A vision
When Professor Stobbe started as associate professor at DTU Electro in 2018, he had a vision for a, to him, new field of research. He wanted to make very carefully designed devices and nanostructures, and push the limits of how precisely they can be made and how much they can be optimised. DTU's state-of-the-art clean room and the significant expertise in the field of nanophotonics here at the department attracted him.
The work Stobbe's group has done on photonic cavities over the past few years, leading to several significant publications, is now well ahead of the international competition.
Last year Stobbe and his team got an article on self-building nanocavities published in Nature. He explains why this piece of work is special to him: “We start with an idea that doesn't exist anywhere - no one suggested doing this. And then we take that idea really far, spending many years of intense work, and then putting it into an article that I am incredibly proud of”, Professor Stobbe says, adding: "It's the best work I've ever done."
The real world
The above has been one of two tracks in Professor Stobbe's vision: basically understanding where the limit is for how small things can be built - especially in regard to photonic cavities, because then you can turn up the interaction between light and matter and hopefully down the line build new information and quantum technology. But also to understand what the properties are in these cavities and how light behaves on such an extreme scale. You could call this the fundamental research track.
But he is also very interested in applied research and commercialization. He has co-founded three start-ups, and explains that he enjoys the filter the real world offers.
Professor Stobbe is hesitant to call his work "research" or "innovation", what he focuses on is impact.
Stobbe loves fundamental research, but he thinks it’s important to consider what is worth spending time on. “For example, if I find a grain of dust, should I spend my whole life trying to understand what it consists of? - because it would be really difficult. Maybe I should try to map out with atomic resolution what this grain of dust consists of. I could spend my whole life on this. And maybe there is something exciting - maybe there isn't. But it doesn't seem very likely that such a research vision would lead me to something absolutely fantastic. And if everyone operates with a research vision that is so blue sky that it doesn’t aim at anything at all, then I am not sure we will get much out of it," he explains.
Therefore, he believes it’s important to consider the challenges of the real world and look at what’s actually missing in the world, because it can help define some questions - "science is about asking the right questions", he says.
Stobbe highlights his collaboration with DTU Construct, who do numerical design that is about optimising geometries and systems, in situations where simple calculations and conventional algorithms fall short. There is a tremendously fruitful interplay between that type of optimization, and Stobbe's vision of pushing the limits of how precise and how small things can be made, and how accurately they can be controlled. Because you can design all sorts of nice things, but if you aren’t able to build them, then they aren’t of much use.
Kill your darlings
Stobbe finds it crucial that basic research takes aim, and believes it has great value to be brutal about one's ideas if they turn out to be bad - and shutting them down.
He has shut down several ideas over the years, because he deemed they weren’t leading anywhere. Last year, his group published an article about topological insulators in Nature Photonics. Stobbe explains that they could have done more work within this field, because now they were in, and theoretically it’s an exciting subject. The issue? Their results showed that it will most likely never work in practice.
Professor Stobbe spent years on this, but with this result, the conclusion was crystal clear to him: Shut it down. It’s not that it wasn’t exciting, but he explains that if he sees a major roadblock somewhere in the horizon, then that’s the end of it. He won’t ask his team to explore something that he himself doesn’t believe in.
He thinks it’s important to always cut down – cutting away everything that doesn’t look promising, and to always take the time to stop and ask himself: “Is this truly the most important or the most ambitious thing we can do right now?” or if he needs to change course.
Stobbe states that in research it’s easy to lull ourselves into a false narrative that something is very important – “but perhaps it’s not important at all”, Stobbe says.
It’s not without challenges to be so willing to pursue the questions, and thus change course in one's research. Obviously, it would be easier to stay in one field. Professor Stobbe recognizes that "starting over" in a new area requires a lot of work.
"But that's how it is with ideals. You always pay a price", he says.
Open source or getting paid? It’s not that simple
There are also limits to research. Some things can only be taken to a certain level, and then it doesn't really make sense to continue within a university environment. But it might still be something that can be of great value to the world, and then you might ask yourself whether you shouldn't just give it all away - "yes, you can give it all away", says Professor Stobbe, but he adds that open source is not always in the world’s best interest.
Stobbe explains that he knows many examples of, for example, software projects that are made open source, and 5 years later they no longer work because they aren’t maintained. Professor Stobbe explains that in many cases the best way to share your research with the world is to turn it into a product and sell it to people. If you make it a product, you take responsibility for it being maintained - but of course it costs some money. "If you're not making them yourself, you're using someone else's," Stobbe says. Someone needs to get paid for maintenance. And of course this depends on what type of research you're doing, but Professor Stobbe thinks that in some instances the road to commercialization is so natural – even idealistic. "I don't start businesses to make money - but they have to make money; otherwise the idea will die”.
Elaborating on his perspective on innovation, Stobbe says that you shouldn’t run a business with the goal of making money, in the same way you shouldn’t do research with the goal of getting an article published in some important journal. You should do it because you have important questions which need to be answered. And if they are indeed important, and you are able to answer them, then publishing an article or selling a product will be a side effect – but it’s just a nice bonus. To Professor Stobbe, creativity is front and centre, “it’s about creating something fun, new, exciting, and beautiful”.
New questions
Stobbe goes where the research takes him, and he’s not afraid to change course.
In his group's work with spectrometers, they are constantly identifying basic research questions. In the nanoelectromechanical systems they build for their spectrometers, they do a mechanical deformation of the chip, meaning that the light is sent through a waveguide, and by mechanically moving these waveguides around, you can control how the light behaves. But it turns out that what limits the resolution in spectrometers is actually that when these things get very close, surface forces cause them to snap together. If you take two things that are extremely clean and flat and put them together, you can't get them apart again because the surface forces pull them together - also called stiction.
But a few years ago, it occurred to Stobbe that what was a pestilence for their work on spectrometers might be a solution elsewhere. Because this stiction was so reproducible, it could be used to create novel methods for nanofabrication of very small cavities.
Because there is a limit to how small optical cavities we can construct ourselves, Stobbe and his team found that they could make them build themselves by bringing two parts together and because of the surface forces that he observed in the group's work with spectrometers, the structure assembles itself. The revelation which was later published in Nature.
A sustainable future
Stobbe explains that there are many connections across his group's research activities, and that most are actually mutations of previous ideas, so everything flows together. But there is one point on which Stobbe calls himself ultra-conservative, and that is materials.
His group exclusively makes silicon structures. In addition to the fact that silicon is the most advanced material for nanotechnology, there is a huge amount of silicon on earth, it’s not associated with problematic mining, and it’s not toxic. Stobbe explains that it’s also somewhat of an ulterior motive with his group's research to only use silicon, oxygen and aluminium, as far as possible, all of which are available, common, and non-toxic materials. "This doesn’t mean that we’re creating sustainable solutions here and now, but it wouldn’t make sense to me to start research projects that would actually lead to a worse world", says Stobbe.
Moving into new fields such as computing architectures Stobbe brings awareness to all the new, extremely energy-intensive AI models. He imagines his group’s fabrication platform would be able to reduce this obscene amount of energy use, hoping that his group's applications, companies, patents, and technologies can help create more sustainable solutions for future societies.