Research facilities possess a mix of studiousness and technology. When walking into the main building of the Diamond Light Source at Harwell, Oxfordshire, the atmosphere is muggy with academic cleverness and engineering endeavour. You’re going through the looking glass into a technological wonderland.
Diamond Light Source is the UK’s national synchrotron and has a reputation of being one of the country’s most under-celebrated technology gems. On a basic level, it’s a large X-ray microscope, powered by electrons, which allows researchers to study molecules and atoms that make up materials. A diverse range of items can be analysed and understood better here, from deadly viruses and great works of art to industrial turbines and battery chemistry.
The modernity of the building’s design is striking. The facility is designed around the synchrotron’s main storage ring, a vacuum tube between 40 and 100mm in diameter, set within hundreds of magnets which focus and bend a stream of electrons travelling almost as fast as the speed of light. The electrons are generated by the linac (linear accelerator) and transferred into the booster ring, where they are accelerated and then shot out into the main storage ring. Spidering off the storage ring at tangents are 33 experiment stations, which make use of the intense light produced by the electron beam, which is 10 billion times brighter than the sun, to peer inside materials.
It’s rare to get inside a synchrotron. The guts of the mammoth machine are normally locked away in a concrete vault to shield the radiation produced. And Diamond Light Source is heavily used by researchers, keen to access the 5,000 hours a year of beam time. Shutdown periods, when engineers and technicians maintain and upgrade the machine, are kept to a minimum.
The basis for the facility, which has cost £450 million to date, and the starting point for the tour, is inside the middle of the ring. Here a conspicuously small electron ‘gun’ fires groups of electrons into the linac at 90,000 KeV (kilo electron volts). The gun, which is about the same size as a small enclosure of computer servers, essentially uses the same technology as an old cathode ray tube television to produce electrons.
The linac increases the energy of the electrons to
100 million eV and focuses them into a beam. Power is increased by ‘injecting’ RF power, radio waves, into the electron beam at various points to speed up the electrons.
From the linac, the electrons are next shot into the booster ring, where their energy is increased to 3GeV (giga electron volts), again by injecting RF. The electrons are then deflected out of the booster into the storage ring, where they accumulate to a stored current of 300 milliamps. The electrons are bent round into the ring, actually a 48-sided polygon, and the photons come off down tangential beam tubes at each bend point.
The electron beam is reduced in size to tens of microns in diameter by strong focusing magnets at each straight section where powerful arrays of permanent magnets called insertion devices are installed. These insertion devices cause the beam to undulate many times and each time another pulse of light is produced. The pulses are additive, making the light produced from them many times brighter that the light which could be obtained from a single bend. The insertion devices are what distinguish Diamond as a third-generation synchrotron.
There are 24 sets of magnets and straight sections around the ring, each representing a cell, within which the tubes branch off to experiment stations. It is the number of cells that ultimately sets the beamline capacity of Diamond.
As the electrons pulse around the storage ring at a rate of 500,000 times a second, the photons, or light, are siphoned off at points and feed into the complex machinery scientists use in their experiments. This light covers a wide and useful spectrum, from infra-red, through ultra-violet to soft X-rays and hard X-rays.
Jim Kay, head of engineering at the Diamond Light Source, says: “In the booster you synchronise the increasing energy of the electrons with increases in the magnetic field to keep them in orbit. This is the origin of the term synchrotron. In the storage ring we build up the stored current to increase the brightness of the light. The electromagnets squeeze the electrons together and resist the forces of the electrons repelling each other which makes the beam size increase.”
The biggest apparent difference between the booster ring and the storage ring is size – the electromagnets are bigger and more numerous and the vacuum tube is larger. This is because the higher energy electrons require a larger force to be ‘squeezed’ together and manipulated and because electrons are stored for days at a time. Any electrons that are lost are absorbed by the walls of the vacuum tube. Any excess light is absorbed by small plates of copper cooled by water.
The storage ring, which has a circumference of 0.5km, is housed in a wide, bright white curving corridor. The ceiling of the corridor, which consists of thick concrete strips, can be lifted in and out of place to enable installation and maintenance of the storage ring and insertion devices. Placed at angles on the floor from the middle of the corridor are large girders, onto which the equipment is placed.
Nigel Hammond, senior project engineer, explains: “The girders upon which the equipment is mounted are set up with ultra-precise positioning and tolerances.
“Each 5-7m long girder has to be machined to within 25 microns flat so that the magnets can be accurately positioned around the storage ring. Vacuum vessels on each girder are attached to vessels on adjacent girders by metal bellows to allow for movement.
“Each girder can be adjusted in five degrees of freedom from a cam system, so we can adjust the position of each girder within an accuracy of 30 microns. The resolution is smaller than we can measure.”
The storage ring itself looks surprisingly bare bones. Nothing is encased, indicative of the uniqueness of the machine and the frequency of upgrades. The vacuum tube is surrounded by various large electromagnets, in different colours to indicate the four different types that are used to manipulate and focus the electron beam. The inside wall of the corridor carries masses of electrical ducting and pipework. Inside the main storage ring, air and water are controlled to plus or minus one degree of 22ºC to limit alignment errors that may happen because of thermal expansion. Demineralised water cools electromagnets and vacuum vessels.
“In my job, one minute you’re listening to a scientist’s idea for some kind of ground-breaking experiment, and the next you’re talking to a plumber about a leak in a pipe. It’s very varied,” says Hammond.
Kay says: “Equipment also has to be renewed because the technology has moved on. The prefix nano is creeping into the titles of equipment and experiments, showing we’re moving from microns to the nanometre scale. This needs an even brighter and stable source of light.”
There are normally five three-week shutdowns a year, but during next summer an eight-week shutdown is planned, during which an entire cell will be stripped down to reconfigure the magnets so that another source of photons can be fitted into the ring. This will enable another beam of light to be siphoned off. Next year will bring a “massive” amount of work, says Hammond. A new configuration will be tried at Diamond, a Double Double Bend Achromat. This will result in one cell, the machines, all of the magnets and vacuum vessels being removed and replaced to a new configuration.
Just as the electron beam is squeezed, engineers at Diamond are determined to squeeze more and more usefulness out it. Kay says: “When Diamond was first built, it was one of the brightest synchrotrons in the world. But we are having to innovate more to catch up with the technology in newer synchrotrons. We’re using more powerful magnets to squash the beam down tighter. We’re making it brighter by placing the pole tips closer to the beam and shrinking the size of the vacuum tube.”
Kay says there is always a high demand for time at Diamond. The range of research conducted is staggeringly broad. Experiments are bespoke designed and built for scientists by the engineers at Diamond. The offices, workshops and laboratories are arranged in a circle around the ring, and the end stations are arranged into ‘villages’ so that common tasks can be more efficiently conducted. There is a surfaces and interfaces village, an engineering and environmental village, a materials village, a macromolecular crystallography village, a spectroscopy village and a soft-condensed matter village.
Biological research makes up more than 45% of the work carried out at Diamond. Researchers use crystallography beamlines to look at medicines, vaccines and diseases. The scientists can examine up to 300 samples in 24 hours, observing the atomic positions of the molecules inside the samples, to see how amino acids fit into protein chains, for example.
The instrument works by shining light onto the sample, which is then diffracted. The diffraction spots are detected using a detector camera on the other side. The sample is rotated and moved so more data and a more accurate image can be built up.
Engineers start by designing the machine from a scientist’s requirements. Martin Burt is one of the engineers who develops the instruments used to look at biological matter. “There is a negotiation between the scientists’ requirements and what is feasible through engineering,” he says. “There’s a focus on getting smaller and smaller, getting more data, and better resolution. The samples also have to be maintained at temperatures below 100K.”
Increasingly, research is done remotely. Samples are sent to Diamond, the scans conducted, and the data sent back to the researcher over the internet. This is part of a wider trend which is seeing the amount of remote handling and automation at the end stations increase. The aim is to increase the throughputs of samples and reduce manual handling, to improve the effectiveness of experiments and make better use of beam time.
Diamond researchers often work with industry. Recent high-profile projects were with Rolls-Royce, Tata Steel and fuel-additive manufacturer Infineum. Much of this industrial research is done using Diamond’s IL12 beam line, which was sent outside of the main building so that a larger research cell could be built. Here, an array of materials have been placed in front of the synchrotron’s light in a bid to gain insight into the various processes happening at a molecular level. Many objects and materials are analysed, from simulated lava, fossilised squid ink and dinosaur feathers, to running motorcycle engines, super alloys in jet engines and welding materials.
Mechanical tests can also be conducted on samples while they’re in the beam, to extract data on fatigue and stress as a process is happening. The researchers also conduct X-ray tomography of components during corrosion, and cracking in composite materials.
Often research is conducted to validate computer modelling. Diamond experiments often reveal key missing information. Kay says: “We did an experiment looking at the dynamics of welding here. We had a molten weld pool dosed with micron-sized beads of ceramic that wouldn’t melt. We tracked the path of the beads in the weld pool to follow the currents of molten metal. It makes for interesting viewing, especially when what you think is going on at molecular level isn’t actually what’s happening at all!”
Particularly interesting research is happening in the area of powder diffraction. It is focused on developing materials for hydrogen fuel cells. Companies want materials into which you can pump gas, hold it, and then release it. There’s also strong demand for materials that improve batteries.
Unique to Diamond is another machine that has been designed to host long-duration experiments. A range of samples is held on a table – including Arctic sea ice, concrete being degraded by uranium, a pharmaceutical drug under different levels of humidity, and a battery. Once a week, the experiment is activated and the samples are tested over the course of several months to years. The resulting data allows researchers to find out how materials change over time.
Diamond is now halfway through its planned life. New technology such as a low-emittance beam, which will increase the brightness by a factor of 10, and the work of its imaginative scientists and engineers, will ensure the centre’s operation for the rest of its expected lifetime and beyond.