Interview: Ursula Keller, saving the sight of millions

Stuart Nathan talks to Ursula Keller, who overcame significant challenges to rise to the top of her profession in physics, specialising in ultrafast pulsed laser technology and helping to improve the vision of millions

Millions of people owe the acuity of their eyesight to Ursula Keller. The semiconductor saturable absorber mirror (SESAM) she invented while working at AT&T Bell laboratories in New Jersey in the 1990s is the crucial component for converting continuous laser light into pulses lasting from picoseconds to femtoseconds. It turns lasers into a surgical tool that can make millions of tiny incisions per second, a property which is used to create diffraction grids on the surface of the cornea in femtosecond Lasik surgery – the most common surgical treatment to correct defects in vision.

In addition, such lasers have become crucial to manufacturing many electronic components and are also finding uses in extremely accurate clocks.

But none of this came easy to Keller. Growing up in a working-class background in socially conservative rural Switzerland during the 1960s and 70s, she had to face the all-too-common prejudice against women studying and entering careers in science. Moreover, she is severely dyslexic. At the time, diagnosis of dyslexia was far less prevalent than it is today with the condition barely recognised.


Despite this, her determination and aptitude for mathematics and physics led her to university, where she studied physics engineering at the Swiss Federal Institute of technology (ETH) in Zürich. After graduating, she secured a scholarship at Stanford University in California, where she switched to pure physics and gained a masters then a doctorate in applied physics.

She went on to work at Bell Labs in New Jersey before returning to ETH as the school’s first female full professor of physics, becoming director of the Swiss National Research Centre for Ultrafast Molecular Sciences and Technologies. She founded the ETH Women Professors’ Forum as a support network to help female researchers advance their careers, both in terms of their position in the university and in their research, and was last year honoured with a lifetime achievement award from the European Patent Office.

Keller explains: “In Switzerland there was a test you took at the end of primary education to decide what sort of secondary school you went to. The top level only took the highest 20 per cent and I didn’t get into that, so I was supposed to go the next level down where you were trained mainly to go into vocational jobs. So I went to the government advisory service for aptitude tests to see what kind of career I would be best suited to.

“They also did some general tests, including maths, and those came out really well because they had graphical sections on geometry and I was really good at those.

“The adviser said I should stay in school. As I got older, my grades were not so bad any more and I got a bit better at languages, although when I started I could barely read and the teachers said ‘you don’t belong here’ but my grades were so good in maths they couldn’t understand it and the other students would say ‘Oh Ursi, you explain this to us’ when the teacher couldn’t teach the problems.”

Keller believes the effects of her dyslexia reduced as she got older. “I was still never able to learn anything by heart and I still have trouble with that, but I learned over the years to organise things in my head so I could remember them better. In a way I think that is what makes me a very efficient data scientist. I can take basic maths and apply it so I can see things very clearly and that’s because, early on, it was the only way I could memorise things.

“This allowed me to transfer to the top school, the ‘Gymnasium’. Things were better there because I could specialise. For example, I never had to do Latin and my languages were good enough to not hold me back – although my final results were top grades in all the sciences and just good enough in the languages.”

Physics was Keller’s primary choice of science. “I was pretty strong in all of them but I felt to progress in biology and chemistry you still had to do a lot of learning by heart and, although I’d found ways of organising information, the way they were taught – at least at the time – didn’t suit me. I can see now ways of learning chemistry but at the time I’d look at a chemical reaction and see all the possible ways it could go and the teacher couldn’t explain to me the way it would actually happen. I’d just have to remember it. Maybe if they’d explained something more about energy levels and quantum mechanics I would have understood it better.”

Ursula Keller
Millions of people who have had laser eye surgery owe their improved sight to the genius of Ursula Keller

After finishing secondary school, many of Keller’s friends took a gap year and went travelling. Although she wanted to do likewise she couldn’t afford it so she applied directly to university.

“I was attracted to physics theory but I was always told it was difficult to get a job if you specialise in theory. I made the decision at the last minute to go towards experimental physics, basically because I needed to get a job. I still wanted to travel after university and I needed to fund that. While I was at ETH, I found out you could apply for a fellowship which would allow you to travel and be funded. I applied for the Fulbright fellowship, graduated top of my class and won a full five-year fellowship. I was accepted at every US university I applied to but decided to go to California.”

A professor at ETH offered to help her because he had been impressed with her grades. He suggested she spend some time in the UK to learn English. She says: “I really like the outdoors and skiing so I decided to ask Heriot-Watt in Edinburgh and I was accepted there. I took some English classes, so I learned with a slight accent, and had a great time.

“It was the early days of optical computing, around 1984, and in 1979 I had seen a demonstration of lasers back in Switzerland and become intrigued by that. At the time, optical computing was desperately oversold because they had only one or two types of logic gate but the potential was obviously there for physics theory to have a major impact on engineering.”

At Stanford, Keller began to study laser physics more intensively and went on to make experimental laser physics the subject of her PhD.

“That stood me in good stead when I went to Bell Labs; because I was an experimental physicist they counted me as an engineer and I had a higher starting salary. At Stanford, there had been no looking down on engineering, everybody was always looking at applications, which was a difference from Switzerland and Germany where physicists tend to be perceived as having a higher status than engineers. At Bell Labs, there was a culture that when you wrote up your paper from your research it would be suggested that you take out a patent. There was a lively start-up culture around both Stanford and Bell Labs and the importance of intellectual property was well understood by everybody.”

It was at Bell Labs where Keller first developed and patented what became the SESAM technology.

“It was called something much more complicated then,” she recalls. “I learned a bit later on that when you make an invention it’s very important to think up a good acronym for it.”

Chain reaction

SESAM is an adaptation of an essential part of lasers. Laser technology depends on a crystal which has the correct optical properties to amplify light as it propagates through the transparent medium; in simple terms, electrons are boosted to high energy levels and, as they fall back to lower energy states, they trigger a chain reaction where they cause electrons in neighbouring atoms to also drop through energy levels.

Mirrors at either end of the crystal bounce a light pulse backwards and forwards, promoting electrons through energy levels and causing more cascades as the energy builds up with each pass. A mirror at one end is only partially silvered, allowing some of the light to escape in a beam.

When I started I could barely read

Keller’s innovation was to replace this partially silvered mirror with a semiconductor film that acted as a mirror but only up to a point. Once this point, known as saturation, is reached the semiconductor becomes completely transparent for a very brief period, allowing a short pulse of 5 per cent of the energy contained within the crystal to escape. This destroys the saturation and the film becomes reflective again; then the cycle repeats. This produces the characteristic short pulses, with the precise composition of the semiconductor determining the saturation level and hence the length of the pulse.

“Bell Labs made a lot of money from the patent in licensing it out to major industry and whatever they paid me as a salary they made back many times over, which I think is good, in fact,” she says.

One major advantage of laser pulses produced using SESAM is they do not heat the material that they strike, but effectively ‘push’ material from the surface into a plasma in a process called ‘cold ablation’. This makes a picosecond laser very effective for cancer surgery, for example – because they do not damage surrounding tissues by heating them – and also for machining very small features in solid materials.

As there is no change in temperature, there is no change in composition or crystal structure surrounding the machined feature, which is very useful for semiconductors where both composition and crystal structure is vital to electronic properties.

The perfect CV

After four years at Bell Labs, Keller decided to return to academia and ETH. There was a drive to recruit more women in senior positions from the president of the university, and Keller was invited back to take up a professorship in physics to carry out research into new areas. “They remembered I had been top of my year and, by that time, I had the perfect CV.”

Once back in Zürich, Keller was surprised to find there was no patent office. “The policy was, you need an industrial partner and will only take out a patent if

industrial partner pays for it. There was no culture of starting spin out companies.” Keller was instrumental in changing the situation and ETH set up a scheme where researchers who want to set up a company can use the university’s facilities for two years.

Another innovation of Keller’s at ETH was the establishment of the Women Professors’ Forum. “We were fewer than 10 per cent women at the university and most of us didn’t know each other. Very often there will be only one woman in a department and comparatively few female students, too. Often we would find the power structure in the departments, when it came to funding and appointments, was concentrated in established circles which is difficult to break into if you don’t have allies. We had no clue what we could actually ask for.”

The forum allows women at the university to build up networks and also creates a group of role models for female students. “It can be hard even to ask for advice but we now have this network where we meet three times a semester to talk about our work and we have people at different levels so someone can say ‘can I see your grant proposal that was successful so I know what to put in mine?’. We can advise on how committee structures at the university work and women know this group exists so they can see it’s not necessary to be intimidated by the establishment. There are several of us who have succeeded with each other’s help.”