Quantum spin excitations in a single crystal of copper germanate

The neutron community has expanded enormously since nuclear physicists made the first measurements of neutron cross-sections in the 1940s. Crystallographers started to exploit the potential of the new technique in the 1950s and were soon joined by condensed-matter physicists. The advent of “cold” neutron sources, which provided a rich flux of long-wavelength neutrons, attracted chemists and later biologists to neutron scattering. And in the last decade or so engineers, materials and earth scientists have joined the club, all attracted by the unique possibilities that neutron scattering is offering to their fields.

Neutron scattering provides unique microscopic information on the structure and dynamics of materials. Within condensed matter, neutrons have made outstanding contributions to our understanding of fundamental phenomena such as magnetism, phase transitions, spin dynamics and quantum fluids. This was formally recognized when Bertram Brockhouse of Canada and Clifford Shull of the US shared the 1994 Nobel Prize for Physics for their neutron-scattering experiments in the 1950s. Neutrons have also contributed greatly to our knowledge of technically important materials such as plastics, proteins, polymers, fibres, liquid crystals, ceramics, hard magnets and superconductors.

Although neutron scattering was pioneered in North America, Europe is now home to the world’s most intense reactor at the Institut Laue-Langevin (ILL) in Grenoble, France, and the world’s most intense pulsed source, the ISIS Facility at the Rutherford Appleton Laboratory in the UK. European neutron scientists are now drawing up plans to maintain their lead in neutron-scattering research by building a next-generation European Spallation Source (ESS).

The ESS would be 30 times brighter than any currently available pulsed source and would lead to a new generation of experiments in many areas of science. Within physics, the ESS could address fundamental questions in fields as diverse as superconductivity and cosmology. The ESS would also be invaluable to researchers in biology, chemistry, engineering, geology and medicine, and could be used to improve our understanding of many industrially important new materials.

Neutron basics

Neutrons interact with matter through all four forces: the strong, weak, electromagnetic and gravitational interactions. However, it is their interactions via two of these forces – the short-range strong nuclear force and their magnetic moments – that make neutron scattering such a unique probe for condensed-matter research. The most important advantages of neutrons over other forms of radiation in the study of structure and dynamics on a microscopic level are summarized below.

  • Neutrons are uncharged, which allows them to penetrate the bulk of materials. They interact via the strong nuclear force with the nuclei of the material under investigation.
  • The neutron has a magnetic moment that couples to spatial variations of magnetization on the atomic scale. Neutrons are therefore ideally suited to the study of magnetic structures, and the fluctuations and excitations of spin systems.
  • The energy and wavelength of neutrons may be matched, often simultaneously, to the energy and length scales appropriate for the structure and excitations in condensed matter. The energy, E, and wavelength, l, are related via the de Broglie relation, E = h2/2mnl2, where h is the Planck constant and mn is the neutron mass. An energy of 20.45 meV corresponds to a wavelength of 2 Å. Neutrons with wavelengths of 0.1-20 Å are ideal for the study of interatomic correlations. In addition, a wide range of energy scales can be probed, from the micro electron volt (µeV) energies associated with polymer reptation, through molecular vibrations in the meV range, to eV transitions within the electronic structures of materials.
  • The use of polarized neutron beams – in which all of the neutron magnetic moments point in the same direction – allows us to separate the nuclear and magnetic contributions to the scattering. In non-magnetic systems, coherent and spin- incoherent scattering, arising from different nuclear spin states in atoms, can be separated.
  • The neutron does not significantly perturb the system under investigation, so the results of neutron-scattering experiments can be clearly interpreted.
  • Neutrons are non-destructive, even to delicate biological materials.
  • The high penetrating power of neutrons allows us to probe the bulk of materials and facilitates the use of complex sample-environment equipment (e.g. for creating extremes of pressure, temperature and magnetic field).
  • The neutron scattering cross-section varies randomly between different elements and even between isotopes. This allows us to observe light atoms such as hydrogen in the presence of heavier ones, to distinguish neighbouring elements in the periodic table easily, and to exploit isotopic substitution and contrast variation methods.

In isotope difference methods, measurements are made on samples that are identical except for the substitution of an isotope. By subtracting the measurements we can “see” the structure of the system from the viewpoint of the substituted element. For example, by substituting a nickel isotope in an aqueous solution of nickel chloride we can “sit on” the nickel ion and see how it is hydrated by the water molecules. Contrast variation exploits the fact that hydrogen and deuterium have elastic scattering amplitudes of different sign. Thus, by varying the H/D ratio in part of a system (e.g. the protein coat of a virus), the scattering density can be matched to that of another part of the system. This can make the protein coat “invisible” to neutrons, which can then be used to locate the nucleic acid inside the virus.

Neutron sources and scattering experiments

Neutrons have traditionally been produced by fission in nuclear reactors optimized for high neutron brightness (usually measured as neutrons per second per steradian per meV). The neutrons from such steady-state sources are produced continuously. After the high energy (MeV) neutrons have been thermalized to meV energies, beams are emitted with a broad band of wavelengths. The energy distribution of the neutrons can be shifted to higher energies (shorter wavelengths) by allowing them to come into thermal equilibrium with a “hot source” (at the ILL this is a self-heating graphite block at 2400 K), or to lower energies with a “cold source” such as liquid deuterium at 25 K. The resulting Maxwell distributions of energies thus have the characteristic temperatures of the moderators (figure 1a).

Accelerator-based pulsed neutron sources are a more recent development. In these sources neutrons are released by bombarding a heavy-metal target with high-energy particles from a high-power accelerator – a process known as spallation. The neutron beams are pulsed because the accelerator beam is pulsed. Spallation releases much less heat per useful neutron than fission (typically 30 MeV per neutron, compared with 190 MeV in fission). The low heat dissipation means that pulsed sources can deliver high neutron brightness – exceeding that of the most advanced steady-state sources – with significantly less heat generation in the target.

Figure 1

As in fission, the neutrons produced in the spallation process have very high energies and must be slowed by many orders of magnitude – from MeV to meV. This is achieved by hydrogenous moderators, such as water at room temperature, methane at 100 K and liquid hydrogen at 20 K. These moderators, which surround the spallation target, exploit the large inelastic-scattering cross-section of hydrogen to slow down the neutrons by repeated collisions with the hydrogen nuclei. The neutron spectrum is more complex than that from a reactor. It has two parts: a “slowing” region of hotter, incompletely thermalized neutrons, and a Maxwell distribution characteristic of the moderator temperature. Since neutrons of different energies interact in a complex way with the moderator system, the shape of the neutron pulse can be tailored and optimized for different classes of instruments (figure 1b).

Thus the characteristics of the neutrons produced by a pulsed source are significantly different from those produced at a reactor (figure 1c). The time-averaged flux (in neutrons per second per unit area) of even the most powerful pulsed source is low in comparison with reactor sources. However, judicious use of time-of-flight techniques that exploit the high brightness in the pulse can compensate for this. Some of the advantages of using pulsed sources are listed in the box.

In a typical neutron-scattering experiment, instruments measure the number of neutrons scattered by the sample as a function of scattering angle. For elastic scattering this corresponds to measuring with diffractometers the momentum change, which provides information about the spatial distribution of nuclei in systems ranging in size and complexity from small unit-cell crystals, through disordered systems such as glasses and liquids, to “large scale” structures such as surfactants and polymers. Spectrometers, on the other hand, additionally measure the energy lost (or gained) by the neutron as it interacts with the sample. These data can be related to the dynamical behaviour of the sample, such as the dynamics of the atoms or molecules in a crystal lattice (phonons) or of the magnetic moments in magnetic systems (spin waves).

To take advantage of the different characteristics of reactor and pulsed sources, experiments performed on the sources differ in detail. In experiments at reactors, for example, a single-wavelength beam is normally used (figure 2). Monochromatic beams can be produced by wavelength selection from a crystal monochromator or by velocity selection through a mechanical chopper. In contrast, “white” beams that contain neutrons with a wide range of wavelengths are generally used at pulsed sources. Energy analysis of the scattered beam is achieved by Bragg scattering from an analyser crystal or by measuring the total time-of-flight. In the latter case, the initial time-of-flight (i.e. the time the neutron takes to travel from the source to the sample), has to be determined as well.

Figure 2

Neutron guide tubes, which exploit the phenomenon of total reflection, can be employed to transport all but the high-energy neutrons over large distances without a significant loss in intensity. As high-resolution measurements tend to require long times-of-flight, guide tubes allow very high-resolution measurements to be made without suffering the fall in intensity caused by the inverse square law.

Achievements of neutron scattering in physics

Neutron scattering has achieved a great deal, not only in physics, but over a very broad range of science: geologists explore the structure of minerals under extreme temperatures and pressures; biologists are increasingly turning to neutrons to collect information that is complementary to that obtained with other techniques, in particular where hydrogen or water is involved; while engineers use neutrons to probe stresses deep inside complex components. In this article, however, we will emphasize some of those areas of physics, in particular condensed-matter areas, in which neutron scattering has been of outstanding importance.

Almost everything we know about magnetic structures – from the early demonstration by Shull of antiferromagnetism in simple systems to the complex structures of today’s hard magnets – has come from experiments with neutrons. The generalized magnetic susceptibility, which is directly proportional to the neutron cross-section, describes all that can be known about the microscopic spatial and temporal correlations of the magnetic moment in a material. This fact, combined with the possibility of using polarized neutrons to separate nuclear and magnetic scattering, has made neutron scattering the foremost tool in fundamental magnetism research.

Figure 3

And when important new materials, such as heavy-fermion systems or high-temperature superconductors, are discovered, neutron scattering is often the first technique chosen to investigate the fundamental physical microscopic properties. Neutrons have, for example, revealed the definitive crystal structures of superconducting materials with high transition temperatures (high-Tc) and have precisely located the positions of the oxygen atoms, where the “holes” that carry the charge reside. Neutron spectroscopy has provided unique information on the nature of magnetism in high-Tc materials, on the interplay between magnetic fluctuations and superconductivity, and on the role of the lattice dynamics. Neutrons gave the first microscopic evidence for the lattices formed by magnetic flux lines in conventional superconductors, and played a major role in accounting for the large heat dissipation in the high-Tc materials, especially at high magnetic fields where no other technique can image flux lines (see figure 3).

The properties of the neutron make it a particularly powerful probe for studies of disordered materials such as liquids and amorphous solids. Among the crucial questions that neutron scattering has solved in this area are the structures of water molecules round ions in solution (literally revolutionizing our understanding of hydration), the microscopic behaviour of supercooled liquids, the detailed structures of binary liquid metals and molten salts, and the structures of amorphous silicon and carbon hydrides that are being used to develop new high-strength materials.

“Soft matter” encompasses a wide range of molecular materials such as polymers, surfactants and colloids, and the range of distance and time scales relevant to the performance of these materials can be remarkable. However, the specific advantages of neutrons have led to many scientific breakthroughs in the field. Examples include the confirmation of the theory of polymer reptation proposed by Pierre Gilles de Gennes, the stabilization of colloids using polymer adsorbates, and the structure of water in lyotropic and microemulsion phases (figure 4).

For more information see the European Science Foundation report and the ESS Scientific Case in further reading.

A next-generation neutron source for Europe

Neutron scattering is very much an intensity-limited technique, and experimenters are pushing hard against the limits of what is possible with today’s sources. In many areas, demonstration experiments are being performed that provide a tantalizing glimpse of what may be possible with more intense neutron beams. Moreover, the broadening and expansion of the neutron-user community – though very much welcome – means that demand is continuously rising. If we want to continue to exploit the power of neutron techniques, a more powerful source is essential.

Many existing neutron sources will soon reach the end of their technically and economically viable life. In 1990 a panel set up by the European Commission recognized this problem and concluded that, unless action was taken, there would be a “drought” of neutrons that would have serious consequences for European science and its exploitation. In 1994, responding to the panel’s report, the commission funded a proposal to examine the technical feasibility and likely cost of a next-generation pulsed source. Thus the European Spallation Source Project – an accelerator-driven neutron source that would be 30 times more powerful than any pulsed source operating at present – was born. The source would take advantage of the huge advances in high-power accelerator technology developed for the US Strategic Defense Initiative and for accelerator-based transmutation of nuclear waste. The ESS Project has just published detailed reports on the scientific case for such a source and its technical feasibility (see further reading). The proposed 5 MW source would cost some ECU 934 m to build (at 1996 prices) and would provide 44 beamlines.

The ESS will deliver a neutron intensity of up to 30 times that of ISIS. Combined with appropriate developments in neutron instrumentation, a total increase of up to three orders of magnitude could be obtained in some experiments. This gain in effective intensity will be used in a variety of ways by the large scientific community that utilizes neutrons to perform “small science” at large facilities.

Figure 4

The high intensity can be used directly to perform experiments on inherently small or dilute samples, or to make measurements over shorter times than are currently possible. In addition, we can trade off some of this enhanced intensity for increased resolution. Exploiting these “generic opportunities” will have a huge impact on all areas of neutron science. Moreover, the ESS will also extend the unique advantages of neutrons to areas where, until now, intensity limitations have prevented their application.

It is difficult to give more than a taste of the exciting possibilities that the European Spallation Source would open up. The ESS Science Case outlines a host of new scientific opportunities in many areas of science (see further reading). Here we will highlight some of the generic opportunities that could be explored in physics.

The need for better resolution is a common quest in physics. Trading increased intensity for higher resolution will enable more accurate structural and dynamical studies of the increasingly complex systems that are central to modern materials science. For example, better experiments on hard magnetic materials (e.g. Nd2Fe14B) are needed to understand magnetic properties on a microscopic scale. This will lead to better materials for the billion-dollar magnetics industry. High-resolution measurements will also shed light on the role that electron-phonon interactions play in “normal” and “unusual” superconductors. The ability to measure small cross-sections will expand studies on systems that carry very small magnetic moments (e.g. molecular magnets and “magnetic” heavy-fermion systems) or experiments in which dilute quantities of paramagnetic ions are used to probe magnetism.

The higher intensity will allow us to measure smaller cross-sections. In disordered materials, we will be able to begin to understand how interactions in aqueous solutions are modulated by changing pH, ionic strength, pressure, temperature and so on – in particular at the low concentrations that are important in many chemical and biological processes. The whole field of neutron Brillouin scattering – inelastic measurements performed at very low momentum transfers – will allow us to explore the dynamical regime between the microscopic and the hydrodynamic for the first time. This regime is of both fundamental and industrial importance. And with so much more flux, the enormous potential of full polarization analysis would become routinely available.

It will also be possible to make measurements over shorter times. By following variations with pressure, temperature, concentration and so on, experimenters will be able to solve real problems, rather than being restricted to single experiments at a single state point that reveals only a small fraction of the picture. For example, the uranium-platinum compound UPt3 exhibits unusual “heavy-fermion” behaviour. Heavy-fermion systems have very large electronic contributions to their specific heat at low temperatures, which can be parametrized in terms of an effective electron mass hundreds or thousands of times larger than those in normal metals. Bulk measurements such as specific heat and thermal expansion as a function of magnetic field, temperature and pressure have established a fascinating phase diagram with strong evidence for unconventional superconducting behaviour in UPt3. However, only very sparse microscopic information obtained by inelastic neutron scattering is presently available to help with the understanding.

Shorter time measurements will also enable real-time studies of kinetic processes. Examples include monitoring the structural changes that occur when a battery is discharged, or watching the phonon spectrum of a crystal during a structural phase transition. The response of a sample under strongly non-equilibrium conditions can also be followed. Measurements of flowing surfaces in sheared systems will give new information on lubrication, while in situ real-time studies of electrochemical and electrode processes will also be possible. And we should also be able to follow, under realistic conditions, the mechanisms – e.g. wall effects during polymer extrusion – that control the processing of soft solids, with a view to optimizing them.

Experiments under extreme sample environments offer the prospect of new and exciting science. For example, we will be able to perform structural and dynamical studies of material phases that only exist in very strong magnetic fields (above 16 T), at very low temperatures (~ nK or less), or at very high pressures (above 25 GPa). All these extreme environments suffer from severe restrictions: high magnetic fields and ultralow temperatures can only be maintained for short times, while only very small sample volumes are possible at very high pressures.

In fundamental neutron physics, a number of experiments can be performed that will help to determine the basic structure of the elementary interactions in nature, to elucidate the history of the universe, and to study a number of questions in quantum and measurement theory. For example, an experiment has been proposed for the ESS that could give a clear-cut answer to the question of the “handedness” of nature. In the late 1950s it was recognized that the weak interaction is exclusively left-handed. However, most Grand Unified Theories – theories that seek to unify the strong and electroweak forces – start with a left-right symmetric universe and explain the evident left-handedness of nature through a spontaneous symmetry breaking at a critical energy. This scenario, if true, would mean that the neutrino should carry a small right-handed component. Although limits on this right-handed component have been derived from free-neutron- and muon-decay experiments, they have not provided the ultimate answer. The proposed experiment investigates the beta decay of unpolarized neutrons into hydrogen. The trick is that one of the hyperfine levels of hydrogen can only be populated where there are right-handed components of the participating particles involved.

An ultracold neutron source could be used to investigate a very wide variety of fundamental problems, including phase topography, Berry phases, time-reversal invariance, strong localization and inertial versus gravitational mass. Experiments currently out of range on existing sources could address interesting questions such as the slope of the quark-quark potential and so throw light on quark confinement. Answers will relate to questions such as the production of heavier elements in stellar processes and the existence or otherwise of an inflationary period in the early phase of the big bang.

The ESS would also produce high intensities of a variety of other particles. Intense pulsed beams of muons are produced via the insertion of a thin graphite intermediate target into the extracted proton beam. High-energy nuclear reactions produce pions over a wide momentum band that decay into muons. These muons could be channelled into an ultraslow beam that would greatly improve the sensitivity and selectivity of muon-spectroscopy experiments on surfaces and thin layers. Likewise, the unprecedented intensity of neutrinos produced would open up a new era in neutrino physics, with a large impact on particle physics, astrophysics and cosmology.

The ESS could also be used to produce beams of short-lived radioactive nuclei that could be exploited in both fundamental and applied studies. The former include nuclear- structure studies as the neutron-to-proton ratio is varied, and measurements of the properties of unstable nuclei that are important in nuclear astrophysics. Possible applications include ion-implantation studies relevant to microelectronics and metallurgy, tumour radiotherapy and the provision of the nuclear data needed for effective transmutation of nuclear waste.

The European Spallation Source

An artist's impression of the proposed ESS campus

The ESS study was carried out by an enthusiastic consortium of neutron users from all over Europe, with representatives from the many areas of science for which neutrons have become an indispensable tool. With funding from the European Science Foundation (ESF), a Europe-wide consultation process was launched. Experts in science with neutrons and in neutron instrumentation met over a two-year period. In parallel, with funding from the European Commission, accelerator and target experts considered the most appropriate technical design.

These activities were overseen by the ESS Council, where ten European countries were represented, chaired by Jorgen Kjems of the Rise National Laboratory, Denmark. One of the authors (JF) chaired the ESS Science Working Group, while Herbert Lengeler of the Julich Research Centre in Germany managed the technical study.

The European Spallation Source (ESS) will be driven by an accelerator producing an average proton beam power of 5 MW (compared with 160 kW at ISIS) at a repetition rate of 50 Hz. The power will be supplied by a 700 m long 1.3 GeV linear accelerator for negative hydrogen ions. After stripping off the electrons, the proton beam will be injected into two accumulator rings that will also compress the length of the proton pulse to 1 us.

The proton beam will be delivered to two liquid-mercury target stations, which will each be differently optimized. One will receive ten pulses per second (and 1 MW of power), optimizing the performance of instruments using mainly long-wavelength neutrons. The second target will operate at 50 Hz/4 MW for high-intensity/high-resolution applications. On each target station, water and liquid-hydrogen moderators will feed 18 independent beam ports, some of which will be equipped with multiple neutron guides. It is currently envisaged that there will be a total of 44 instruments.

The technical study established a site-independent, “green field” design (above) and delivered a cost estimate for the construction and operation of the ESS. The total construction cost is ECU934m (1996 prices and accurate to 20%). Having submitted the technical feasibility study and the scientific case, the ESS project is now moving into the R&D phase.

Five leading European institutions – the Commissariat a I’Energie Atomique in France, the Council of the Central Laboratory of the Research Councils in the UK, the Forschungszentrum Julich in Germany, the Paul Scherrer Institut in Switzerland and Risa National Laboratory in Denmark – have agreed to collaborate in providing the database for the engineering design, and to further minimize the costs and the technical risk. The engineering phase could start in the year 2000 and – if all goes well -the ESS could be fully operational by 2010.

Next steps

Neutrons have made pivotal contributions to a range of scientific areas. This was underlined in a recent critical assessment by the European Science Foundation (see further reading), which also foresaw increasing demand in traditional and new areas, both fundamental and applied. Noting that other non-neutron tools, such as synchrotron radiation sources, could not substitute this demand, the ESF concluded that unless appropriate action is taken, sources of neutrons are likely to decrease in the next ten years (figure 5). The realization of the ESS would resolve this problem in Europe.

This same message has been received in the US and Japan, and both countries have well developed plans for next-generation neutron sources. In 1995 the US abandoned plans for a more powerful reactor facility, the Advanced Neutron Source, but is currently developing proposals for a Spallation Neutron Source at the Oak Ridge National Laboratory in Tennessee. The SNS would be powered by a 1-5 MW accelerator. Meanwhile, Japan is working on two schemes for spallation sources, also in the 1-5 MW range. For Europe to maintain the lead in the field that it has had for the last 25 years or so, it must plan now for new sources, since the lead-time for building such facilities is long.

The scientific case for the ESS shows a tremendous amount of exciting science that would be possible with the source. However, it is always difficult to predict the future. No-one foresaw high-temperature superconductors, and likewise we cannot foresee – despite foresight exercises – the new science of the first decade of the next millennium.

We must, therefore, make sure that the techniques will be available with which to address the scientific questions of the 21st century. Neutron scattering is and will continue to be a tool of paramount importance to the scientific community.