Presentation of Astro @ Paris-Saclay

• Frédéric GALLIANO (DAp, CEA Paris-Saclay, France)

I will introduce this annual day by briefly reviewing what is the Astrophysical Axis of the Graduate School of physics. I will present our actions and the opportunities we offer.

Optical Systematics in CMB Spectral Distortion Instruments

• Morgane LOQUET LE GALL (IAS)

While CMB temperature anisotropies have been characterised with precision, the CMB frequency spectrum has not been remeasured since the FIRAS instrument aboard COBE in the early 1990s. FIRAS showed that the spectrum is extremely close to a blackbody, with deviations from a perfect Planck spectrum below ΔI/I ∼ 10⁻⁵. Such small departures, known as spectral distortions, encode the thermal and energetic history of the early Universe, including epochs prior to the release of the CMB itself. Spectral distortions are exceptionally faint signals: detecting them requires an exquisite understanding of the instrument, and anticipating from the earliest design stages how instrumental systematics will limit the final measurement. FOSSIL (FTS fOr CMB Spectral diStortIon expLoration) and BISOU (Balloon Interferometer for Spectral Observations of the primordial Universe) are two projects targeting their detection. To assess the impact of systematic effects on the science measurement, we present an instrument simulator for FOSSIL and BISOU. This model forecasts the error induced on the science measurement by several instrumental systematic effects. This talk focuses on those linked to the optics. A laboratory breadboard of the instrument has been developed at the Institut d'Astrophysique Spatiale (IAS). It will allows us to characterise these systematics experimentally and to compare the simulator's predictions with measurements, a key step towards validating the model ahead of future missions. This presentation outlines the optical modeling process of the BISOU instrument. I will begin with an overview of the optical concept of its Fourier Transform Spectrometer (FTS), followed by the first overall configuration based on geometric optics. The design was refined and optimized through iterations using Gaussian beam propagation, enabling more realistic modeling of the instrument’s optical performance. The outcome is an optical configuration that incorporates key constraints such as size and thermo-mechanical limitations. I will present the current concept and the simulation results concerning the co and cross polarisation beam performance, including the ellipticity, together with the remaining steps required to finalize the design and prepare for the associated breadboard implementation.

Dialogue between worlds

• Xavier BERTOU (IJCLab)
• Maria LOVINO (IN2P3/IJCLab)

We will present an interdisciplinary project that bridges astroparticle physics and fashion design. Co-developed with Argentine fashion creator Panni Margot, “Dialogue Between Worlds” aims to explore new, impactful ways of communicating science and engaging society. The project establishes a partnership between fundamental astroparticle physics and creative design, transforming complex cosmic concepts, from quantum origins to dark matter phenomena, into tangible artworks. More than a stylistic experiment, the collection provides an innovative framework for scientific communication. Each garment is conceived as a technically informed medium that translates abstract data, physical principles, and experimental methodologies into meaningful and accessible learning experiences. By engaging audiences who may not typically interact with scientific research, the project opens new pathways between specialized knowledge and broader public understanding.

What does C₂H₂ trace in planet-forming regions of disks?

• Pacôme ESTEVE (Institut d'Astrophysique Spatiale)

Despite the tremendous number of exoplanets detected, planet-forming scenarios are not yet able to fully explain the large diversity of planets. They form and obtain their composition from gas and dust in protoplanetary disks. The elemental composition of disks is known to vary spatially due to physical (ex: pebble drift) and chemical (chemical reactions, sublimation) processes. Thus, the elemental composition of planetary atmospheres is expected to give valuable insight about their formation, in particular the C/O ratio. However, the chemical composition of planet-forming regions of disks is still poorly constrained, making it difficult to link exoplanetary atmospheres and planet formation history. In this context, the James Webb Space Telescope opens an unprecedented observation window on the inner disks (<10 au), by detecting plenty of new molecules, especially hydrocarbons. In particular, the JWST spectra show a large dispersion of H₂O and C₂H₂ emission in TTauri disks (Grant et al. 2025), potentially highlighting chemical processes at stake. We now need thermo-chemical models to understand these observations and quantify fundamental parameters of disks. I will present an overview of my work based on the thermo-chemical model Dust And LInes (DALI) to investigate the relationship between the observed variation of H₂O and C₂H₂ emission and the C/O ratio. After mentioning several modelling improvements, especially a refined Carbon chemistry, I will show that the C₂H₂/H₂O line flux ratio is very sensitive to the C/O ratio, but also other parameters such as the elemental abundances O/H and C/H, or the dust size distribution. These results could put strong constraints on our understanding of the structure and evolution of protoplanetary disks.

Zoom-in simulations from the kpc scale to star forming cores

• Georges ABBOUDEH (AIM - CEA Saclay)

Zoom-in simulations from the kpc scale to star forming cores - Abstract: :: Star formation plays a crucial role in shaping its environment and drives the evolution of planets, the interstellar medium, galaxies, and the Universe as a whole. We investigate this process using numerical simulations that follow star-forming structures across a wide range of spatial scales. We first simulate a kiloparsec (~1000 pc) portion of a galaxy using the adaptive-mesh refinement code RAMSES. From this large-scale simulation, we identify and extract multiple star-forming regions, which inherit their physical conditions from the initial galactic environment. We then zoom-in on these regions, reaching spatial scales of ~200 AU. Within these zoomed simulations, we identify dense cores and analyze their statistical properties, including the core mass function, turbulence, and fragmentation. By studying a large sample of regions with self-generated initial conditions, we capture the diversity of star-formation environments within a galaxy and quantify how core properties vary from one region to another. Finally, we compare our numerical results with observational data to assess how well the simulated core populations reproduce observed properties and how their characteristics depend on environment.

Modelling the EUV coronagraphic observations of the Full Sun Imager on board Solar Orbiter

• Ganushan GANESARATNAM (Institut d'Astrophysique Spatiale)

Because of its mass and abundance, helium plays a fundamental role in the dynamics of the solar wind. The properties of helium are characterized in the photosphere with spectroscopic surveys and in the heliosphere with in-situ measurements. However, very few measurements of the helium abundance exist in the corona above 1.1Rs where the solar wind is accelerated. The EUI and Metis instruments on board Solar Orbiter can map the spatial distribution of helium abundance from 1.5Rs to 7Rs by simultaneously observing the Lyman alpha lines of H0 et He+. These measurements will help constrain models of solar wind acceleration in the solar corona. We need a comprehensive model of the coronal emission in the passband of each instrument in order to estimate the helium abundance from EUI and Metis data. In this work, we focus on modelling the EUV observations of the Full Sun Imager (FSI) from EUI in its two narrow passbands: the 304A Lyman alpha line of He + and the 174A line of Fe9+. We built a state of art model of EUV emission lines to simulate the stray light-free images provided by the coronagraph mode of FSI. This allows us to quantify the relative contributions of the collisional excitation and photo-excitation processes. We present comparisons between FSI observations and MHD-based forward modelling.

Ultraminiaturized Gas Chromatograph Based on Lab-on-a-Chip Micro-electro Mechanical Systems for Space Exploration

• Lylia SKANDER (LATMOS (Laboratoire ATMosphères et Observations Spatiales))

Gas chromatography is a widely used analytical technique in space exploration for the separation and identification of organic compounds in complex samples. Space missions impose stringent constraints on onboard instrumentation, which must be as compact, lightweight, and low-power consuming as possible. In this context, ultra-miniaturized chromatographs based on lab-on-chip represent a promising technology for future space missions. Consequently, we developped a lab-on-chip gas chromatograph based on MEMS (Micro Electro Mechanical Systems) technology. Each analytical component (preconcentrator, column, detector) for the gas chromatograph is fabricated on silicon chips and integrated onto a fluidic interface adapted for space requirement, with the aim of achieving performant analysis of planetary samples. This work will focus on the preconcentrator unit, which is a critical part of space gas chromatographs used for in situ analysis. Solid samples are usually pyrolyzed over a given period of time and the preconcentrator is mandatory to focalize the analytes during the sampling phase, and then to inject them at the head of the chromatographic column after a thermal-desorption step. Toachieve this, the preconcentrator is integrated onto a printed circuit boards (PCB) allowing its thermal heating and temperature monitoring. To evaluate the analytical performance of MEMS preconcentrators filled with TENAX TA® as an adsorbent, analytical tests were carried out by coupling the preconcentrators to a commercial gas chromatograph. Two preconcentrators were characterized. One was packed with the adsorbent and stored for two years before testing, while the other was freshly packed with the adsorbent before the experiments. These two preconcentrators were tested to study the effect of time on their adsorption capacity and aging. To evaluate the performance of these preconcentrators, several tests were carried out, including the optimization of adsorption and desorption conditions, as well as breakthrough curve studies. These results demonstrated promising performance in line with our objectives. Both preconcentrators exhibited good linearity between the desorbed number of analytes and the injected one with with R2 > 0.991, while the freshly packed preconcentrator showed a higher adsorption capacity. The use of new adsorbent materials for the sample preconcentrator will also be investigated in order to extend the range of compounds that can be tested. This work shows that we have a component with good analytical performance. The next step will be to continue its qualification for use in space applications.

Direct dark matter detection with Transition Edge Superconductors (TES) with single-electron sensitivity for the TESSERACT Collaboration

• Alec COCHARD

The nature of dark matter is one of the most important unsolved puzzles in fundamental physics, driving searches beyond the Standard Model. Current theories suggest that this matter primarily interacts via gravitation, but also weakly with ordinary matter. However, due to the very low interaction cross-sections, it is very hard to observe and detectors sensibilities need to be increased. The TESSERACT (Transition Edge Sensors with Sub-EV Resolution And Cryogenic Targets) international collaboration aims at pushing back these limits and increase the resolution of current detectors (Billard Julien et al., Nuclear Physics B, 2024). To achieve this, teams are developing cryogenic detectors coupled with superconducting Transition Edge Sensors (TES) through several technologies (such as liquid helium, semiconductor crystals and others). At IJCLab (Orsay), our focus is on the Cryosel detector, a germanium (Ge) crystal paired with TES sensors, optimized for phonon collection via the Neganov-Trofimov-Luke (NTL) effect. By applying a non-linear electric field, even small interactions as small as a single electron-hole pair can be detected (Marnieros Stefanos et al. (2026), To be published). Moreover, the final objective is to measure both the ionisation of an interaction but also its calorimetry, allowing discrimination between the different kinds of particles. Current efforts prioritize TES design refinement and phonons simulations with G4CMP to optimize detector performance and background discrimination. By 2028, Cryosel and other TESSERACT detectors will be deployed at the Modane Underground Laboratory (LSM), leveraging its ultra-low-radioactivity environment for the first direct dark matter searches. This project also aligns with RICOCHET, targeting coherent elastic neutrino-nucleus scattering (CEνNS) at the ILL reactor, further advancing quantum-limited calorimetry and low-energy particle physics. This presentation will provide an overview of the TESSERACT collaboration, then dive in the Cryosel germanium detector developed at the IJCLAB. The quantum physics processes will be explained and discussed, linked with experimental results and simulations.

An Overview of the Ionic Composition and Energy Distribution measured by MSA during BepiColombo Swing-bys of Mercury

• Sébastien VERKERCKE (LPP)

Mercury is one of the two terrestrial planets in our Solar System possessing an intrinsic magnetic field. Being the closest planet to the Sun and having a weak magnetic field, the Hermean magnetic environment is rapidly changing and forms a tightly coupled system with the planet’s core and surface. Its closeness to the Sun causes Mercury’s surface to undergo extreme conditions, with continuous space weathering that can eject atoms from the surface to form a tenuous atmosphere called an exosphere. As these atoms become ionized, they drift along the magnetic field lines and populate the magnetosphere of Mercury. Part of these ions can re-impact the surface and eject surface atoms, forming a strong coupling between the magnetosphere, the exosphere, and the surface. More than fifty years after the Mariner 10 mission confirmed the presence of Mercury’s magnetic field (Ness et al. 1974), the structure of Mercury’s magnetosphere and its ionic composition remain poorly understood. The ESA/JAXA BepiColombo mission is currently cruising towards Mercury to study the Hermean environment, from the planet's surface up to its magnetosphere. The Mass Spectrum Analyzer (MSA) on board the ESA/JAXA BepiColombo mission collected some data during its cruise phase and during the different planetary swing-bys. The instrument measures the ions’ energy, and the Time-Of-Flight (TOF) of the ions entering MSA TOF chamber. Using the data acquired during the different swing-bys of Mercury and taking advantage of their unique trajectories, we identified different regions of Mercury’s magnetic environment and their associated ionic populations. This work will help better understand the dynamic magnetosphere of Mercury's and sets the path for the analysis of the data that will be acquired by MSA during the orbital phase starting in 2027.

TBD

• Clément PELLOUIN (CEA - AIM)

Probing Cosmic Web Structures with Generative Models

• Mehdi NOOR (Institut d'Astrophysique Spatiale)

The cosmic web, consisting of an intricate network of voids, walls, filaments and clusters, contains key information about cosmological parameters and the evolution of large-scale structures in the universe. In this context, very large ensembles of cosmological simulations are required to analyse upcoming galaxy survey data. To circumvent the associated computational cost, we aim at constructing a generative diffusion model to emulate the cosmic web while preserving its statistical properties. In this context, we explore how self-attention maps, which are learnt neural network features embedded within the diffusion model, can offer a new way to probe the underlying structure of the cosmic web. In addition to emulating high-fidelity simulations, our approach could also provide a quantitative assessment of the performance of generative models.