​Thorvald Andersson, Professor Em at MC2. Photo: Michael Nystås

Thorvald put Chalmers on the world map for semiconductors

Thorvald Andersson was one of the people who helped to put Chalmers on the world map for semiconductors in the 1980s. He has now been retired for a few years, but has not yet cut all ties with MC2. “It’s been successful – and above all fun,” says Andersson, reflecting on the past.

He was born outside Karlstad in 1946. After completing a technical upper-secondary school programme in Örebro, he went on to study Mathematics at Uppsala University.
“After that I came to Chalmers and the University of Gothenburg and studied Physics and Mathematics. I had good teachers and found it so interesting that I carried on down that path,” explains Andersson with a smile.

Offered a doctoral studentship
In the early 1970s he was offered a doctoral studentship in the former Physics Department led by Professor Gösta Brogren, and he publicly defended his doctoral thesis there in 1976. Andersson then found time for a year at teacher training college before Chalmers called him back. Professor Brogren asked whether Andersson wanted to take over and develop the MBE activities at Chalmers and build up the MBE Group. MBE stands for Molecular Beam Epitaxy and is a method of making advanced layered structures at atomic level.
“I barely knew what it was, but I decided to accept. With the help of Gösta we first bought an MBE system in 1977. The funding didn’t come from Chalmers, but instead mostly from individuals in Stockholm and Lund. They had been to conferences and/or had completed their doctorates at places such as Cornell, so they understood the development. At Chalmers it was surprisingly quite conservative,” says Andersson.
“An additional MBE system was purchased in 1990, this one with more backing from Chalmers. Eventually students started to complete their doctorates. It was a fun time with so many exciting things going on,” recalls Andersson.

Being emptied
We meet on the fourth floor of the MC2 building. Andersson has a small work corner among removals boxes and clutter near to the laboratory. The premises are now being emptied.
“Before I retired I had my office one floor up; Professor Johan Liu then moved in there. We used the room we’re sitting in now as a coffee and meeting room for the MBE Group,” says Andersson.
He explains further:
“It all started in the Physics Department, but when the physics building was due to be rebuilt, we decided to move to these empty premises. I got everyone on board with the idea that it was better to move than to stay and enclose the machines inside the physics building. Organisationally, the MBE Group then became part of the Division for Microelectronics (MEL) at MC2, but that was mainly a formality. We have always been an offshoot who have looked after ourselves.”

Learnt more in the USA
As soon as he had returned to Chalmers in 1977, Andersson travelled to the USA to learn more about MBE technology. There, he went on several study visits to various laboratories at MIT, Bell Labs, IBM, Wright Patterson AFB and Varian. After returning home, it was time to open the wallet. The first machine cost 180,000 dollars to buy at that time.
“Gösta Brogren forced me to go to the central administrative office at Chalmers and make the payment on the same day that the dollar dipped below five Swedish kronor. The payment was successful despite a technical glitch,” recalls Andersson.
When the time came to buy an additional machine in 1990, the price had soared to nearly SEK 10 million. It was jointly financed by the banks’ newly established research fund and Swedish Central Government, which provided grants for heavy equipment.
“Following my contacts with the above labs, I realised that modern, advanced equipment would be very expensive to purchase. Many people were initially critical, but today equipment costing ten times as much is purchased unhesitatingly. It has been an exciting development.”

The world’s first commercial machine
At MC2 the machines are known as “MBE I” and “MBE II”, but their official names are “Varian MBE-360” and “Varian GEN II Modular”. Varian is the name of the manufacturer, which was based in Palo Alto, California. It remains comparable with “Mercedes class” even today.
“'MBE I' was the world’s first commercial machine bought by a university, and it was the fourth machine of its kind in the world. The first was purchased by Varian’s own research lab, the second by WPAFB in Ohio, the third went to Germany (for use by the postal services), the fourth to Chalmers, the fifth to Cornell and the sixth to Santa Barbara.”

Unique in Sweden
The MBE Group was unique in Sweden when it was founded in 1977, and it continued its activities until the end of 2015. Thorvald Andersson led the group until he retired in 2013.
“The first few years consisted purely of research, which aimed to lead to applications for component groups, but component research was never a big hit in Sweden. It’s impossible to follow the Asian, Japanese ventures, because there is a market there in a completely different way. The USA has the needs of the defence industry and you can obtain money there to supply it with materials; in Asia you have the consumer market. We have none of those things here, and Europe barely does either,” he explains.
Nevertheless, for a time his group was at the forefront of material research into semiconductor materials (quantum structures) and they advanced the research in this field.
“We were working on the technology from such an early stage and were regarded as some kind of pioneers in Europe at first. For a time I was a speaker at several important conferences in the USA, Asia and Europe.”

Wanted a small group
Andersson had the ambition of constantly keeping the group relatively small, with 5–10 members.
“I didn’t want there to be too many of us because that creates other problems. It’s said that a group of more than six people develops what you could call ‘social’ problems. If fewer than five people make up the group, it’s very easy for everyone to get close to each other, and it’s therefore easier to solve any problems. If there are more than six members, there’s always a bit of friction and unfriendliness – it’s what I’ve heard many people say. That’s why I never wanted there to be too many of us, so that it would get unwieldy. The technical side of things was difficult enough so I didn’t want any other problems. But I think it went well and was very interesting.”

Like the pages of a book
Using both MBE machines it was possible to produce layered structures with a thickness ranging from one up to about twenty atomic layers. The key is therefore to manufacture specifically thin layers with an extremely high degree of control. Layer upon layer upon layer.
“You can compare it to a book in which the pages represent the thin layers. When the layers get thinner than a few nanometres, their properties differ from the norm for the host material. MBE is therefore a technology dedicated to making atom-thin layers with quantum properties. Thicker layers require different technologies.”
What are the layers used for?
“We used them to study various quantum properties in physics. Gallium arsenide and gallium nitride can be used for LEDs, such as white LEDs. Other areas of use include high-frequency components in mobile phones, high-frequency transistors that previously ran at megahertz speeds but now run at gigahertz speeds and frequencies a thousand times higher. LEDs and transistors are the largest commercial areas of application today.”

Stellar career path
Thorvald Andersson has had an academic career that went from strength to strength. He became an Associate Professor in 1983 and was promoted to the position of Professor of Physics in 2001. His CV is an impressive, 10-page list of his achievements; prizes and accolades, Visiting Researcher posts, citations, and names of doctoral students.
What do you value the most?
“In the 1980s I had two articles that were very highly cited, and researchers at Bell Labs commented that through those articles I had put Chalmers on the world’s semiconductor map. That was nice. I wrote the articles together with a Russian researcher who was here at that time. The 1980s were perhaps the heyday. Within a short period, some of my articles were referenced more than 100 times, which was quite unusual in those days,” says Andersson.
The group members were the first in the world to produce the results presented in both articles, which were about quantum wells.
“We were also among the first to understand quantum dots. And we worked on antimonides very early on. In this we were on a par with the largest laboratories. Naturally we can’t compare ourselves to the world’s major laboratories; they had completely different resources – not just for expansion but also for analysing materials, and we never had that,” says Andersson.
He is critical of how government funding was distributed between the higher education institutions.
“The funding was often divided between Chalmers, Lund, Stockholm and Linköping. This led to dilution, which were the rules of play that we had to live with, but I never thought that it was particularly good. We might start something, but could never really finish it as it took too long, and in the meantime the major international laboratories that tackled it made much more rapid progress. We saw many such situations very early on here,” says Andersson.

A Visiting Researcher in many countries
As a Visiting Researcher, Andersson worked in departments in large parts of the world for a total of nearly two and a half years: in the USA, Japan, China, Korea, Taiwan, Thailand, Singapore, Poland, Germany, Greece, France, the UK...
“I’ve been in all those countries as a Visiting Researcher for short or long periods of time – visits ranging from a few days to up to a year. It has been extremely exciting.”
He has also received several accolades for his work. In 1999 he was awarded a prize for best business concept in the Venture Cup (a competition for the entrepreneurs of tomorrow), in 2009 he received SEK 100,000 from Teknikbrostiftelsen (a foundation promoting contact between academia, industry and the local community), and as recently as in 2011 he came third in Lärosätenas Idétävling Väst (a competition within higher education institutions for utilisation of ideas). Among others. 
“I have also worked on other types of semiconductor materials. Towards the end of my career I worked on molecular semiconductors. They constitute an area for the future that is big in Linköping and Stockholm, for example.”
In conjunction with his nearly one-year period as a Visiting Researcher in Japan in the mid-1990s, Andersson started working on gallium nitrides, which were a novelty at that time.
“NTT outside Tokyo, where I was, is one of the largest laboratories in the world.”
Over the years Andersson has supervised more than 30 doctoral students and licentiate students. In the 1980s and 1990s there was a great deal of interest among students, but it has subsequently waned here as in most Western countries.
“Today it is probably difficult to attract students to this area. Just like in the rest of society, young people like to follow what is covered in the media and perhaps see the flashiest topics, and less of what may be interesting in the long term.”

Took part in establishing MC2
Thorvald Andersson was very highly involved in establishing MC2. Initial discussions began already back in the 1970s. A formal steering group was set up in 1990, with Andersson among its members. He has been involved throughout.
“I was the person to come up with the initial idea of building a new laboratory in 1977, which much later became the cleanroom and MC2. I had seen major laboratories, mainly in the USA, and understood what was happening. But at Chalmers initial discussions were only about equipping a room, moving a room, adding a room and suchlike. The opinions from Physics were solely negative. So I suggested to Gösta Brogren that it was imperative to seriously consider daring to construct much larger purpose-built premises, and to combine the Physics part of the Electrical Engineering Department and the Electrical part of the Physics Department. Gösta seized the opportunity and tackled it; at first, he was doubtful, but he soon changed his mind,” says Andersson smiling.

Started predecessor
He also helped start the Centrum för halvledare och mikroelektronik (a centre for semiconductors and microelectronics), which was also a predecessor of MC2. Its past chairs included Professors Torkel Wallmark and Olof Engström.
“But as soon as something becomes formal, many people get involved and it can be difficult to identify the actual mutual interest. And that’s when I often lose my commitment and shift my focus to something else instead. It’s not a straight line but a very winding path. Everyone has to agree. I’m more informal in my way of working,” says Andersson.
However you look at it, you can see that the MBE technology and the machines, the idea of constructing the new premises and a “centre” acted as a catalyst for establishing MC2.

Continuing in the cleanroom
The MBE activities are now continuing in the Nanofabrication Laboratory – the cleanroom – which already contains two equivalent systems. The old machines are being phased out and dismantled; a few components that can still be used are being saved.
“In principle, you could clean them thoroughly and sell them. But I expect that there are so many of these for sale around the world that it would probably be difficult to find a buyer,” says Andersson.
How do you feel about things now?
“I’m not a sentimental person. The machines are basically just stainless steel. People have often developed a personal relationship with old equipment, but the nature of development today is such that what was once hot will eventually become old. Two to three new research fields will emerge in the space of 10 years. The only possible fear is that it is taking too long to get new research fields into the laboratory.
You don’t feel that it is knowledge that risks being lost?
“Yes, of course, but that’s all part of the game,” he says.

Kept in contact
Thorvald Andersson has been a Professor Emeritus since 2013. He has not lost contact with MC2 entirely, although he has not retained any teaching duties.
“I visit occasionally and meet old colleagues. We have quite a lot to discuss, sometimes for several hours.”
He spends most of his time with his wife, Ingegerd, who is a retired psychologist, the families of their three children including their eight grandchildren, their sailing boat and their holiday cottage in Bergslagen.
“It’s all go, all of the time. I also read a lot when I’m at home. At the moment I’m reading three books in parallel, including Herman Lindqvist’s De vilda Vasarna (a violent history of the Vasa family). I’m also reading a book about de Gaulle and one about the Kremlin. Before that I read Äntligen diktatur, (Finally dictatorship), which is about Thailand. I sometimes read novels, no crime novels, but mostly books that are linked to reality.

Writing down his memories
Andersson has also started documenting his experiences, precisely as a way of harnessing the knowledge that he has acquired over the years. He doesn’t yet know what the new project will result in, or even whether it will ever be completed. It might end up being a combined history and autobiography.
“It’s not exactly going to be finished tomorrow – I’ve only completed a few pages so far. When I actually sit down and write, I keep going off on small tangents that you can meander down endlessly. There’s no plan; I’m just writing, seeing where it leads to and whether it can be used. I hope to have something finished in the course of 2017. It’s all quite exciting,” concludes Andersson.

Text and photo: Michael Nystås

Published: Tue 14 Feb 2017. Modified: Wed 15 Feb 2017