The European Theoretical Spectroscopy Facility is a collaborative research network focusing on the theory and simulation of various types of electron-based spectroscopies, applied to advanced materials, nanostructures, molecules, etc. The ETSF is based on a multitude of individual collaborations, some of which go back to the 1980s. For the early history of the network I focus here on the European collaboration that stemmed from the invention of the GW approximation in many-body perturbation theory by Lars Hedin (Lund University) in 1965, and which led, together with additional participants, to the EPSI, Nanophase and Nanoquanta networks (see below) and the creation of the ETSF itself.
The GW approximation found immediate application to the homogeneous electron gas, but ab initio applications to more realistic systems were beyond the scope of computers in the 1960s. The field started to heat up in the 1980s, when papers from Berkeley and Bell Labs showed that the biggest supercomputers such as the Cray-X/MP (with computing power similar to a modern smartphone, incidentally) were finally capable of applying GW to realistic materials. The field developed by combining ideas and techniques from the “band structure” community (then starting to recognise the relevance of ab initio density-functional theory to some – but not all – of their calculations), and the “optical properties” community, who had for some time been performing calculations based on linear-response theory using various simplified methods and empirical parameters.
Lars Hedin and his colleagues Carl-Olof Almbladh and Ulf von Barth from Lund, together with Rodolfo Del Sole from Rome, were prominent in helping to organise workshops, often based at CECAM (then located at Orsay, near Paris), from the late 1980s onwards. Lucia Reining was by then based at CECAM, having completed her PhD in Rome; she was an active collaborator with numerous other workshop participants including Friedhelm Bechstedt (Jena) and Rex Godby (Cambridge, later York), and was soon to move to the Ecole Polytechnique, Palaiseau, France. To cement these collaborations, Rodolfo Del Sole coordinated a successful application to the European Commission’s fledgling science programme for a Human Capital and Mobility network "EPSI" (for “Electronic Properties of Semiconductors and Insulators”), involving those mentioned plus Olle Gunnarsson (MPI Stuttgart) and four other nodes, which helped to oil the wheels of more intensive collaboration, as well as continued annual workshops, from 1992-1995.
A more substantial EU research training network grant, Nanophase (for “Nanoscale photon absorption and spectroscopy with electrons”) which ran from 2000-2004, saw the groups of Matthias Scheffler (FHI Berlin) and of Angel Rubio and Pedro Echenique (San Sebastián) join the consortium. The Nanophase network published around 80 papers per year, of which about 20 were collaborations involving multiple Nanophase nodes.
In May 2004 the Nanophase network held its first Young Researchers' Meeting at the École Polytechnique, exclusively for network scientists without permanent positions. The experiment was very successful and continues annually until this day.
Building on Nanophase’s success, the Nanoquanta EU Network of Excellence (2004-2008) saw the creation of an ETSF, with a mission to reach out to experimental and industrial collaborators, as well as continuing the network’s strong tradition in research and training. New core groups joining the Nanoquanta network were led by Hardy Gross (Freie Universität, Berlin) and Xavier Gonze (Louvain-la-Neuve), in part reflecting the increasing relevance to spectroscopy of density-functional theory, and in particular time-dependent DFT, alongside methods such as GW and the Bethe-Salpeter equation. Nanophase postdoc Giovanni Onida now had a permanent position in Milan, forming another of the ten Nanoquanta core groups. The Nanoquanta network continued to grow to around 120 scientists, and produced around 160 papers per year, of which about 40 were Nanoquanta collaborations.
The ETSF issued its first call for proposals for collaborative projects with external experimentalists and industrial researchers in spring 2007. This was to be the first of a biannual series of calls, leading to over 100 research and training projects with external users, with the ETSF consciously mirroring the operation of experimental synchrotron facilities, including being organised around seven “theoretical beamlines" for the purpose of user proposals. (These beamlines were Optics, Electron Loss Spectroscopy, Photoemission, Quantum Transport, Time-Resolved, Vibrational, X-Ray.)
The network’s 2008-11 grant from the EU was specifically focused around the ETSF as a scientific “e-infrastructure”, funding continued access by external users to the ETSF for research and training, in parallel with maintaining the network’s development and application of its own research and software. Now around 30% of the ETSF’s publications involved multiple ETSF nodes.
By 2011, the network had trained dozens of scientists who were building their own research groups across Europe, and wished to remain within the ETSF’s collaborative network, along with their research students and postdocs – a total of well over 250 researchers. Other collaborators, including some in the USA, also wished to be part of the network. Evidently, the original structure of a small number of “core nodes” was no longer appropriate, and the ETSF adopted its present structure, consisting of independent permanent scientists (“Research Team Leaders”, 71 at the last count), along with relevant members of their research groups.
The network’s commitment to fundamental science, training of young researchers, novel applications of theory, innovative software, and openness to collaboration with external scientists continues unabated, and the deep connections that are generated during 30 years of cooperation between numerous European groups should ensure the ETSF’s ongoing role in catalysing significant research in this thriving field.
York, 23 April 2019