by Panos Charitos. Published: 22 January 2013

Detlef started his career as atomic physicist. He studied atomic and X-ray physics in Dresden/Germany and spent many hours as a user of ion sources. Even before finishing his PhD he had applied atdifferent laboratories and that’s how he got a contract with CERN where he worked for the first time as an operator of ion sources. Detlef’s PhD thesis was on simulation and diagnostic of the electron impact on ion sources, in this special case ECR ion sources and EBIS.

A.M. Can we say that the study of “ion sources” is a particular field in physics?

D.K. I wouldn’t call the study of “ion sources” a particular field of physics as it is rather a mixture of physics and engineering. It is based on knowledge atomic physics, plasma physics but also a lot of engineering, material science and so on. It is a truly interdisciplinary field and that’s why I think it is so interesting.

A.M. How can you describe the ion source used at CERN?

The source that we have here is of ECR type which stands for Electron Cyclotron Resonance. In this type of source you have a magnetic bottle created by a set of coils and a permanent magnet and we inject some oxygen gas, some lead vapour and microwave with a frequency of 14.5 GHz. The microwave in principle heats electrons which then create the plasma, confine the ions and also make the ionization to the higher charge states that we need.

In the source oxygen is the main part while lead is only a small fraction of the total composition of the plasma. The oxygen has two main effects, it creates on the plasma chamber wall an oxide layer which increases the secondary electron emission. In that way the density of the electrons in the plasma can be increased. That’s why oxygen is very good otherwise we could use nitrogen or different gases but it has been proven that oxygen is the best!

In addition, the oxygen creates an ion cooling effect. The ionization is stemwise - Pb+1, Pb+2 and so on up to Pb29+ which is produced in our plasma after about 30 Milliseconds. So since it is a stemwise ionization you need to keep the ions for some time in the plasma. Due to collisions the ions are heated to a certain temperature and can escape the plasma, but when the heavy lead ions collide with the lighter oxygen ions they transfer energy to them. Then the oxygen ions leave the plasma and take away some kinetic energy – but we don’t care about that - while the lead ions in which we are interested stay longer in the plasma. This is the cooling effect.



Fig.1: Detlef Kuchler at CERN in front of "his" machine, LINAC 3. (Image: CERN)


A.M. Can this source work with many different types of ions?

Most of the sources are multi-purpose sources and they can be used as sources for many different species. From very light elements (i.e. hydrogen) up to the very heavy ones like uranium. If you find a way to put the material in the source then it can be used to give you what you want.

A.M. What changes when we use the source for heavy ions compared to protons?

As I mentioned it is the way that the material can be put in the source. So in the case of protons which are extracted from hydrogen it is very easy since hydrogen is a gas and we only need a gas bottle. But for lead, since lead is a metal we are using an ohmic heated micro-oven which has the size of a finger (a palm-sized cylinder) and there we actually evaporate the metallic lead. The heating needs to be controlled very carefully as otherwise liquid lead will spill out into the plasma chamber of the source.



Fig.2: Detlef Kuchler holds a purified sample of lead used to create heavy ions for the LHC


A.M. Is this an expensive method?

The use of the micro-oven itself is not very expensive. The main cost comes from the fact that we use isotopical pure lead 208Pb. As the natural lead has four stable isotopes we would lose almost half of the beam already in the low energy part of the Linear Accelerator due to the composition of the natural lead. That’s why we are using an enriched isotope, the 208Pb which comes at an expense of about 1$ per milligram. Of course this is a rough estimation. Think that if you order 10 grams they will come at a cost of $13.000 dollars but they are enough for almost two years of operation. So it is not that expensive compared to other things as e.g. electricity!

A.M. For the upcoming proton-lead run which are the special requirements that you need to take care of?

There are always two things: the first is the stability over a long period and the second is the intensity of the beam. They both depend on the source settings. In order to set up the source in the right way I should mention that experience plays a great role as you always need to balance between the two. So if you push the source to a very high intensity then your stability will get limited to 10 minutes or so. After that your intensity will drop and you have to retune the source. That means that you can go up to a certain limit which currently guarantees stability for almost 10 hours of operation. Keep in mind that the source needs to be constantly retuned. We don’t set up once and then leave it running.



Fig.3 Detlef holding a piece of lead used for ions (Image:CERN)


What is the difference between the ion source used at CERN and in other laboratories?

It is the technology on which the source is based. So at RHIC for example they are using an EBIS (Electron Beam Ion Source) and not an ECR one. During the last years they developed this new source, which is quite sophisticated and is a very special source. In fact we tried to scale their results to see if we could use a similar source but it seems that for our requests this source does not deliver enough intensity. Perhaps I could add that the main difference is that an ECR source delivers a high or medium current of high or medium charge states while the EBIS delivers a lower current of very high charge states. So you have to balance what you really need. For example in LINAC 3 we get Pb29+ from the source, accelerate and send it through a stripper where we lose more than 80% of the beam due to the stripping efficiency for the different charge states but we still get a sufficient current that could be sent to the LHC. At the end of the LINAC we get a current of 20-25 µA of Pb54+ during a pulse of 200 μs.



Fig.4:This small bottle contains the lead source for Linac 3, which provides lead ions for collisions in the LHC (Image: CERN)


A.M. Are you planning to increase this current following the recent upgrade plans for ALICE? Which are the main constraints?

We are certainly planning to do that but we need to perform several studies to understand how the beam behaves in the machine (beam diagnostics). We also have some ideas on how the current from the source could be increased and to accelerate several charge states in the machine. However we need man power and more time to do that. Bear in mind that for the next years we won’t have to deliver only lead ions for the LHC proton-lead run but also argon and xenon for the NA61 experiment. In addition we want to modify the machine to deliver light ions for biomedical experiments.

A.M. How fast do we change from one type of ions to the other?

At the moment we have only one source and it takes some weeks to change the type of ion that the source delivers. Especially with the metallic ions it takes several weeks to stabilize the source. We are also only a small group with only four source experts while of course there are several technicians and engineers working on the different parts of our machines.