A spectrometer for the Extremely Large Telescope
Employeneur Annemieke Janssen of TMC Physics is closely involved in a project to build the world’s largest optical telescope at the European Southern Observatory (ESO) in Chile. The ‘Extremely Large Telescope’ (ELT) has a primary mirror diameter of no less than 39 metres. A project of this scale presents some interesting challenges.
Annemieke is a Systems Engineer at the NOVA (Nederlandse Onderzoeksschool voor Astronomie) Optical Infrared Group – which is facilitated by Astron, a radio observatory in Dwingeloo, where she is stationed. As an expert in the field of astronomical instrumentation, Annemieke is helping design one of the instruments for the telescope.
‘The design and construction of the telescope and instruments is a very intensive process and takes more than 10 years,’ Annemieke says. ‘Major international consortia are involved in the instrumentation projects, and I regularly travel to Paris or Munich. Once every two years there is a conference on astronomical instrumentation. A select group with the same background attend this conference, so I meet a lot of acquaintances.’
The projects, on which the NOVA Optical Infrared Group works, are temporary and therefore only a limited number of people are hired on a fixed contract. ‘For me it is ideal to work as an Employeneur on this project through TMC. Because project money is used and it is a temporary assignment, the chance of a permanent contract is slight.’
The experimental phase
A telescope collects light and focuses it in a focal plane. To make pictures or spectra, an instrument must be attached to the telescope. This is often a camera or a spectrometer. Annemieke explains that MOSAIC could become one of the ELT instruments. MOSAIC is a combination of two identical spectrometers that work in visible light (450–900 nm) and two identical spectrometers that work in the infrared range (900–1800 nm). ‘At the moment this project is still in an experimental phase, in which I’m working on the spectrometer for visible light. This year it will be decided whether the construction of the instrument is feasible and whether it meets the wishes of the astronomers. If the concept is approved, we will make a detailed design and build the instrument.’
The challenge of seeing the big picture
People often wonder why the process of designing and manufacturing takes so long. There are already many spectrometers in this wavelength range for other telescopes, such as the ‘Very Large Telescope’ (VLT). It may be a possibility to place an instrument from the VLT on the ELT but Annemieke explains why this is impractical: ‘The first problem is the size of the focal plane, where the telescope and the instrument are connected. Suppose you want to observe a part of the sky the size of the moon (0.5° x 0.5°). The focal plane on the VLT is then slightly more than 1 m x 1 m, while on the ELT it is almost 6 m x 6 m. The difference is a factor of five in length and 25 in area. Either we accept that the visible part of the sky has become 25 times smaller or the instrument has to be bigger. Unfortunately, an instrument cannot simply become five times larger, because the size of detectors and optics, such as lenses, mirrors and prisms, is limited.’
In MOSAIC, the focal plane becomes 1.5 m long. It is not feasible to make spectra of the entire surface. Instead, using glass fibres allows the positions in the focal plane to be selected, where stars and galaxies are located. The glass fibres then ‘transport’ the light from the focal plane to the spectrograph. ‘This helps,’ Annemieke says, ‘but the instrument still tends to get bigger. Not only does the focal plane increase by a factor of five compared to the VLT, but the projection of a point source in the focal plane also increases by more than a factor of five. For example, a star is imaged at 3 mm x 3 mm, which is too large to fit into one glass fibre. That is why several glass fibres are needed to transport the light from one star to the spectrograph.’
Building bigger optics
Building a large spectrograph with large optics may be a solution, but the production takes time. Annemieke: ‘The production of a lens with a diameter of 40 cm alone can take more than a year. First the glass must be melted in an oven and then it takes more than six weeks before the glass has cooled down. The glass has to cool very slowly to remain homogeneous. Then the glass is inspected, polished and coatings can be applied.’ However, most of the time is spent in a detailed design before the lens can be manufactured. Annemieke continues: ‘A design arises after years of interaction between engineers, astronomers and technicians, who try to develop the ideal instrument. Such an instrument, which is also feasible and affordable, can be used for interesting research. I hope to travel to Chile sometime to see the telescope in real life.’