On its 30th anniversary, the European Synchrotron Radiation Facility is dismantling its flagship storage ring to make way for a revolutionary X-ray source.
If there is one area where particle accelerators have had the most significant scientific impact, it is surely X-ray science. Each year, tens of thousands of experiments are carried out at the 50 or so synchrotron X-ray facilities worldwide, spanning a wide range of disciplines – from materials science and archaeology, through structural biology, planetary science, environmental science, nanotechnology and more.
Nestled between two rivers in northwest Grenoble, France, the 22 partner-nation European Synchrotron Radiation Facility (ESRF) is one of the world’s leading light sources. It is based on a 844 m-circumference 6 GeV electron storage ring with 44 specialised experimental stations serving around 5000 users per year, and so far experiments there have contributed to over 32,000 peer-reviewed publications. Established in 1988 and inaugurated in 1994, the ESRF was the first third-generation synchrotron, using periodic magnetic arrays called undulators to deliver the world’s brightest X-ray beams. Since then, numerous national light sources have sprung up, alongside flagship third-generation facilities in Japan and the US.
A decade ago, the ESRF embarked on a 330 million euro upgrade programme to help ensure Europe’s leading position in the light-source world. At its core is a first-of-its-kind storage ring with an increased X-ray brightness and coherent flux 100 times higher than before, in addition to new X-ray beamlines, instrumentation, computing and other improvements. The ESRF upgrade will allow users to probe complex materials at the atomic level in greater detail, with higher quality, and much faster. The final beam of the current ESRF storage ring will be extracted on 10 December, and operations will cease for 20 months while the new machine – called the Extremely Brilliant Source (EBS) – takes shape.
Synchrotron radiation, which spans a broad spectrum up to the hard X-ray region, is famously something particle physicists try to avoid, as it limits the energy of circular colliders. But the power of X rays to elucidate the structure of matter had been known since the early 1900s, and it wasn’t long before this by-product of particle physics was put to use. Following the rise of storage rings during the 1960s, nascent synchrotron X-ray users carried out experiments “parasitically” by placing their apparatus in the path of synchrotron radiation from circular particle colliders. By the 1970s, dedicated second-generation sources started to be built, but it was not until the 1990s that third-generation light sources such as the ESRF were possible. Instead of just producing X-rays from the curved trajectories of electrons passing through a dipole, these facilities use long zip-like arrays of alternating-polarity magnets called undulators to produce much brighter, more spectrally coherent, X-ray beams. This allows users to probe matter at shorter spatial and temporal scales.
Key to increased X-ray brightness at a synchrotron is a parameter called the beam emittance, which is a measure of the size and spread of the electron beam as it circulates. Third-generation storage rings reduce the vertical emittance to the X-ray diffraction limit, thanks to a lattice design that decouples the vertical and horizontal motions of the electrons. Reducing the horizontal component is more challenging. One approach is to build a bigger storage ring (the 6.3 km tunnel at Fermilab that housed the Tevatron collider has been mooted as a possible site for an ultimate X-ray source, for example). However, that option isn’t always suitable, and the ESRF had to find a smarter solution.
The existing ESRF storage ring consists of 32 cells each 27 m long and made up of sequences of magnets, vacuum chambers and position monitors. Technically it’s called a double bend achromat lattice, since it relies on two dipoles per cell and because electrons of different energies are bent and focused in the same way (achromatically) resulting in very collimated and stable beams.
The new EBS lattice has the same footprint as the previous machine and will leave the present beamline layout unchanged. It is based on a hybrid multi-bend achromat (HMBA) with seven, as opposed to two, bending magnets per cell and optics that maximise the stable phase space volume available for the electron beam, reducing the horizontal emittance. The result is a tighter packing of electrons, increasing the brightness and degree of coherence of the X-rays by two orders of magnitude. This gives the EBS beams laser-like properties approaching those of X-ray free-electron lasers (XFELs) such as the European XFEL (CERN Courier July/August 2017 p18), and will make EBS the first high-energy fourth-generation synchrotron light source.
Inspired by SuperB
The EBS lattice uses concepts developed for the former SuperB project (an asymmetric electron—positron collider for flavour-physics studies) which, unusually for a collider, had an optimal horizontal emittance close to zero. The physicist behind the design, Pantaelo Raimondi, who led the SuperB project at INFN Frascati (and who is also the originator of the crab-waist technique to maximise collider luminosity, which is currently being considered for almost all future high-energy colliders) became director of accelerators at the ESRF in 2012 and set about the task of building a new kind of light source. An important moment in the evolution of the HMBA came at a workshop at CERN in 2011, when Raimondi realised the synergy with the multi-bend achromat design for an advanced synchrotron in Lund, Sweden, called MAX IV (CERN Courier September 2016 p39).
The challenges facing the ESRF’s accelerator physicists, engineers and technicians are huge. The new storage ring requires 1000 innovative magnets – nearly twice as many as in the previous ring. These have to be squeezed into the same space inside the accelerator tunnel, so each of them has to be more compact and generate magnetic fields up to three times stronger than existing models. The vacuum chambers also have had to be redesigned to fit the limited space in and around the magnets, while the mechanical tolerances of some components have shrunk to within a hundredth of a millimetre. Non-evaporable getter (NEG) technologies pioneered at CERN are to play an important role in overcoming these challenges. EBS will use permanent-magnet technology for its 128 dipoles, and around 90% of the existing ESRF infrastructure will also be reused. The reduced beam energy loss due to unwanted synchrotron radiation and the optimised magnet design have also resulted in a decrease of the power consumption of the synchrotron by about 30%.
Last year, engineers built an entire EBS cell, consisting of girders, magnets, vacuum chambers and other components. This allowed the team to confirm the engineering principles of the new arc and to start series production of the storage-ring components. The existing storage ring will be dismounted by April 2019 and EBS will be installed later in the year and commissioned by March 2020, with user service resuming by September that year.
Four brand new beamlines will be designed to take full advantage of the EBS, while others will be refurbished, and the project has a special development programme to address the optics that transform the raw X-ray beam into a form suited for experiments. The X-ray detectors themselves are also being developed to handle the increased flux, with one technology based on the MEDIPIX3RX chip developed by the Medipix3 collaboration, of which the ESRF and CERN are partners. Control systems, mechatronics and software are other key areas of work to ensure the EBS is ready when it enters user mode two years from now.
Lattice travels
The EBS lattice has inspired other major light sources around the world. In addition to MAX IV, which was inaugurated in 2016, several multi-bend achromat synchrotrons are under planning or construction. This includes Sirius in Brazil but also upgrades at high-energy synchrotrons such as the Advanced Photon Source (APS) at Argonne National Laboratory in the US. Meanwhile, the Advanced Light Source (ALS) in California is moving towards a conceptual design report, and SPring-8 in Japan is pursuing a HMBA that will enter operation on a similar timescale. The long-term goal is to one day build an “ultimate” storage ring in which the horizontal emittance reaches the fundamental X-ray diffraction limit, as is the case today for the vertical emittance. Meanwhile, XFELs are starting to so sprout up in the way storage rings did around the turn of the millennium.
Particle physics will continue to play a vital role in X-ray science by driving accelerators forward and sharing vacuum, detector and other technologies. In fact, when workers come to dismantle and rebuild the ESRF storage ring in the coming weeks and months, they will also be ripping out and remounting parts of the radio-frequency (RF) system of CERN’s Large Electron–Positron (LEP) collider, the forerunner to the LHC. LEP’s RF system was the starting point of the ESRF design and teams at both laboratories, located just 150 km apart, collaborated closely to turn third-generation light sources from an idea into reality.
