil fabricated this summer by ATAP’s superconducting magnet program, the researchers built and tested a magnet made from Bi-2212 wire. The “racetrack” coil, made of 17-strand “Rutherford-style” cable, carried a record 8.2 kA while generating a peak field of 3.3 T, quadrupling performance of a dozen coils made before 2015 and representing a 60% increase over two coils made and tested in 2016 and earlier in 2017. The overall quench current density in the cable was 730 A/mm2 and the wire’s engineering current density was 930 A/mm2, which are practical current densities for applications.

Realizing the promise of HTS

High-temperature superconductor, whether used as inserts to augment traditional superconductors or (as in this case) as the principal conductor, has long offered tantalizing prospects to the developers of magnets for high-energy physics and other uses. The attraction is not their ability to superconduct at relatively high (though still cryogenic) temperatures—in fact, the work is done at the usual liquid-helium temperatures—but their high-field potential.

Increasing the field of an electromagnet calls for more current, but the wire and cable can only handle so much and remain superconductive. Traditional conductors, such as niobium-titanium (NbTi) and the higher-field niobium-three-tin (Nb3Sn) materials that are now coming into use, have a critical current density that decreases rapidly as magnetic field increases. The critical current density of Bi-2212 decreases much more slowly as the field goes up. The engineering current density of this coil is expected to remain above 500 A/mm2 even at 20 T — far above the magnetic fields achieved thus far in accelerator-style magnets.

Thanks to a multi-institutional, public/private sector team effort at all levels from materials through wire and cable fabrication, racetrack coil RC-05 achieved critical current an order of magnitude greater than its predecessors of several years ago and half again that of its immediate predecessors in the current effort.

This high critical current density, together with new magnet designs such as the canted cosine theta technology also being explored by MDP, makes it possible to envision 20-T-class accelerator magnets for future high-energy colliders, such as an energy upgrade of the Large Hadron Collider or the notional Future Circular Collider. “This is great news not only for high-energy physics, but for spinoff applications throughout science, wherever high-field magnets are needed,” said ATAP Division Director Wim Leemans. Examples include 25 T solenoids for >1 GHz nuclear magnetic resonance (NMR) magnets.

Potential long-term implications are numerous. “Our interests include expanding all the frontiers of what’s possible,” Leemans adds, “including the cost-effectiveness and feasibility frontiers as well as absolute performance. “One of our overarching interests is bringing the accelerator to the application, and higher fields can mean smaller machines as well as new capabilities.”

According to Prof. David Larbalestier, Chief Materials Scientist of the National High Magnetic Field Laboratory (NHMFL) at Florida State University and Director of its Applied Superconductivity Center, “The parameters and performance of this coil and solenoidal coils made at NHMFL show that Bi-2212 is now a high-field conductor, ready for magnets that can enable superconducting magnet applications impossible with any other superconductor.” Looking beyond high-energy physics, NHFML has an application interest of its own in mind: developing 1.3-1.6 GHz NMR magnets using this round, multifilamentary, twisted, fine filament and macroscopically isotropic wire.

Click for larger version
Engineering a top-performing superconducting magnet requires mastery of seemingly fractal detail, from the overall design down to the formulation of materials. Superconducting material (powdered, in this case) is made into filaments, which are formed into wire, which is then woven tightly into a flat cable.

Partnering is key to the field

The record-setting magnet was the work of many hands. Dr. Soren Prestemon, head of ATAP’s superconducting magnet program and Director of the U.S. Magnet Development Program and the Berkeley Center for Magnet Technology, described it as “an example of what is possible when U.S. national lab, universities, and private industry come together to push technologies.”

The LBNL-fabricated coil coil was heat-treated at the NHMFL. The wire was made by Bruker OST LLC in New Jersey from a precursor Bi-2212 powder, which in turn had been manufactured by nGimat LLC in Kentucky using an innovative nanospray technology. Bruker OST LLC and nGimat LLC were supported by the U.S. Department of Energy’s Small Business Innovation Research (SBIR) program, working closely with the teams at LBNL and NHMFL, under the framework of the MDP.

“I’m extremely grateful for the dedicated and talented teams at LBNL, NHMFL, Bruker OST, and nGimat,” said Dr. Tengming Shen, the ATAP researcher who led the effort. The record-setting achievement is a culmination of his Early Career Research Program (ECRP) work, which began in 2012.


Measuring particles in the NEG/ion pump Berkeley Lab contributions to Linac Coherent Light Source II, a free-electron laser being built at SLAC by a partnership of SLAC and Berkeley Lab, Fermilab, Jefferson Lab, and Argonne, has seen continued progress in areas that include soft- and hard-X-ray FEL undulators and the VHF gun and low-energy beamline for the injector source.

SXR and HGVPU Undulator Production

Another pair of soft-X-ray undulators (SXRs) arrived at SLAC in September, making a total of six units delivered from vendors Keller Technology Corporation and Vacuumschmelze GmbH. The LBNL team continues to work with the vendors and with SLAC to monitor performance of the devices so that labor-intensive tuning is minimized and the installation schedule is maintained. To facilitate tuning, magnet modules are now delivered to LBNL for magnetic measurements and optimization before being shipped to a vendor for integration.

The HGVPUs (horizontal gap vertically polarized undulators) for the hard-X-ray beamline are moving through the production process as well. The first two sets of production magnetic modules built by Neorem Magnets Oy in Finland have been measured at LBNL and shipped to Keller Technology Corporation and Motion Solutions Co. LLC for integration of the magnetic and mechanical assemblies. These vendors are well advanced in their preparations and have assembly lines in place. A first article from one vendor is expected in late October, with the second following in November.

The pre-production HGVPU has undergone environmental testing and will be shipped to SLAC for long-term tests, which will exercise the movement of the adjustable gap and make periodic measurements to determine if the performance is impacted by the gap changes.

Two HGVPU production girders (blue) and one pair of strongbacks (long rectangular silver blocks atop the left girder) in the assembly area at Keller Technology await integration with the measured and optimized magnetic modules that will be shipped from LBNL. Our engineers have worked closely with each vendor and have visited their sites to assist in preparation and to verify that the processes developed here are being properly employed.

Injector Source Fabrication and Particle-Free Assembly

A vacuum leak found in the VHF gun assembly has been repaired, and the assembly has been fully re-cleaned using particle-free procedures and rebuilt; peripheral components are now being installed.

All components for the low-energy beamline downstream have been delivered to LBNL and subjected to particle-free cleaning. The buncher cavity and its power couplers have been tuned and matched, and final components are being mounted before bakeout.

A substantial documentation package describing the final as-built hardware is being finalized by the QA team, to be delivered together with the equipment. Delivery to SLAC, marking the completion of injector source production, is expected soon. The hardware will be installed in the SLAC linac enclosure in preparation for early injector commissioning, which is planned to start in January 2018. Members of the LBNL team will participate in the installation and commissioning.

The buncher cavity center wall and outer cells are shown at a vendor, ready for final brazing. The completed assembly is now at LBNL for frequency tuning, cleaning, and installation in the low-energy beamline. ATAP’s Fusion Science and Ion Beam Technology Program applied the titanium-nitride coating to this ceramic “break” or insulator in the integrating current transformer of the LCLS-II low-energy beamline. A combined nonevaporable getter (NEG) and ion pump for the VHF gun undergoes particulate measurement. Extreme cleanliness is a must to avoid potential contamination of the superconducting linac cavities in LCLS-II.

Gun Amplifier Production Making Progress

LBNL engineers again visited R&K Co. Ltd. in Fuji City, Japan, the vendor for the gun RF power amplifier. The meeting was to discuss technical details of the controls and operations of the amplifier, and to plan for acceptance tests to be completed in November.


Based on an article in the September 27, 2017 issue of ALS News.

In the last year since the ALS Upgrade Project (ALS-U) received approval of mission need (CD-0) from the U.S. Department of Energy, I’m happy to report that we’ve made a lot of progress. I’d like to share some updates with you as well as offer opportunities for our users to engage in the next phases of the project.

The ALS-U Project team has been building momentum as we set up the project, develop the conceptual design, and carry out key R&D. A number of Berkeley Lab divisions are contributing to the efforts. Project team staff include members of ALS, Accelerator Technology and Applied Physics, Engineering, and Facilities, and we have received additional assistance from various operations divisions. The enthusiasm and support we’ve received from across Berkeley Lab for this project is a true testament of the “team science” spirit that E.O. Lawrence himself engendered and upon which the Lab was founded.

In addition to internal expertise, we’ve benefited from the input of external advisory committees. This past summer, we held the inaugural meetings of our Machine Advisory Committee (MAC), which is charged with providing advice on the technical design and R&D activities related to the accelerator systems and insertion devices, and Experimental Systems Advisory Committee (ESAC), which is providing advice on the technical design and R&D activities related to the beamlines. Overall, the feedback was very positive. Mentioned in particular were the quality of the team and the progress of the R&D as well as conceptual design. The ALS-U team received many detailed comments that will help strengthen our efforts moving forward. In addition, the ALS Scientific Advisory Committee (SAC) continues to provide important guidance on scientific directions.

In January, we held a workshop on “Solving Scientific Challenges with Coherent Soft X-Rays.” The event convened more than 170 U.S. and international scientists to discuss early science enabled by ALS-U. The recently released report contains a number of exciting opportunities presented by ALS-U’s new capabilities that address longstanding scientific challenges.

As we move into the next phases of the project, including the selection of new and rebuilt beamlines that will be part of the project scope, it will be critical to get input from ALS users. Over the next few months, we’ll be soliciting that input primarily through online user forums and crosscutting beamline reviews.

I look forward to continuing to work with you as we design and build the future of ALS.


ATAP’s Hand in Squeezing Microwaves for More-Sensitive Magnetic Resonance Spectroscopy

Conceptual illustration of magnetic resonance
A. Bienfait, Univ. Paris-Saclay
A multi-institutional effort led by Dr. Audrey Bienfait of the University of Paris-Saclay has demonstrated a “quantum squeeze” technique that games the Heisenberg uncertainty principle for more-sensitive magnetic resonance spectroscopy. The work is the subject of a new article in the journal Physical Review X.

Researchers led by Dr. Thomas Schenkel in ATAP’s Fusion Science and Ion Beam Technology Program, a co-author of the paper, conceived of, designed, and made the special spin material (bismuth-doped silicon-28, a silicon isotope with no nuclear spin of its own) that enabled this study.

To learn more…

MDP, BACI, AMP Refresh Web Presence

The US Magnet Development Program (MDP) announces the launch of its website,, while the Accelerator Modeling Program and the Berkeley Accelerator Controls and Instrumentation Program — heirs to the legacy of the Center for Beam Physics — have pages of their own now on the ATAP website. We invite you to visit and learn more about their goals and achievements.

k-BELLA Workshop Report Available

The report of the Laser Technology for k-BELLA and Beyond Workshop is available online.

The topic of the workshop, held at LBNL May 9-11, was near- and long-term technology prospects for ultrafast lasers that could operate in the multi-kW to even tens-of-kW average power range. Such laser performance is needed for k-BELLA, further stepping stones to a laser-plasma accelerator relevant to high-energy physics, and spinoff benefits en route.

Memorandum of Understanding for High-Luminosity LHC Accelerator Upgrade Project

A formal Memorandum of Understanding between Fermilab and LBNL on the High-Luminosity LHC Accelerator Upgrade Project has been signed by both Laboratory Directors. The MOU defines the management structure, LBNL roles, and commitment to safety and quality.

The Fermilab-led AUP is the framework for US efforts in the overall HL-LHC project, which will greatly increase the luminosity of the CERN collider. The AUP was reviewed recently and given a Critical Decision 1/3a recommendation. The review found that it is thoroughly ready for the next phase — work toward the Critical Decision 2 milestone of a project baseline and a robust preliminary design — and that long-lead-time procurement of conductor should proceed.

Specific LBNL Statements of Work will be released by the AUP office. More than 12 years of work by ATAP as a partner in DOE-OHEP’s LHC Accelerator Research Program (LARP), with critical support and partnership with the Engineering Division, has resulted in two major roles in fabrication of interaction-region quadrupole magnets.

(but come score some points for yourself and our division anyway)

Inclusion Insights game cards can be posted as conversation starters, as Ina Reichel has done at her office
As a follow-up to the talks given by Dr. Steve Robbins on July 26-27, 2017, the Diversity and Inclusion Office (DIO) has launched a Labwide inclusion program called “Inclusion Starts with a Conversation.” DIO, along with the Diversity and Inclusion Council, invite all Lab employees to watch the inclusion videos and play the “Inclusion Insights” game. If you have attended one of Steve’s talks, you know that you can expect an experience that’s fun and and engaging as well as informative.

Each month, a set of short inclusion videos (2-8 minutes each; total no more than 10 to 15 minutes) will be announced in Today at Berkeley Lab and released on Accompanying each set of videos is a short game question that you can answer if you’ve watched the videos.

When you correctly answer the question, you’ll receive a playing card, which you are encouraged to display at your workstation. In keeping with the theme of “inclusion starts with a conversation,” when people ask you about the card, engage!

You get a point for each card, plus 5 bonus points if you were among the first 10 people to correctly answer each question in that round. Each Division at the Lab will also receive 10 bonus points for each facilitated group discussion that it hosts.

An Inclusion Conversation with LBNL Director Mike Witherell is the prize for the top ten point earners every 3 months (each individual can earn this prize only once). The overall top 3 point winners will also receive a prize, as will the Division with the most per-capita points. Ultimately, though, everybody who plays is a winner — and so is the whole Lab, because when more people know and care about inclusion, we all benefit.

ATAP will host a facilitated discussion for the November videos on Wednesday, November 8th from noon to 12:30 in the 71-264 conference room. Feel free to bring your lunch. If you would like to be added to the meeting on calendar, please email Ina Reichel ( If your group would like to have an additional facilitated discussion either during one of its regular meetings or a special meeting, Dr. Reichel can arrange that as well.

Let’s start the conversation!


Wildfire or Other Emergency Evacuations: Would You Know What to Do?

Berkeley fire, 1923
Northside Berkeley, 9/17/1923: from a silent-era newsreel via Wikipedia.
The catastrophic fires in Napa and Sonoma Counties earlier this month brought up a question that should be on everyone’s mind: what should we do if there is a fire near Berkeley Lab?

Some of us went through this on August 2 during a precautionary evacuation. Those who lived and worked in these hills in 1991 — or 1970 — know that it is not an idle question. And then there was 1923, when a conflagration driven by the Diablo winds reached the intersection of Hearst and Shattuck. The Laboratory takes a variety of precautions to reduce its wildfire risk, and of course has its own fire department, but a “Diablo wind” is always a reminder that such threats are only mitigated, never eliminated.

Announcements of whether and how to evacuate

When a wildland fire is expected to impact the Lab, we will be notified through our Building Emergency Teams, building Public Address Systems, and the Lab Alert applications on our phones. The instructions may be to shelter in place or to evacuate the site, by vehicle or on foot.

Ready, steady, go

If there is time:
• Safely shut down your equipment.
• Gather your essential items (keys, cell phone, laptop, medications, wallet and ID, etc.). You may not be returning anytime soon — nor to an intact building.
• If you have a “go bag” for emergencies, ready with survival items such water, snacks, clothing, flashlight, etc., get it ready.

Leave only when your building is instructed to do so. This reduces traffic congestion. Evacuations may be called by zone (see the blue Site Map tab on the Emergency Guide flip charts) so know your zone! It is useful information in many emergencies. Shown below is a zone map with notes on principal ATAP buildings, taken from the Lab’s Emergency Guide.

emergency-zones map
Click for larger version
Zone 1 – 71 complex
Zone 2 – 88
Zone 3 – 46 complex, 47, 53, 58 complex, ALS buildings
Zone 4 – 77 and 77A

How to actually evacuate

• Bicycles can be a fast and efficient mode of evacuation.
• If you have a car, offer rides to others who do not have cars on site.
• As the fire approaches, the evacuation mode may change — be prepared to pull over, park , and continue on foot if necessary.
• You may be instructed to evacuate on foot. Have comfortable, sturdy walking shoes available (perhaps in your go-bag…)
• If you are unable to evacuate on foot, discuss this with your Building Emergency Team Leader in advance so that you can work out an assistance plan for your evacuation.

Be aware of multiple routes to exit LBNL and ways to get home.

What’s next?

There may be a designated off-site gathering area for people who cannot get home to meet and receive assistance.

After evacuating safely, contact your supervisor, if possible, and let him or her know that you are OK.

Monitor LabAlert, visit, or call 1-800-445-5830 for information about when it is safe to return.

Be In the Know in Emergencies with LabAlert

Timely and accurate information is a precious thing in an emergency. LabAlert brings the latest news on what’s happening and what to do straight to your cellphone. Click here to learn more and sign up. The free service is activated only in emergency situations and the information is direct and to the point. We encourage employees, students, and onsite affiliates to sign up.