by Panos Charitos. Published: 20 July 2013

Q: Which were the main developments of stochastic cooling since it was originally invented and applied at the ISR and what's the role of CERN?

A: The invention of stochastic cooling at CERN in the 1970s was a high point of the lab’s history. As everyone knows, it was crucial to achieving antiproton beams of sufficiently low emittance (or beam size) to be collided in the SPS, leading to the discoveries of the W and Z bosons.

A high bandwidth pickup detects fluctuations in the particle distribution of a beam and selectively suppresses them with a kicker at another point in the ring. Over time this reduces the spread in amplitudes of transverse or longitudinal oscillations in the beam, making it “cooler” and smaller. But there are many subtleties in all this and design and construction of the necessary hardware can be very challenging.

A good account of the physics and some of the history can be found in the recently published book by our late colleague Dieter Möhl, one of the pioneers, along with Simon van der Meer, Lars Thorndahl and others. It was implemented on various storage rings at CERN and later in other labs.

For many years, most applications were to rings with coasting beams, in which the particles are spread continuously around the circumference rather than being bunched by an RF system as they are in the LHC. Stochastic cooling of bunched beams was studied in the SPS in the early 1980s by Swapan Chattopadhyay and others but turned out to be very difficult for the parameters of those beams. However bunched beams had been cooled in the very first implementation in the ICE ring at CERN and were again more recently at Fermilab and RHIC.

The cooling of full-energy, bunched, colliding heavy ion beams (gold and uranium) at RHIC has been spectacularly successful. It is the first time that hadron beams have been cooled in collision so you can see the luminosity growing with time from the beginning of a fill. Previously at RHIC (and at the LHC) it would decay as the beam emittance grew due to intra-beam scattering. The integrated luminosity collected by the end of a fill is much higher with the cooling.

Stochastic cooling works much better for heavy ions than for protons because the number of particles per beam is generally smaller so the charge fluctuations that the system works on are more significant. The large charge of the individual particles also helps.


John Jowett in the CERN Control Room during one of the heavy-ion runs at the LHC.

Q: How did you become interested in applying stochastic cooling technique to heavy ions? Have you worked in this subject before?

A: I have not personally worked on the subject before but, following the success of the stochastic cooling of heavy-ion beams at RHIC, it is natural to ask whether the technique could also be exploited at the LHC. Indeed I recall some coffee conversations with Dieter Möhl in recent years where he was cautiously optimistic about the possibility. I had previously collaborated with Mike Blaskiewicz who was one of the driving forces behind the RHIC cooling system. We discussed the possibility in 2011 and he presented some first ideas at the International Particle Accelerator Conference that year.

Recently we invited him over to spend a couple of weeks at CERN to look at more details and discuss with other CERN experts, including Fritz Caspers, Wolfgang Höfle, Manfred Wendt and others. He also set up simulations together with Michaela. We have greatly benefited from his advice and experience.

Q: Do you think that stochastic cooling could be applied to the LHC? In which areas?

A: These first studies show that a cooling system could be very effective for the heavy ion beams in the LHC. This is perhaps not obvious since the LHC is different from RHIC in that the luminosity decay is dominated by the so-called “burn-off”, the inevitable removal of particles from the beam by the collisions themselves, rather than emittance growth due to intra-beam scattering. Still, the reduction of the emittance from the cooling would allow more collisions to be squeezed more quickly out of the remaining beam population, substantially increasing the effective average luminosity.

Q: Is your team currently working on these issues and which are the open topics that you are studying?

A: However there are certainly some substantial technical problems to be solved. For example, the kickers would have to open and close around the beam only after it is accelerated to full energy. As Fritz pointed out, we could not simply copy the RHIC design as these structures, even when open and inactive, would overheat in the presence of the high-intensity proton beams of the LHC. The high-bandwidth amplifiers needed would also be quite demanding to build. On the other hand, it seems that it would not be necessary to set up microwave links to transmit signals across the LHC ring above ground. Fibre optic links over shorter distances in the tunnel should be adequate.

Clearly, it would take a few years to design and build such a system. The concept is at an early stage and still has to be transformed into a real project with all the necessary collaborators. Our heavy-ion team is quite small and we are also working on other ideas for boosting the heavy-ion luminosity in future years.