ISOLDE goes from strength to strength

24 November 2004

In 1964 CERN took the initiative to develop the means for studying short-lived nuclei. Forty years on, ISOLDE continues to be a world-leading facility for research with radioactive beams.


If you can look into the seeds of time,
And say which grain will grow and which will not…
Macbeth I, 3

Experiments with radioactive beams are attracting a great deal of interest these days. In the US, a radioactive-beam facility – the Rare Isotope Accelerator (RIA) – has been approved as one of the top-priority projects in physics for new construction. In Europe, there are several new projects that are well advanced in securing funding, such as SPIRAL 2 in France and the Facility for Antiproton and Ion Research (FAIR) in Germany. In the next decade, a high-intensity installation called EURISOL could become a powerful successor to CERN’s ISOLDE (Isotope Separator On Line).

In the year of CERN’s 50th anniversary, it is interesting that the field had its beginnings at the organization 40 years ago. Even earlier, an experiment carried out at Niels Bohr’s Institute in Copenhagen in 1951 had proved the feasibility of connecting an electromagnetic isotope separator to a cyclotron. Its scientific aims lay primarily with neutrino physics, but a decade later it was becoming clear that the same technique would allow an attack on the problem of unstable nuclei with very short lifetimes, which were predicted theoretically. It was also clear that this would require a major effort. During 1963-64, CERN’s director-general, Victor Weisskopf, consulted leading European nuclear physicists, and early in 1964 he issued a call for proposals for nuclear experiments at CERN’s 600 MeV Synchrocyclotron (SC).


There was already an informal European network active in the areas of isotope separators and nuclear structure, and within this a collaboration emerged to prepare a proposal for studying unstable nuclei. Among its prominent members in the early phase were René Bernas (Orsay), Wolfgang Gentner (Heidelberg), Karl Ove Nielsen (Aarhus), Alexis C Pappas (Oslo) and Goesta Rudstam (Uppsala). Later the same year, a proposal prepared by the collaboration was presented to Weisskopf, and on 17 December 1964 he invited the groups to go forward with the proposed programme.

On 16 October 1967, the first experiments were carried out at ISOLDE. The experimental arrangement that had emerged from the collaboration’s interactions with the leaders of the SC Division, Giorgio Brianti and Ernst Michaelis, is shown in figure 1. An important and far-sighted feature of the design was that the extracted proton beam was taken to a shielded area approximately 6 m below ground in order to reduce the external radiation levels, although this increased the construction costs considerably. Without this feature, it would have been impossible for ISOLDE to make use of the 100 times stronger beam that became available after the SC Improvement Programme (SCIP) in 1973-74.

The SC closed in the early 1990s, and the ISOLDE facility moved to the PS Booster where, at the time, there was sufficient capacity to enable operation with an average of 2 μA of protons at 1 and 1.4 GeV. The pulsed proton beam from the PS Booster initially caused major concerns, as many target types deteriorated quickly owing to the high instantaneous beam power. Technical improvements and inventions would eventually help to overcome these problems, turning the pulsed beam into a unique feature for release measurements of radioactive species from the targets and for physics that could benefit from a semi-pulsed radioactive beam.

Technical development to keep CERN at the leading edge in this field has always been important at ISOLDE. The development of a Resonant Ionization Laser Ion Source (RILIS) in collaboration with the Institute of Spectroscopy in Troitsk (Russian Academy of Sciences) had already started while ISOLDE was still at the SC, and the decision to continue this work has proved essential for the facility’s competitiveness. Today the RILIS system can be used to selectively ionize 25 elements, and it is now employed in more than half of the physics shifts that the facility delivers. The development of a two-stage target with a durable primary target directly hit by the proton beam – resulting in a shower of neutrons irradiating UC and ThC secondary targets – has also been essential. With this type of target, the total yield of secondary radioactive ions is smaller, but the suppression of many reaction channels in the secondary target results in a pure beam of fission fragments. The technique was first proposed by Jerry Nolen of Argonne National Laboratory, and is the leading principle for at least two proposals for future radioactive beam facilities, SPIRAL 2 in France and SPES in Italy.


The fundamental understanding of the nucleus requires precision measurements of its properties at the edge of stability, and this requires that the radioactive beams are accelerated to energies of several million electron-volts per nucleon. In the 1990s, a collaboration between several European institutes proposed a compact linear accelerator for radioactive ions at ISOLDE, the Radioactive Beam Experiment (REX). This system, in particular the low-energy part, is highly innovative. The radioactive ions from ISOLDE are delivered at 60 keV predominantly in charge state 1+. At REX the ions are captured in one of the world’s largest Penning traps (REX trap) and subsequently undergo cooling and side-band cooling; both techniques have been developed at the ISOLDE ISOLTRAP experiment. In the next stage, the ions are injected into an Electron Beam Ion Source (EBIS) for charge-state multiplication. This is an ultra-high vacuum system with an intense electron beam guided by a strong magnetic field. The ions are trapped in the electron beam and will lose electrons through collisions. The fact that the EBIS system operates at a very good vacuum results in a pure and highly charged radioactive ion beam in which only isobaric contaminants from the previous stages can potentially cause pollution of the beam.

A mass separator after the trap-EBIS system permits the selection of a charge state for further acceleration in a linear accelerator. This device is very compact thanks to the high charge states used, and consists of a radiofrequency quadrupole (RFQ) followed by an interdigital H-type (IH) structure and multi-gap resonators. It is very similar to LINAC 3 at CERN, which is used for the heavy-ion programme, the main differences being that at REX-ISOLDE the isotopes accelerated are frequently changed and the beam intensities are very low. The main experimental device operating at REX is MINIBALL, a compact and highly efficient array of segmented germanium detectors. In recent experiments the pure beams have permitted precise measurements of the shape of nuclei that are very rich in neutrons, such as 32Mg, 78Zn and 126Cd. The results are very encouraging and show that the full REX system is unique and provides a powerful tool for the investigation of the structure of exotic nuclei.


ISOLDE’s success has two main ingredients. First, CERN provides the infrastructure and manpower that make it possible to operate a large facility serving many users, and second, the collaboration of many European physics institutes leads to a broad programme of a high scientific quality that justifies the effort. ISOLDE is today delivering some 350 eight-hour shifts of radioactive ion beams per year. The variety and availability of such beams far exceeds that of any other low-energy facility in the world.

The ISOLDE collaboration remains a major driving force behind the evolution of the facility, and it makes an important contribution to both its operation and its development. The large amount of knowledge in target and ion-source chemistry and technologies that has accumulated over many years has resulted in 700 isotopes from 70 elements being available at ISOLDE. Combined with important technical developments such as the RILIS and REX, this makes ISOLDE a unique facility at the forefront of research in nuclear physics and allied fields. ISOLDE is also well integrated in the European research structure through the EURONS infrastructure initiative.

The future of ISOLDE and REX-ISOLDE will be determined by the proposed upgrades of the injector accelerators at CERN to provide a primary proton beam of much higher intensity. The ISOLDE hall is currently being extended, and CERN is undertaking a major consolidation of the facility itself, including a new laboratory for target handling. In addition, plans are under way to increase the energy of the REX post-accelerator and make highly charged and cooled beams available for other experiments. ISOLDE also plays a central role in the European Union design study for the third-generation radioactive ion-beam project EURISOL. CERN would be an ideal site for this facility, with unique synergies with other areas of research through the multi-megawatt proton driver that is required. The link to neutrino physics within EURISOL (the beta-beam concept) makes an interesting connection to where it all began in Copenhagen – the study of neutrinos through nuclear methods that demonstrates the close alliance between nuclear and particle physics.

In 1981, D Allan Bromley of Yale University, later science advisor to President George H Bush, wrote in an external assessment requested by the CERN Directorate: “The question sometimes arises as to why other major activities of the scope of ISOLDE have not been mounted in the US and elsewhere. I believe that the answer is rather simple. The ISOLDE group got such a head start on the rest of the world activity in this field that people were very reluctant to attempt to mount a competitive operation.” Since then the field of radioactive ion-beam physics has expanded enormously, and ISOLDE’s lead has been followed by major investment in new facilities in three continents.

Further reading

P G Hansen 1996 “The SC: ISOLDE and Nuclear Structure” in History of CERN vol. III ed. John Krige (Elsevier) 327.

Peter Butler, CERN, P Gregers Hansen, National Superconducting Cyclotron Laboratory and Department of Physics and Astronomy, Michigan State University, and Mats Lindroos, CERN.

bright-rec iop pub iop-science physcis connect