IBM scientist Kirsten Moselund Awarded Prestigious European Grant

Back in 2007 the European Research Council (ERC) emerged as the first pan-European organization funding frontier research through individual grants. 

The multi-million Euro, multi-year grants are highly coveted by scientists and competition is significant. In fact, since it started, 50,000 ERC grant applications have been submitted, yet only 5,000 projects received funding.

This didn’t deter IBM scientists in Zurich from participating and the Swiss lab currently has nine scientists with ERC grants, including three which have just been announced in the past several weeks including Kirsten Moselund, Leo Gross and Abu Sebastian.

Moselund, who is originally from Denmark,
joined IBM Research in 2008.
In December, Moselund, received her grant to fund research in developing nanowire lasers for highly efficient silicon computer chips in a project called PLASMIC. The result of her research could have an impact on the development of low power, compact light emitters, which may find applications spanning the Internet of Things to high-performance computing.

I recently had a chance to catch up with her to ask about her grant research.

1. Can you explain what the focus of your research is under the ERC grant?

Kirsten Moselund (KM): The focus is to explore the ultimate scalability of III-V light sources on silicon by using metallic cavities. Conventional lasers use a photonic cavity usually some sort of Fabry-Perot or another resonant cavity design such as a micro-disc. The reflectance and output can be tuned to optimum performance and the widespread use of semiconductor lasers, for a great variety of applications, is a testament to the great success of this technology. 

However, when trying to scale down these cavities beyond the diffraction limit, i.e. in principle, to less than a few wavelengths in dimensions, the photonic cavities become lossy, and eventually the optical mode is no longer contained. Plasmonic or more simply metallic cavities provide a means for confining light to sub-wavelength dimensions to address this, but the price to pay is high optical loss in the metal, which must be overcome to achieve lasing.

Based on the ERC grant I plan to use the Template Assisted Selective Epitaxy (TASE) process developed within IBM Research for the integration of active III-V photonic NW material, which will form the gain material for the nanolasers. The versatility of this approach gives us an advantage in terms of freedom in device design as opposed to conventional III-V integration methods. Access to the Binnig and Rohrer Nanotechnology Center at IBM and the most advanced fabrication facilities is also important, as very tight process control is required to reduce roughness at the nanoscale.

2. What are some of the major milestones you hope to achieve?

KM: On a materials level, TASE was developed as a tool for electronic devices, and we know from experiments that the III-V material has excellent electrical properties, such as high mobility. The first milestone is to assure that the material quality in terms optical properties is also sufficient to provide the optical gain required. If it's not then we need to optimize the technology to achieve this at the device dimensions required.

The first device milestone would be to achieve lasing from a cavity which is sub-wavelength in at least one dimension, based on optical pumping and likely at low temperatures. From there, we still have a long way towards the ultimate goal of electrically actuated lasing in a sub-wavelength cavity at room temperature directly coupled to a silicon waveguide. So far, this is something which has not been achieved in this field.

III-V lasers on silicon is a well established technology, these are high performance devices with cavities in the 100s of µm range, which are usually heterogeneously integrated. 

The field of monolithically integrated nanolasers is actually not that developed, and I believe there is much room for innovation. 

Moselund, left, with her colleague Antje Rey, participated
in the Swiss Night of Research in 2009 to explain
and demonstrate her research to the general public.
The application space is also very different, we will certainly not achieve the same in terms of performance. Ultimately, the goal of being able to integrate thousands of tiny light emitters densely on chip will hopefully open up new avenues for exploitation.

3. Why is this important for Europe?

KM: Nanolasers are certainly interesting for microelectronics, for on-chip optical signaling etc. and this is currently the target application. However, they also have applications beyond this in sensors, quantum computing and Internet of Things. These are fields which are predicted to become increasingly important in the near future, and where Europe has a strong tradition. 

Also, unlike the microelectronics industry which is extremely consolidated, Europe may support many start-ups or SMEs in these domains to help drive future economic growth.

4. Any tips or advice for other scientists who are applying for ERC grants?

KM: For everyone the reasons for applying for an ERC grant are different, but regardless it's a lot of work, so you should be truly motivated. 

For me it was a desire to change my research topic and work on something new, and unlike most other funding mechanisms, the ERC Starting Grant actually makes this possible. 

On one hand this means you do not have any baggage, in terms of convincing the committee that you will be independent and that this is not just your supervisors/managers idea.  On the other hand, I believe it is harder in terms of work you have to invest to make a convincing proposal in a field which you are less familiar with.

Two pieces of practical advice I would give are: 
  1. Involve others – in evaluating your ideas, reading the proposal, or stylistic input. I received a lot of helpful input for which I am extremely grateful from managers, fellow group members, in particular my office mate Heinz Schmid, but also people outside of my immediate field like Govind Kaigala, Emmanuel Lörtscher, Catherine Trachsel, Elad Koren and my PhD supervisor. Actually the unfamiliarity with the topic allows them to focus on other issues, and made me rethink my focus and improve the proposal. Also Euresearch association provides great support in terms of preparation courses and proofreading, and most of these offers are free of charge. Furthermore, the entire process is a bit like running a marathon so having involved others make you less likely to give up at times when your proposal seems completely hopeless.
  2. Secondly, the ERC makes it decision based on two parts a) your personal profile and b) the project. You cannot change what you have been doing the past ten years or more, but you can change the way you write a proposal to suit a specific committee, and depending on how you formulate the topic you can usually fit it in a few different panels. So, I would always choose the panel which is most likely to look favorable at your personal profile and achievements based on the composition and background of its members. For example in my case, with an electrical engineering background, I chose an engineering panel which is usually more application oriented rather than a physics-based panel.

Stay tuned for Q&A articles with the other two IBM ERC grant winners in the coming weeks.

Labels: , , , ,