Editor's note: This is one of four stories describing the experience of Dutch PhD students working in both academia and industry. In the first instalment Erik Kieft, 3 years into his PhD, tells us why he chose to do a PhD in a commercial setting.

My PhD project focuses on the analysis of new light sources called extreme ultraviolet (EUV) radiation-producing plasmas. One potential application for these lies in the production of computer chips, allowing them to bear more transistors and thus increase the speed of computers in an unprecedented way. This research project is a cooperation between ASML, the world's largest manufacturer of lithography systems, and the Eindhoven University of Technology (TU/e).

My interest in doing research in a commercial environment started when I was looking for a place to do my master's project in June 2000. I knew that I wanted it to involve plasma physics, but I also liked the idea of going outside the confinement of our own department and get a taste of what it is like to do research in a different working environment. Therefore, I gladly accepted a graduation project at the Netherlands Organisation for Applied Scientific Research (TNO), a large Dutch research institute that acts as a mediator between universities and the industry.

When my master's was nearing its end, I started looking for a PhD position in the same field. Soon enough, I heard about a cooperation between EPG, the other of two plasma physics groups at TU/e than the one I was a member of, and ASML on the subject of the characterisation of EUV plasmas. My current supervisor at university asked me if I would be willing to fulfill one of the two PhD positions. I had done a short internal traineeship in his group 2 years earlier and apparently I had left a good impression. Although it is not uncommon for both of the university's plasma physics groups to do research funded by industry, this project was different in that a large part of the actual work would be done at ASML itself.

At that time I had already considered other positions, but I chose this particular PhD for a number of reasons. In the first place, I thought (and still do) that it was a highly exciting project, being a new application of plasma physics. Then, it offered me the opportunity to approach a problem in different ways, and to use a wider range of diagnostics tools. Finally, I also saw the cooperation with an industrial partner as an added bonus.

Since I've started my PhD I've been dividing my time more or less evenly between the university and the EUV laboratory at ASML. Thus, while so far I have done all my experiments at ASML, I go to the university regularly to meet my supervisor there (I also have one at ASML), work on the theoretical aspect of my project, write reports and papers, and visit the library.

A detour for practical reasons

At the moment I'm also preparing a new experimental setup there, which will be moved to ASML later. Two practical reasons forced me into this little detour; one is that TU/e is paying for the setup's parts; the other is that they have more lab space available at the moment. I feel lucky to be given good technical support both at TU/e and ASML. I also think it is only natural that ASML 'borrows' my expertise every now and then for experiments that are not directly related to my thesis when quick results are needed.

Even more importantly, ASML gives me all the freedom that I need to determine my own approach to my research. I never felt that ASML was pushing me too much in one direction or another, which I know is one of the common criticisms that comes with industry-funded research. Supervision both at ASML and the university takes place in a rather informal way. Especially in the beginning of the project, my supervisors and I had regular meetings. Naturally, they look at my results from a different perspective: My supervisor at the university focuses more on scientific output, while the one at ASML is more interested in practical implementation. In general it is my responsibility to keep both supervisors up-to-date of developments 'on the other side'.

My colleagues in the project are, at the moment, one other PhD student and one master's student, who I supervise. We have regular discussions about different aspects of work, but we generally don't carry out any experiments together. Each of us has his own project and responsibilities, even though we all work on EUV sources.

So, what does it take to do a PhD in an industrial environment? I believe that the kind of skills I use when doing my research at ASML are not much different from the ones I would use if I was doing a more traditional PhD solely within a university. The main difference is that in my position you need to be more able to keep track of your own progress. This is a natural consequence of working in two different locations: Neither of your supervisors--at the university or in the company--will be involved in all parts of your work on a daily basis, and therefore you are ultimately the one who can assess whether you're going in the right direction or not. If you find it difficult to work independently or ask for help when you need it, then it might be wise to choose a different setting for your PhD.

Also, generally speaking, while within university you are often in a group of people with the same scientific background, in a company you might be the only one who has a certain expertise. That's probably why the company wanted to have you as a PhD student in the first place. At ASML, for example, while most people have a technical background, they are usually not experts in plasma physics per se. The result is that I have to put more efforts into explaining my work, which helps me hone my communication skills.

In addition to the opportunity to gain extra skills, doing a PhD in a commercial environment also implies an intense, daily contact with people from the industry, which is to my eyes one of the greatest advantages of my PhD. Although the work that I do is of a relatively fundamental nature, I find it very inspiring to be working so close to the place where, hopefully, the results of my work will one day be applied to create new and better computer components.