“Getting a high-quality vacuum in an accelerator is a bit like cleaning your house,” vacuum instrumentation engineer Anthony Meunier tells me. “You can put the vacuum cleaner round fairly quickly to get rid of the worst of the dust, but it’s much harder and takes much longer to get into all the corners for those last little bits!” Anthony should know – he’s in charge of procuring all the different components required to obtain, measure and maintain a pressure of 10-9millibar in over 840 metres of vacuum chamber – that’s a trillion times lower than the Earth’s atmospheric pressure!
This is to ensure that when the electrons are whizzing around the vacuum chambers, they don’t ‘bump’ into stray air molecules. If an electron is deflected by another particle, such as is present in residual gas, it decelerates; releasing electromagnetic radiation in the process (also called Bremsstrahlung, from the German bremsen “to brake” and Strahlung “radiation”). The high-energy photons released can interact with the matter of the chamber walls, causing it to emit neutron radiation (amongst other types), for which special safety shielding is required.
Fortunately, Anthony and his colleagues are able to assure an ultra-high vacuum state, using a combination of pumps, gauges, valves and controllers. “First, we pre-pump for a few hours, using turbo pumps to evacuate air molecules to the exterior and to reach a pressure of around 10-6mbar,” explains Anthony. “Next we use a combination of sputtering ion pumps and Non-Evaporable Getter (NEG) pumps to obtain lower than 10-9mbar. NEG pumps get rid of small-mass molecules, such as hydrogen, whereas sputtering ion pumps work better for larger mass molecules and are especially able to pump noble gas such as argon.”
The vacuum system is backed up by highly sensitive gauges that sound alarms in case of pressure increases, and programmable logic controllers (PLCs) that close valves if a vacuum threshold is passed, isolating the rest of the machine from an air leak. In the current machine, around half the cells have remote valves and half manual, but for the new accelerator all valves will be remote.
It will also be an opportunity to upgrade the diagnostic software such as for the Residual Gas Analysers (RGAs) and the hardware and software that allows Anthony and his colleagues to monitor the vacuum in real time in the Control Room. “We have over 1000 pumps and gauges to survey,” he says. “So we need to be sure that any problems are immediately visible.”
There’ll be 14 ion pumps and 10 NEG pumps per cell of the new storage ring, instead of 11 ion pumps and 11 NEG pumps today.
“We can never get a ‘perfect’ vacuum,” admits Anthony. “There will always be molecules of hydrogen, carbon monoxide and dioxide left, but we can do a pretty good job. The volume of all the vacuum chambers combined is around 5000 litres and if we brought all the remaining molecules of gas together, at atmospheric pressure they’d be no bigger than a micron grain of pollen!”
The calls for tender for most of the instrumentation will go out in January 2016 and components will be delivered in batches until October 2017 at the latest, when the assembly phase will commence. “We’ll do as much as possible beforehand, including pumping and pre-baking. Just the final baking will be done once the girders are installed inside the storage ring.”
There will be lots of knowledge and experience to acquire on the new system but Anthony is prepared for surprises, perhaps both good and bad. “One of the most important qualities for working on vacuum systems is rigour, to maintain equipment and ensure that all is working as it should be,” he says.
“I’m lucky to be surrounded by a great team – a mix of experts in vacuum mechanics and instrumentation who’ve been here since the construction of the ESRF and innovative new recruits who are ready to take the baton.”
“My personal wish for the EBS project is to push the development of analysis tools to be able to better see and diagnose vacuum events,” he states. There may be a special reason behind this… As is surely the case in many particle accelerators, a ‘friendly debate’ often rages on between experts over whether vacuum instabilities disturb the beam or whether beam instabilities disturb the vacuum. “Higher resolution diagnostic equipment and fast data analysers will certainly help us to identify the origin of pressure peaks faster and more easily,” he laughs. “Even so, I’m sure this debate will continue regardless for the next 30 years!”