GUSTAVO TÍLIO / LNLSElectrons produce visible radiation, X-rays and ultraviolet light GUSTAVO TÍLIO / LNLS
Brazil’s largest scientific and technological research instrument is to get an even greater and more powerful version by 2015. The project for a new source of light, which is to adopt innovative construction solutions, is almost ready. It is being designed by researchers from LNLS, the National Synchrotron Light Laboratory in the city of Campinas, inner-state São Paulo, which is open to Brazilian and foreign researchers from academia and from firms, engaged in technological projects or studies that require that they investigate with synchrotron radiation the atomic structure of materials such as polymers, rocks, metals, proteins, medication or cosmetic molecules, and even tridimensional images of fossils or of cells. This radiation is generated by electrons produced within an accelerator and then inserted into a metal ring with a 93-meter circumference (the new ring is to have a 460-meter circumference), in an ultra-high vacuum environment. They go round almost at the speed of light and when they pass by the magnets along the ring, the magnetic field deflects them, causing them to release photons, thus forming the so-called synchrotron light, consisting of electromagnetic waves, such as those with the frequency of X-rays, ultraviolet light and even visible light (the latter seldom used in scientific experiments). This light is employed by researchers in LNLS at 14 workstations with light lines, distributed around the ring.
The new source has already been named Sirius – a name chosen from the staff’s suggestions – in reference to the most brilliant night sky star. It is important to build this because the current instrument is becoming obsolete. The Brazilian synchrotron is about to complete 13 years of service and technological and scientific needs demand more up-to-date equipment. “The evolution is necessary because, at the end of the day, science is competition. The important, relevant questions in the areas served by the Synchrotron are always new, as several of the old ones have been answered already. The new ones, however, require more sophisticated equipment”, says the physicist Antônio José Roque da Silva, LNLS director since July 2009 and a professor at the Physics Institute of the University of São Paulo (USP). One of the advantages of a laboratory such as the Synchrotron is its cross-disciplinary nature, involving people who do research into biology, materials science, technology, energy and paleontology. “With the LNLS, the country can compete in several areas and use the same laboratory simultaneously, the year round, to carry out its experiments”.
The new Synchrotron project is being entirely designed in Brazil and is to be a third-generation laboratory, as opposed to the current, second-generation one. There are currently some 50 sources of synchrotron light in the world, of which 16 are third generation that went on-stream as from 1994. They are characterized by their more brilliant radiation, with more light being generated and with low emittance, this being a unit of measurement that determines the size and the divergence (spreading) of the focus of the light source. “The lower the emittance, the greater the possibility of focusing the beam produced”, explains the civil engineer and physicist Ricardo Rodrigues, technical director of the new source project. He took part in the building of the first Synchrotron, inaugurated in 1997. Sirius is being designed to have a 1.7 nanometer-radian (nm.rad), whereas the current synchrotron source has 100 nm.rad. This means greater brilliance on a smaller radiation beam and also with a smaller angle of opening. It should be one of the brightest sources in the world. For example, the Soleil Synchrotron in the town of Saint-Aubain, France, which was inaugurated in 2006, has emittance of 3.7 nm.rad, while the Diamond Synchrotron, in Oxfordshire, England, which went on-stream in 2007, has 2.7 nm.rad.
“Of the 50 sources of synchrotron radiation in the world, only 30 are open to researchers from outside the institution to which the laboratory belongs. There are 11 in Europe, 7 in the United States, 10 in Asia, 1 in Australia and 1 in South America, which is the LNLS. If the second source is not built, Brazil and South America will disappear from the world’s synchrotron radiation map”, says the French physicist Yves Petroff, LNLS’s scientific director since December 2009 and the person responsible for the scientific objectives of the new light source project. Of the 1,656 LNLS users in 2009, 20% came from Latin America. Out of this total, 14% were Argentine. These studies led to approximately 250 articles in scientific journals.
“Smaller countries, such as Spain, South Korea and Taiwan are building third-generation sources”, says Petroff. At the age of 73, he has had a long career in synchrotron laboratories around the world. From 1993 to 2001 he was the director-general of the European Synchrotron Radiation Facility (ESRF) in Grenoble, France, and has also worked in similar laboratories in the United States. Additionally, he has held the position of scientific director of LURE (the acronym in French for the Laboratory for the Use of Electromagnetic Radiation) and of CNRS (the acronym in French for the National Center for Scientific Research). He sits on several synchrotron scientific committees, including that of LNLS, since its implantation stage back in 1988. “An interesting point is that the number of users from the Energy Department in four American synchrotrons grew by 40% (from 6,000 to 8,400) between 2000 and 2008, while the users of the French ESRF grew by 36% between 2003 and 2009”, he says. “A substantial part of this growth is due to using this radiation to study biological structures. All pharmaceutical companies, for instance, use the lines of light for this purpose”. He also reminds us that recently the exploration of specific properties of the X-rays produced by synchrotron machines have enabled the obtainment of three-dimensional images of any object, with resolutions smaller than a micrometer (one thousandth of a millimeter), in paleontology, archeology and environmental studies. “I invited Yves Petroff to help restructure the LNLS scientific division and to help with the scientific objectives to use the new source and the new lines of light, as they are becoming increasingly sophisticated”, says Roque.
EDUARDO CESARLight lines: work stations for researchers from academic institutions or companies EDUARDO CESAR
Permanent magnets
Besides meeting the required specifications of a third generation synchrotron light source, the project also plans to achieve a much bigger reduction in power consumption. To this end, new solutions are being tested in the LNLS, based on innovative technologies. The first is to adopt permanent magnets, a global novelty for this type of laboratory. These magnets are to be used in the construction of the dipoles that produce the magnetic field that deflects the electrons from their path within the ring to form photons and hence synchrotron light, which is captured and filtered, among the various electromagnetic waves found in the beams of the lines of light. These dipoles currently function with electromagnets, consisting of metals wrapped in wires that turn into magnets once they are electrified. This system dictates that a number of other instruments be attached , such as a refrigeration system and coils, which consume a lot of power. “The permanent magnets are similar to refrigerator magnets”, compares Rodrigues. They don’t need electric energy to work and are sold commercially worldwide. They are made out of ferrite, which is cheap, and alloys that include neodymium, iron and boron. To date, only one machine in the word, an antiprotons accumulator at Fermilab, in the United States, runs with permanent magnets. “Nobody has yet had the courage to do this in synchrotrons, although the understanding of these materials has progressed quite a lot”, says Rodrigues. Cutting power consumption carries a lot of weight in this decision. With the permanent magnets, savings of 6.5 gigawatts-hour (GWh) are expected, or R$4.5 million a year.
Another innovation developed at the Synchrotron, in collaboration with Soleil, the French laboratory, is to be used both for the new source and the old one. It is a radically different radiofrequency (RF) system that will save more than R$1 million in power a year. The laboratory’s current power bill is some R$3.5 million a year. The RF system replenishes the energy that the electrons lose in the form of synchrotron light. Though resorting to the most advanced technologies, almost all such laboratories around the world operate using electronic valves almost one meter long, each of which costs US$150 thousand. Such valves were heavily used in electronic devices before the commercial advent of power transistors. In the synchrotron case, they are especially made in England to supply the high energy needed to amplify the 476 mega-hertz (MHz) frequency. This electromagnetic wave, rather than expanding in space, as in a radio station, for instance, is imprisoned within chambers called resonating cavities along the ring. The current source uses two such RF generators of 30 kilowatts (kW) each. “To date, the only way of joining high power and high frequencies was this valve”, says the electronic technician Claudio Pardine, coordinator of the LNLS radiofrequency laboratory.
Pardine, in collaboration with the French researchers from Soleil, developed a new system called solid state amplifier, comprised of hundreds of small electronic boxes with 250 watts of power. “As early as 2001, LNLS was the world’s first laboratory to replace the valve by the solid state amplifier in a system of one kW for a synchrotron light injector”, says Pardine. There are countless advantages to this, but the main one is really to save power. “To supply 30 kW, the traditional valve system requires power of 170 kW; the new solid state one needs 60 kW.” At present, the RF system consumes almost 1.8 gigawatts-hour (GWh) a year, meaning that the cost of power for the RF equipment amounts to R$1.3 million a year. With the implementation of the new system, there will be savings of about 50%, not to speak of eliminating the need to replace the valve every five years. “We stop being hostage to the manufacturer. Maintenance becomes easier and cheaper”.
Soleil, the French laboratory, was the first to install a solid state amplifier with tens of kilowatts. “We built some of the components of these amplifiers at LNLS in 2005”, recalls Pardine. “We sold the parts at cost to make these prototypes that we then developed jointly”. Pardine follows the guidance of the Chinese researcher Ti Ruan, who currently works at Soleil and who previously taught at the University of Paris. Ruan persuaded the French laboratory directors, during construction, to use the solid state amplifier. However, another major laboratory, Diamond, in England, inaugurated in 2007, preferred the valve. Pardine highlights that the idea of using solid state amplifiers is an old one, but this only became feasible now, thanks to the evolution of materials and electronic equipment. To develop and build the new RF towers, he got financing from Finep, the Studies and Projects Finance Agency, under a program for electric energy equipment, in the amount of R$1 million.
WILLIAM BARBOSA / LNLSNew source, shaped like a circle, is to be built next door to the current oneWILLIAM BARBOSA / LNLS
Solid ground
To run exemplarily, Sirius will require super-stability in the large storage ring, to ensure that the electrons are not deflected by more than one thousandth of a millimeter (micrometer) from their projected orbit. The same super-stability also applies to the light lines. Any variation or metal dilation can disturb the electron beam. The air conditioning equipment, for instance, when it causes the temperature to fluctuate by half a degree, dilates the ring’s concrete and steel support by micrometers. This is not good for the electrons. “There are small soil variations that are imperceptible under normal circumstances, but when one works on a micrometric scale, they become very important”, says Ricardo Rodrigues. The project therefore includes a hardened super-floor, 200 meters in diameter and 1 meter thick, with no joins. “Nobody has yet made this type of flooring in Brazil. It’s 20 thousand cubic meters of concrete to be produced in one week, 24 hours a day. The layers are poured one over another and the curing (drying) of the material can’t be fast”. These layers are moist and must not dry until further layers have been added. Therefore, ice is going to be added to the process. Special logistics will be set up, including a concrete plant and an ice plant next to the building site of the new source.
The initial budget estimate for Sirius is approximately R$400 million, spread over six years. The money should come independently from the Ministry of Science and Technology or in partnership with other federal institutions. The Ministry has a management contract with ABTLuS (the Brazilian Association of Synchrotron Light Technology), a social organization that is in charge of the Synchrotron and of two other laboratories on the LNLS site: LNBio (the National Biosciences Laboratory), a former Synchrotron center that is now autonomous, and CTBE (the National Laboratory of Bioethanol Science and Technology). Both of them also use synchrotron radiation in some of their experiments. All three are coordinated by CNPEM (the National Center of Research into Energy and Materials), whose head, as from June, will be professor Walter Colli, formerly a professor at the Chemistry Institute of the University of São Paulo (USP).
LNLS has an experienced team heading the project that already knows how a synchrotron is built. Ricardo Rodrigues was one of the first three researchers hired in August 1986 by CNPq (the National Scientific and Technological Development Center), which managed LNLS at the time, to build the laboratory. “We had Cylon Gonçalves da Silva as director, Aldo Craievich to look after the researchers’ use of the lab, and me to look after the project and the construction, which took 10 years”, Rodrigues recalls. According to him, it was not only the fact that funds were constantly put under contingency that delayed the project. “I can’t blame it all on the budget. Acquiring the experience and the know-how took a long time. I think we used the best method for learning something. The hired staff, an engineer or physicist fresh out of university, was given the following message: ‘You have to do this. We’ll help as much as possible, we’ll work together’. Nobody went to do a doctorate or take special course. We had them take trips to visit other labs and they’d ask: ‘How do you do it?'”, says Rodrigues.
Physicist Liu Lin was one such professional on the initial team. “In 1985, when I was doing my master’s degree at USP’s Physics Institute, in São Carlos, I worked on the project of the ring’s magnetic network and I did dynamic simulations of the electron beam. Afterwards, I was on the team that spent three months at the Stanford Linear Accelerator Center (SLAC) at Stanford University, in California, in the United States”, Liu says. “We learnt a lot because there they make instruments and we had the opportunity of designing a fictitious machine that enabled us to get acquainted with the physics of accelerators”, says Rodrigues. This same ideal of building instruments and systems used in the construction of the first ring will be employed in the upcoming one. “We have designed and purchased a number of things, but in financial terms only 16% of the first machine was imported”.
As the current leader of the LNLS Accelerator Physics Group, Liu studies the dynamics of electrons under the effect of the electromagnetic field. “We designed these fields so as to ensure that a strong electron beam with high energy can be stored in a stable manner, producing synchrotron light for several hours. To achieve this, we must specify among other things a magnetic network, which will determine all the properties of the synchrotron light beam produced”, says Liu. For Rodrigues, the project is almost complete and the outlook is for construction to take half the time the first machine took. “Now it’s no longer urgent to train personnel; the group of people coordinating the project is still young”.
