22 THIS ARTICLE HAS BEEN PEER-REVIEWED. COMPUTING IN SCIENCE & ENGINEERING

C R O S S T A L K

The European Theoretical Spectroscopy Facility (ETSF) is a distributed knowledge

network that gives researchers access to state-of-the-art computer simulations for

electronic excited states in matter. Focusing on the fundamental knowledge of matter

at the quantum-mechanical level, ETSF seeks to transfer this understanding to the future

design of technologies in multiple areas.

The ETSF: An e-Infrastructure That Bridges Simulations and Experiments

Understanding both the interaction of matter and radiation—including visible or ultraviolet light, x-rays, and electron beams—and our capa-

bility to analyze and predict the materials’ reac-tion enhances our ability to design new materials, improve devices, and understand our environ-ment. Experimental spectroscopies detect the materials’ electronic, magnetic, and vibrational responses, which originate a wide range of result-ing phenomena and processes, including the color of materials, photosynthesis, and light emission.

Theoretical spectroscopy combines quantum-based theories and computer simulation to understand the interaction of radiation and matter. Recent advances in theoretical and computational formal-isms and increases in computational power have led to great progress in theoretical spectroscopy,

letting researchers predict and interpret complex experimental spectra from quantum-mechanical �rst principles. In the past decade, the �eld has clearly made an impact on application-oriented ar-eas, providing important insight into the behavior of complex materials critical to the development of new technologies in many areas, including photo-voltaics (solar cells), LEDs, smart materials, and biological markers for medicine. However, the contacts between experimentalists/industrial re-searchers and theoreticians, although fruitful, have been made on an individual basis. To build a more ef�cient bridge between simulations and experi-ments, an organizational approach is necessary.

The European Theoretical Spectroscopy Facil-ity (ETSF; www.etsf.eu) is a distributed knowledge network, headquartered in Louvain-la-Neuve, Belgium, that provides theoretical expertise and simulations to both public and private sectors. Comprised of more than 200 scientists from 10 core nodes across Europe and associated groups world-wide, the ETSF is an electronic infrastructure (e-infrastructure) that aims to provide a structured answer to the theoretical spectroscopy needs of aca-demic, government, and industrial laboratories.

As Figure 1 shows, the ETSF nodes contrib-ute to the creation of open source software codes, theoretical simulations, research and background knowledge, training, customized support, and collaborations to enhance user studies of complex or nanoscale materials’ electronic properties.

A.Y. Matsuura, Nicola Thrupp, and Xavier Gonze

Université Catholique de Louvain, Belgium

Yann Pouillon

University of the Basque Country, Spain

Gaëlle Bruant

Centre National de la Recherche Scienti�que, France

Giovanni Onida

Università degli Studi di Milano, Italy

1521-9615/12/$31.00 © 2012 IEEE

COPUBLISHED BY THE IEEE CS AND THE AIP

JANUARY/FEBRUARY 2012 23

Simulations and Experiments: Filling the GapThe ETSF emerged from 15 years of collabo-rations among European research groups spe-cializing in the simulation of materials systems. Researchers in both the public sector and private companies were constantly challenging theorists to provide methods and software to explain exper-imental �ndings and material properties. Between 2004 and 2008, a Nanoquanta Network of Excel-lence was started with European Union funding to create a virtual facility that �lled the gap between theory/simulations and experiments by providing answers to these queries and by making it easier for interested experimentalists to �nd theoretical collaborators. The ETSF was established and be-came operational during this time.

To build an effective bridge between simula-tions and experiments, the ETSF was created with a structure analogous to an experimental synchro-tron light-source facility. A synchrotron facility consists of beamlines along which light travels to experimental stations, where experiments are performed to determine the electronic structure of materials and devices. Just as a synchrotron has beamlines where experiments are performed to investigate the interaction of materials with x-ray light, the ETSF has virtual beamlines (shown in Figure 2) that, based on quantum-mechanical the-ories, provide simulations of what happens within

a material when it interacts with light or other per-turbations. The ETSF is a virtual user facility that accepts proposals from users for the allocation of theoretical/simulation support, just as a synchro-tron is a physical user facility that accepts propos-als from experimentalists for the allocation of time at the synchrotron to perform experiments.

The ETSF works in close collaboration with researchers, companies, and governments to de-termine the best course for employing ETSF resources and software to solve scientific/ technological problems. ETSF services are adapted to accommodate the customer’s particular needs.

Time-resolved

Vibrational

Transport

Energy loss

Optics

X-ray

Photo-emission

Figure 2. The ETSF’s virtual beamlines, which are analogous to those in an experimental synchrotron light-source facility. Like a synchrotron facility, the ETSF accepts proposals from users for theoretical/simulation support.

Figure 1. The European Theoretical Spectroscopy Facility eases the use of state-of-the-art theoretical tools for nonspecialists. ETSF includes more than 200 scientists from 10 core nodes across Europe and associated groups worldwide; its goal is to meet the theoretical spectroscopy needs of public and private laboratories.

Fundamentalknowledge

Integratedtools Services

Development of thetheory and codes

Keep the calculationsef�cient

Open access to all researchers,including experimentalists

24 COMPUTING IN SCIENCE & ENGINEERING

This customized user support can range from providing complete theoretical calculations, simu-lations, and analysis of a scienti�c/technological issue for clients who aren’t experts in the �eld, to simply providing training courses on ETSF open source software for the most sophisticated users. In this way, the ETSF brings simulation expertise to experimentalists and other researchers.

User AccessUsers can interact with the ETSF in several ways. The ETSF offers direct access to open source software codes through its website (www.etsf.eu) for researchers familiar with theory and code based on quantum-mechanical �rst principles. Re-searchers who aren’t yet familiar with ETSF codes, but still wish to use them directly, can bene�t from

r� training material available on the codes’ websites;r� regularly organized training sessions; andr� training projects that let users spend a few

weeks at an ETSF node to be trained individu-ally (indirect access to ETSF codes is also pro-vided through collaborative research projects).

These user projects can be as simple as running requested calculations and providing results to users or as complex as giving users a theoretical collaborator who will develop theory and code on demand. Collaborative research projects gen-erally last between two and 12 months.

Training days for experimentalists are also of-fered. Although the ETSF was constructed to be analogous to an experimental synchrotron, language issues between theorists and experi-mentalists create a barrier that the ETSF strives to overcome. Through presentations of the main theories and ETSF beamlines, these training courses help experimentalists understand both the capabilities and limitations of ETSF simulations. At the same time, ETSF scientists learn from the experimentalists about their speci�c needs, and the developments in theory and code that would be most relevant for them today.

Theoretical Spectroscopy TheoriesScienti�cally, the ETSF employs a range of theor-etical methods to study electrons in materials and

their interaction with radiation, particularly density- functional theory and many-body perturbation theory. Over the last decade, the �rst principles theory of electron spectroscopies has been in rapid development and expansion internationally, and the ETSF has contributed strongly to this, both in terms of applications and in terms of the develop-ment of the underlying theory and software.

The Density-Functional Theory (DFT)1 method is an important preliminary step for most of the investigations performed by ETSF. While DFT isn’t suf�ciently accurate for most spectroscopy applications, it’s relatively inexpensive from a computational point of view, and the quantities computed from DFT are usually necessary start-ing points for more computationally demanding and accurate calculations, like those based on many-body perturbation theory (MBPT).2 In the rather complex framework of MBPT, the main techniques—the “GW” method for the one-particle spectral function for photoemission or inverse photoemission, and the Bethe-Salpether method for optical absorption and electron en-ergy loss spectra—have advanced from exotic theories for simple prototype systems to power-ful tools that can be applied to complex molecules, nanostructures, and advanced materials. MBPT delivers accurate quasi-particle electronic struc-ture as well as optical and x-ray spectra. The ETSF’s other main technique—time-dependent density-functional theory (TDDFT)3—has de-veloped suf�ciently to be applied successfully to response functions, optical properties, and time-dependent phenomena. It’s especially well suited for describing time-dependent phenomena in the femtosecond range.

A couple of recent scienti�c breakthroughs in theoretical spectroscopy should be noted. For in-stance, theoretical spectroscopy is now applicable to a wider range of scienti�c domains, which now includes biological systems.4 In addition, combined theoretical and numerical tools now are capable of realistic simulation of complex experiments such as Inelastic X-Ray Scattering (IXS),5 Angle Resolved Photoemission (ARPES),6 and surface spectroscopy.7

SoftwareHighly ef�cient scienti�c software plays a cru-cial role in bridging the gap between theoretical methods and real applications. In the past decade, scientists in the ETSF community have developed many different codes for modeling real materials from �rst principles (see www.etsf.eu/resources/software). Initially, these developments could be conducted independently. Increasing user

Highly ef�cient scienti�c software plays a

crucial role in bridging the gap between

theoretical methods and real applications.

JANUARY/FEBRUARY 2012 25

demands, both in terms of capabilities and size of physical systems, have produced many challenges for theory and software development. The ETSF has explored new ways of calculating interesting physical quantities, while at the same time taking advantage of existing, proven methodologies.

Keeping the calculations ef�cient, even while increasing the size and complexity of the systems, is one issue faced by ETSF code developers. If the system becomes too large, the codes used to cal-culate atomic-scale physical properties decrease in performance. The ETSF thus consistently works on algorithmic development and is currently work-ing with high-performance computing (HPC) centers across Europe to optimize ETSF codes so that they’re compatible with massively parallel ar-chitectures and GPU computing. When working solely at the atomic scale (less than 1 nm), it’s pos-sible to perform a whole series of calculations using only one code. However, to study systems of nano-scopic size (1 to 100 nm), it’s often necessary to use the capabilities of different codes in a synergistic way. Complementary use of different software ap-plications to study different phenomena in the same material—or even to study the same phenomenon in the same material—can be informative as it of-fers a way to validate results and software codes.

An important component of the ETSF is thus the integration of code development in the form of standards and libraries, as each research group in the network provides complementary theoretical capabilities and software tools. Signi�cant progress has been made in four areas in the last �ve years:

r� de�nition of the ETSF coding standards, to ensure a minimum quality level for all of our software;

r� uni�ed access to more than 200 exchange- correlation functionals (the basic building blocks of DFT) by writing a speci�c, robust, and well-documented library, code-named LibXC;

r� handling output data to improve data portabil-ity and increase interoperability between ETSF codes, as well as to solve various archiving is-sues, by designing a �le format and developing its associated library, code-named ETSF_IO; and

r� providing portable access to atomic data �les, by participating in the design of an XML-based �le format and writing a converter between various formats. This converter is currently being rewrit-ten in the form of a library, code-named Libpspio.

The organization of ETSF software activities is exible, regularly adapting to the evolving needs of software developers. The ETSF gives them a

forge and a test farm, ensuring that they don’t need to install and maintain dedicated computers. The forge consists of remotely accessible storage space equipped with version -control systems, Web applications, and backup tools, letting ETSF developers manage their source code safely and exibly.

The ETSF test farm covers various combina-tions of computer architectures, as well as a broad range of running and development environments, which makes it especially useful for the early de-tection of portability issues. To make codes more easily available and user friendly, the ETSF has created live CDs featuring many tutorials, thus let-ting users experiment with the software and their underlying concepts. The ETSF is now migrating toward virtualization, which offers more exibility and reliability. Some ETSF codes have attracted the attention of major Linux representatives, re-sulting in collaborations between the ETSF and these distributors. Currently, the ETSF’s largest project in this area is the creation and maintenance of two Debian metapackages for nanoscale phys-ics (for an example, see http://blends.alioth.debian.org/science/tasks/nanoscale-physics).

These activities are made possible by trained ETSF developers. Locally, the developer teams provide �rst-hand training to the members of the local ETSF node. When people move from one node to another, they take this knowledge with them. Every other year, the ETSF organizes a tutorial in which a limited number of researchers can learn the basics of scienti�c software engi-neering and maintenance from anywhere in the world. Finally, ETSF users can apply to receive personalized training through visits to one or more of its nodes. In the future, the ETSF Devel-opers’ Reference, a document that ETSF volunteers are now writing, will provide free practical details on all fundamental aspects of scienti�c software development in the ETSF context.

From a more philosophical viewpoint, the development, maintenance, and distribution of ETSF software is essentially based on openness and freedom. The ETSF distributes open source

Anyone with the required skills should have the

opportunity to verify any publicly available

scienti�c result and discover how the result has

been obtained.

26 COMPUTING IN SCIENCE & ENGINEERING

software only, typically under the GNU Gen-eral Public License (GPL; www.gnu.org/licenses/gpl.html). Sharing our source code and allowing people to study it, copy it, modify it, and share modi�ed versions is an optimal way to improve the overall quality of research, both in our �eld and for more general purposes. In addition, ETSF utilities and libraries are distributed with even more permissive licenses so that they can be in-cluded in proprietary software.

ETSF developers believe that anyone with the required skills should have the opportunity to verify any publicly available scienti�c result and discover how the result has been obtained. In prac-tice, this philosophy leads to the earlier detection of critical bugs, improved coordination of software developments, and a friendly and informal atmo-sphere among the people involved. We simplify this further by letting local ETSF nodes retain ownership of the software they’ve developed.

Proposal ProcessIn addition to developing theory and providing training and open source software, the ETSF also accepts proposals for user and training proj-ects that could bene�t from theoretical expertise. Connecting the ETSF to experimentalists and other researchers who could bene�t from theo-retical spectroscopy simulations is the goal of the ETSF call for proposals. The CFPs were inten-tionally created to work in an identical manner to an experimental synchrotron light-source fa-cility’s CFPs. By using the same language as an experimental facility, the ETSF becomes easier for its experimentalist users to understand, thus helping to bridge the gap between simulation and experiment. Potential users are asked to submit a two-page proposal, in which they explain their research topic, their problem, and their needs in terms of simulation. Training proposals are eval-uated within two weeks by the ETSF beamline coordinators, and directed to the most appropri-ate ETSF node according to the applicant’s needs. Collaborative research proposals are evaluated

twice per year. Because potential users aren’t familiar with theoretical spectroscopy, they’re asked to discuss their proposal with us before submitting to avoid infeasible proposals.

After the submission deadline, ETSF beamline coordinators evaluate each proposal’s feasibility with regard to the requested simulation work, the foreseen duration, and the required CPU time. An independent panel of evaluators ranks the pro-posals based on scienti�c excellence and the feasi-bility report. For proposals from industrial users, the impact for society is added as a second evalu-ation criterion. The highest-ranked proposals are distributed to the most appropriate scientists by an internal panel, with representatives from each ETSF node. In some cases, proposals require the expertise of several ETSF nodes.

The ETSF provides scientists and CPU time to successful proposals. Most of the user proj-ects result in a joint publication between the us-ers and the ETSF scientists. To date, the ETSF has received more than 200 proposals, completed 80 user projects, and has more than 1,000 ETSF software users.

Applications: ETSF Theories and Software CodesUser projects from academic and industrial users let us work in �elds as varied as chemistry, min-eralogy, nanotechnology, materials science, astro-physics, biophysics, photovoltaics, and optics. The ETSF has worked on materials ranging from nanomaterials to biomolecules to surfaces and bulk materials. Calculating properties of materi-als through simulation lets users understand the material structurally in addition to predicting the material’s electronic properties. Ultimately, this leads to the understanding of certain phenomena in more detail, the selection of materials suited to speci�c purposes, and to the design of new ma-terials with better properties for speci�c applica-tions. We now describe some examples of how ETSF theories and simulations have contributed to such experiments.

Biophysics

The simulation of biomolecules at an atomic or molecular level contributes to our understanding of how these entities interact with their environ-ment and how they function. For instance, the ab-sorption of light by rhodopsin—a protein in the eye—causes this protein to change shape, thereby setting off a cascade of events that subsequently allow us to see. By creating a virtual environment that mimics the protein’s environment and adding

User projects from academic and industrial

users let us work in �elds as varied as

chemistry, mineralogy, nanotechnology,

materials science, astrophysics, biophysics,

photovoltaics, and optics.

JANUARY/FEBRUARY 2012 27

the protein’s structure to this environment, we were able, over nanosecond time periods, to simu-late what happens to the protein and thus learn more about what occurs at an atomic level.8

Geophysics

Another collaboration with experimentalists fo-cused on understanding the electronic properties of the materials that form the earth’s lower man-tle to learn about aspects such as seismic activity. Experimental techniques determined the chemi-cal structure of mantle substances, and we could then simulate the material at extremes of pressure and temperature—something that isn’t easy to do experimentally. The results of these simulations added to our knowledge of the material’s prop-erties and helped the interpretation of further experimental data, leading to an explanation of some of the mantle’s phenomena.9

Materials Science

Applications such as optical data storage (as in re-writable DVDs) are based on a phase change from a crystalline state to an amorphous state in certain telluride alloy materials. The use of �rst-principles calculations10 contributed to our understanding of this phase change, and predicted that the materi-als’ changing composition would improve the data storage devices’ stability.

Nanosciences

Microelectronics rely almost exclusively on today’s silicon-based technology, but recent discoveries have demonstrated the possibility of electronic devices made entirely of nanostructured carbon. Nanostructured materials are known to possess novel electrical, magnetic, optical, and thermal

properties relative to their bulk counterparts. How-ever, the physics behind these interesting prop-erties isn’t well understood. To implement them effectively into marketable products, it’s necessary to gain insight into these materials’ properties.

Experimentally, one of the most important dis-coveries has been the production of carbon in the form of linear atomic chains, thereby realizing 1D nanowire structures. However, the spectral data generated by the production of these carbon nanowires11 was complex. A theoretical investiga-tion, provided through an ETSF user project,12,13 successfully explained the experimental data and determined the chains’ detailed structure, which was subsequently con�rmed by experimental data.14 Moreover, our calculation results suggest the possibility of controlling properties such as nanowire conductance (see Figure 3), as well as magnetic and optical properties. Linear carbon chains could hence acquire an important role in future carbon-based electronics.

ETSF simulations15 have also been bene�cial in understanding related materials such as graphene, which exhibits interesting properties such as high electrical conductivity, high optical transparency, and high mechanical strength. These character-istics make it an ideal candidate for many appli-cations, including integrated circuit components, touchscreens, and liquid crystal displays.

Gas Sensors

Gas sensors can be used to detect small concen-trations of a speci�c chemical compound. The applications here are widespread, ranging from the use in home, medical, and agricultural envi-ronments to monitoring environmental pollut-ants or chemical processes in industrial settings.

Figure 3. Simulated image of carbon nanowires connecting two graphene fragments. The yellow, red, and blue conformations represent carbon atoms with different properties. The system can be designed to behave as (a) an insulator or (b) a conductor.—Figure courtesy of Zeila Zanolli.13

(a)

(b)

28 COMPUTING IN SCIENCE & ENGINEERING

Carbon nanotubes (CNTs) have a large surface area-to-volume ratio, which should make them particularly sensitive to molecular adsorption and excellent candidates for ultra-sensitive gas sensors. However, undoped CNTs don’t bind well with other molecules and thus don’t make good sensors. A possible solution is the “doping” of these CNTs (adding different atoms to the CNT), which might offer a way to control CNT reactivity to and selec-tivity of speci�c chemicals (see Figure 4).

We used computational methods to screen the potential sensing capabilities of CNTs doped with different transition metals.16 With the ad-vancement of computer performance and simu-lation techniques, computational screening has become faster and less expensive than using ex-perimental techniques to screen materials. The study predicted that nickel-doped CNTs could be a promising candidate for carbon monoxide (CO) detection. The next step is to test these nickel-doped CNTs experimentally.

Light-Emitting Diodes

White LEDs are considered one of the most prom-ising light sources, with much less toxicity and bet-ter energy ef�ciency than incandescent or compact uorescent lamps.17 However, the original blue LED device offers an insubstantial color-rendering index and insuf�cient color reproducibility for gen-eral lighting and LCD backlights.

However, theoretical simulations in one of our industrial user projects might help elucidate the mechanism behind the luminescence properties of new red and green phosphors for white LED. The purpose is to predict the structure of phosphors with ideal properties before synthesis and to use a computational screening process to determine the

most promising materials. This process will re-duce the cost of research and development, which has so far been based on a trial-and-error experi-mental screening method.

Thermal Barrier Coatings

Gas turbines are essential for power generation and propulsion. Their ef�ciency is currently limited by the maximum operating temperature that the struc-tural components can sustain, while jet engines are already operating above the melting temperature (in particular during take off and landing). The most promising solution here is to deposit a ceramic thermal barrier coating (TBC) onto turbine com-ponents to insulate the turbine thermally, which in turn allows an increase in the operating tempera-ture. This substance should not only be resistant to heat and act as a good insulator; it should also be “compatible” with other turbine components.

One such material is yttria-stabilized zirconia. Computing different properties through simula-tions can help us understand why this material has proven to be an effective TBC. In addition, this understanding can help engineers design materi-als with improved properties.

In a �rst step, we worked with Siemens AG to determine the atomic and electronic structure for different crystalline phases of zirconia. The re-sults are in good agreement with experiments.18 Calculations also provide an understanding of the observed stabilization of the crystalline phases, which exist at high operating temperatures when yttrium is added. Future calculations will take the high operating temperatures into consideration, and we’ll be able to simulate a range of similar materials and predict the most suitable insulating TBCs for gas turbines.

Figure 4. A carbon nanotube (CNT) gas sensor. “Doping” nanotubes—that is, adding impure atoms—changes the properties of the nanotubes. In this case doping makes them more likely to bind well with other molecules, and thus makes them excellent candidates for ultra-sensitive gas sensors. —Figure courtesy of Angel Rubio.

Target moleculeBackground Active sites

Sensing property

JANUARY/FEBRUARY 2012 29

The ETSF is at a critical juncture: it’s transitioning from groups of people with expertise in theory and materials simulation who work well together,

to a structured virtual consortium that forms partnerships and working relationships at the or-ganizational level. Outreach to organizations and connections to new users, who can both strength-en and widen the bridge between simulations and experiments, is an ETSF priority.

We’re focusing on reaching out to experimental facilities and HPC centers. Over the past decades, light sources have advanced tremendously in reso-lution and power. These experimental facilities have many users who gather tremendous amounts of data, much of which remains unanalyzed and unpublished due to the lack of theory. Many real systems are too complex for current theories to be described accurately. The ETSF is now forming partnerships with synchrotron light sources to advance theories and software that better comple-ment experimental techniques and better describe more realistic complex systems so that they’re more bene�cial to a larger set of users.

The ETSF-synchrotron partnerships will also result in joint CFPs for the simultaneous allo-cation of both experimental and theoretical re-sources to users. Concurrently, the ETSF is in discussions with HPC centers to have Web por-tals on HPC webservers to access ETSF software. These strategic partnerships will take the ETSF closer to its goal of becoming a virtual analogue to a synchrotron, the ultimate bridge between simulations and experiment.

Companies will also play a vital role in the ETSF’s future. The ETSF currently has projects with industry in areas such as LEDs, photovolta-ics, biological markers, optical storage devices, and �ber optics (see Figure 5). We’re making a concentrated effort to increase the number of corporate-driven projects over the coming years.

If ETSF might be useful to you, contact us and visit the ETSF website (www.etsf.eu). We’re always looking to improve our theory and software to meet the needs of new customers and technologies, and we’re always interested in new partnerships.

AcknowledgmentsWe’re grateful to all of the ETSF members who have

contributed to the writing of this article. The ETSF is

funded by the ETSF-I3 project (contract 211956) as an

Electronic Integrated Infrastructure Initiative under the

European Union’s Framework Programme 7, as well

as numerous national funding sources. Yann Pouillon

receives funding from the Spanish Ministry of Science

and Innovation (contract PTA2008-0982-I) and Min-

istry of Education (contract FIS2007-65702-C02-01),

as well as Consolidated Basque University Groups of

the Basque Government (contract IT-319-07), the e-I3

European Theoretical Spectroscopy Facility Project

(contract 211956), and Etortek-inanoGUNE.

References1. R.M. Martin, Electronic Structure: Basic Theory and

Practical Methods, Cambridge Univ. Press, 2004.

2. G. Onida, L. Reining, and A. Rubio, “Electronic

Excitations: Density-Functional Versus Manybody

Greens-Functions Approaches,” Rev. Modern Physics,

vol. 74, no. 2, 2002, pp. 601–659.

3. M.A.L. Marques et al., eds., Time Dependent Density

Functional Theory (TDDFT), Lecture Notes in Physics,

vol. 706, Springer, 2006.

4. D. Varsano et al., “A TDDFT Study of the Excited

States of DNA Bases and Their Assemblies,” J. Physical

Chemistry B, vol. 110, no. 14, 2006, pp. 7129–7138.

5. H.-C. Weissker et al., “Signatures of Short-Range

Many-Body Effects in the Dielectric Function of

Silicon for Finite Momentum Transfer,” Physical Rev.

Letters, vol. 97, no. 23, 2006, doi:10.1103/PhysRevLett.

97.237602.

6. F. Bruneval et al., “Exchange and Correlation Effects in

Electronic Excitations of Cu2O,” Physical Rev. Letters,

vol. 97, no. 26, 2006, doi:10.1103/PhysRevLett.97.267601.

Figure 5. Example spectroscopy applications. Spectroscopy can be used to examine numerous materials in many �elds, including (a) nanotechnology, (b) biomolecules, (c) LEDs, and (d) photovoltaics.

(a) (b) (c) (d)

30 COMPUTING IN SCIENCE & ENGINEERING

7. M. Palummo et al., “Re�ectance Anisotropy Spectra of

the Diamond (100)-(2x1) Surface: Evidence of Strongly

Bound Surface State Excitons,” Physical Rev. Letters, vol. 94,

no. 8, 2005, doi:10.1103/PhysRevLett.94.087404.

8. J.S. Frähmcke et al., “The Protonation State of Glu181

in Rhodopsin Revisited: Interpretation of Experimental

Data on the Basis of QM/MM Calculations,” J. Physical

Chemistry B, vol. 114, no. 34, 2010, pp. 11338–11352.

9. R. Caracas and C.E. Cohen, “Ferrous Iron in Post-

Perovskite from First-Principles Calculations,” Physics

of the Earth and Planetary Interiors, vol. 168, nos. 3–4,

2008, pp. 147–152.

10. W. Welnic et al., “Origin of the Optical Contrast in

Phase-Change Materials,” Physical Rev. Letters, vol. 98,

no. 23, 2007, doi:10.1103/PhysRevLett.98.236403.

11. L. Ravagnan et al., “Cluster-Beam Deposition and

in situ Characterization of Carbyne-Rich Carbon

Films,” Physical Rev. Letters, vol. 89, no. 28, 2002,

doi:10.1103/PhysRevLett.89.285506.

12. L. Ravagnan et al., “Effect of Axial Torsion on SP Carbon

Atomic Wires,” Physical Rev. Letters, vol. 102, no. 24,

2009, doi:10.1103/PhysRevLett.102.245502.

13. Z. Zanolli, G. Onida, and J.C. Charlier, “Quantum

Spin Transport in Carbon Chains,” ACS Nano, vol. 4,

no. 9, 2010, pp. 5174–5180.

14. C. Jin et al., “Deriving Carbon Atomic Chains from

Graphene,” Physical Rev. Letters, vol. 102, no. 20,

2009, doi:10.1103/PhysRevLett.102.205501.

15. A. Rubio, “Hybridized Graphene: Nanoscale Patch-

works,” Nature Materials, vol. 9, 2010, pp. 379–380.

16. J.M. García-Lastra et al., “Modeling Nanoscale Gas Sen-

sors under Realistic Conditions: Computational Screen-

ing of Metal-Doped Carbon Nanotubes,” Physical Rev. B,

vol. 81, no. 24, 2010, doi:10.1103/PhysRevB.81.245429.

17. M. Mikami et al., “New Phosphors for White

LEDs: Material Design Concepts,” IOP Conf. Series:

Materials Science and Eng., vol. 1, no. 1, 2009,

doi:10.1088/1757-8981/1/1/012002.

18. H. Jiang et al., “Electronic Band Structure of Zirconia

and Hafnia Polymorphs from the GW Perspective,”

Physical Rev. B, vol. 81, no. 8, 2010, doi:10.1103/

PhysRevB.81.085119.

A.Y. Matsuura is chief executive of the European Theo-

retical Spectroscopy Facility through the Catholic Uni-

versity of Louvain, Belgium, and an adjunct professor

in the Department of Physics at Boston University.

Matsuura has a PhD in physics from Stanford Univer-

sity. Contact her at anne.matsuura@uclouvain.be.

Nicola Thrupp is an administrator at the Université

Catholique de Louvain, Belgium, and a research

network administrator and project manager at the

European Theoretical Spectroscopy Facility. Thrupp

has an MS in molecular biology and computational

chemistry from the University of Stockholm. Contact

her at nicola.thrupp@uclouvain.be.

Xavier Gonze is president of the European Theo-

retical Spectroscopy Facility and a professor in the

Nanoscopic Physics Department at the Université

Catholique de Louvain, Belgium. He also initiated

and coordinates the Abinit software project. His

research interests include the study of dynamical,

thermodynamical, dielectric, and electronic prop-

erties of the crystalline state with applications to

high-technology materials thanks to �rst-principles

simulations. Gonze has a PhD in applied sciences

from the Université Catholique de Louvain. He is a

fellow of the American Physical Society. Contact him

at xavier.gonze@uclouvain.be.

Yann Pouillon is on a three-year research engineer

contract at University of the Basque Country, San

Sebastían, Spain, and coordinates the software activi-

ties of the European Theoretical Spectroscopy Facility.

His research interests include the atomic-scale study of

water and biomolecules, as well as the fundamental

description of electron–electron interactions at the na-

noscale. Pouillon has a PhD in materials physics from

the Louis Pasteur University of Strasbourg, France.

Contact him at yann.pouillon@ehu.es.

Gaëlle Bruant is a research engineer at the Centre

National de la Recherche Scienti�que (CNRS) in

France, where she’s responsible for the European

Theoretical Spectroscopy Facility call for proposals

at the ETSF Center for Users and Technology. Bruant

has an MS in European professions from the Paris

Institute of Political Studies and an MS in European

projects from the University of Cergy-Pontoise,

France. She’s a member of the French European Proj-

ects Managers Club. Contact her at gaelle.bruant@

polytechnique.edu.

Giovanni Onida is a professor at the Università degli

Studi di Milano’s Solid State Physics Theory Group and

has coordinated the European Theoretical Spectros-

copy Facility’s Milan node since 2004. His research

interests focus on ab initio studies of structural and

electronic properties of low-dimensional systems and

nanostructured materials. Onida has a PhD in physics

from the University of Milan. Contact him at giovanni.

onida@unimi.it.

Selected articles and columns from IEEE Computer

Society publications are also available for free at

http://ComputingNow.computer.org.

Focus on

Your Job Search

IEEE Computer Society Jobs� OLSWZ� `V\� LHZPS`� ÄUK�

H� UL^� QVI� PU� 0;�� ZVM[^HYL� KL]LSVWTLU[�� JVTW\[LY� LU-

NPULLYPUN�� YLZLHYJO�� WYVNYHTTPUN�� HYJOP[LJ[\YL�� JSV\K�

JVTW\[PUN�� JVUZ\S[PUN�� KH[HIHZLZ�� HUK� THU`� V[OLY�

JVTW\[LY�YLSH[LK�HYLHZ�

New feature:�-PUK�QVIZ�YLJVTTLUKPUN�VY�YLX\PYPUN�[OL�

0,,,�*:�*:+(�VY�*:+7�JLY[PÄJH[PVUZ�

Visit www.computer.org/jobs to search technical job

openings, plus internships, from employers worldwide.

http://www.computer.org/jobs

;OL�0,,,�*VTW\[LY�:VJPL[`�PZ�H�WHY[ULY�PU�[OL�(07�*HYLLY�5L[^VYR��H�JVSSLJ[PVU�VM�VUSPUL�QVI�ZP[LZ�MVY�ZJPLU[PZ[Z��LUNPULLYZ��HUK�JVT-

W\[PUN�WYVMLZZPVUHSZ��6[OLY�WHY[ULYZ�PUJS\KL�7O`ZPJZ�;VKH �̀�[OL�(TLYPJHU�(ZZVJPH[PVU�VM�7O`ZPJPZ[Z�PU�4LKPJPUL��((74���(TLYPJHU�

(ZZVJPH[PVU�VM�7O`ZPJZ�;LHJOLYZ��((7;���(TLYPJHU�7O`ZPJHS�:VJPL[`��(7:���(=:�:JPLUJL�HUK�;LJOUVSVN �̀�HUK�[OL�:VJPL[`�VM�7O`ZPJZ�

:[\KLU[Z��:7:��HUK�:PNTH�7P�:PNTH�

(FU�UIF�CVJMEJOH�CMPDLT�ZPV�OFFE��

5BLF�ZPVS�DBSFFS�UP�UIF�OFYU�MFWFM�JO�TPGUXBSF�EFWFMPQNFOU �TZTUFNT�

EFTJHO �BOE�FOHJOFFSJOH�XJUI�

"SUJDMF�DPMMFDUJPOT�GSPN�UIF�*&&&�$PNQVUFS�4PDJFUZ

.BUFSJBMT�GSPN�)BSWBSE�#VTJOFTT�4DIPPM�1VCMJTIJOH

$PNQVUFS�EJTDPVOUT

0OMJOF�DPVSTFT�BOE�DFSUJ¾DBUJPOT

0VS�FYQFSUT��:PVS�GVUVSF�

°

°

°

°

XXX�DPNQVUFS�PSH�CVJMEZPVSDBSFFS

Is your careerfoundation

solid?