Thèse Laser Direct Additive Manufacturing Of Natural Inorganic Minerals For Space And Terrestrial Applications Lamista H/F - Doctorat.Gouv.Fr

Établissement : IMT Mines Albi École doctorale : MEGEP - Mécanique, Energétique, Génie civil, Procédés Laboratoire de recherche : ICA - Institut Clément Ader Direction de la thèse : Thierry SENTENAC ORCID 0000000324592525 Début de la thèse : 2027-10-01 Date limite de candidature : 2026-11-23T23:59:59 Renewed international ambitions to establish a permanent human presence on the Moon - driven notably by NASA's Artemis program, whose Artemis II mission marks a recent milestone - place in-situ resource utilization (ISRU) at the centre of exploration strategy. Sustaining astronauts and lunar infrastructures in such a resource-scarce environment requires minimizing dependence on materials shipped from Earth. Lunar regolith, abundant and readily available on site, together with solar energy, constitutes one of the very few exploitable local resources and is therefore a natural feedstock for on-site manufacturing and repair of technical parts.
Most experimental work on regolith-based additive manufacturing to date has focused on Mare-type simulants, yet current lunar base scenarios point toward a South Pole location dominated by Highland-type terrain. This PhD project addresses this under-explored question: how does Highland regolith behave under a direct, binder-free additive manufacturing process such as Selective Laser Melting (SLM), and how must the process be adapted to it? Unlike indirect routes (vat photopolymerization, binder jetting, fused filament fabrication), which require an organic binder transported from or produced on the Moon and a subsequent debinding/sintering stage reaching temperatures above 1500°C, SLM proceeds in a single step without added binder or waste heat - a decisive advantage for a resource- and energy-constrained lunar environment, though one that raises major scientific challenges, since ceramics are poor heat conductors with brittle behaviour, prone to cracking, warping and porosity.
The project is organized in five work packages. WP1 selects and characterizes a reference Highland simulant (thermal, thermo-optical and phase-transformation behaviour by FTIR, DTA/TGA/DSC, laser-flash diffusivity, SEM/TEM and XRD), benchmarked against the Mare simulants (BPY, EAC-1) already studied at ICA. WP2 develops optimized metal-ceramic and/or carbon-metal-ceramic composite feedstocks that valorise the metallic and carbon-based by-products of regolith reduction processes (pyrolysis, molten-salt electrolysis), improving toughness while limiting brittleness. WP3 optimizes the SLM process itself for the Highland simulant and its composites, addressing thermal-gradient management to limit cracking, in-process monitoring of the melt pool (thermal and optical instrumentation), and an in-situ post-processing protocol (a second, non-melting laser pass replacing ex-situ furnace annealing). WP4 validates the manufactured parts through geometric (3D scanning), volumetric (X-ray computed tomography), microstructural (SEM/EDS) and mechanical (compressive, flexural, fracture-toughness, Weibull statistics) qualification, against functional specifications co-defined with ESA for target applications such as heat exchangers, irrigation pipes or particle filters. WP5 quantifies the environmental impact and energy balance of the process through a Life Cycle Assessment (ISO 14040-14046) and identifies realistic terrestrial applications, particularly in construction and resource-constrained industrial contexts.
This work is carried out within an international, multidisciplinary framework bridging the Toulouse ISRU Task Force (ICA, IRAP, LGC, TBI, CNES), University of Cologne (DE), ESA's European Astronaut Center (EAC, Cologne (DE)) and Cologne university, and is expected to take the form of a co-supervision and a one-year stay at the EAC, with further collaboration opportunities with Hochschule Aalen and the Laser Zentrum Hannover Institute. Beyond its space application, the project directly answers the sustainable-transitions pillar of the BEST program: manufacturing parts directly from local mineral resources and solar energy, with no organic binder and minimal waste, offers a model of resource-efficient production whose relevance extends to Earth, reinforcing the project's dual Moon-to-Earth impact.
1. Context and objectives of the international collaboration
1.1 International context
Renewed international ambitions to establish a permanent human presence on the Moon - most notably through NASA's Artemis program, whose Artemis II mission marks a significant recent milestone - have placed in-situ resource utilization (ISRU) at the center of lunar exploration strategy. Sustaining astronauts and infrastructures in such an extreme and resource-scarce environment requires minimizing dependence on materials launched from Earth. The Moon offers very few exploitable local resources: lunar regolith, water in still-uncertain physical and chemical forms (permanently shadowed regions), and solar energy. Regolith is abundant, ubiquitous and directly usable as a feedstock for producing components for lunar habitats and for technological equipments required for human activities, making it a natural focus for ISRU-oriented research. These ambitious objectives have helped to identify technological and scientific bottlenecks that need to be addressed within a tight timeframe for the space sector, whilst taking particular account of the numerous and significant constraints associated with space applications. Consequently, several scientific teams are working on such topics, particularly in Asia, North America and Europe.

1.2 Toulouse University context
Within this international context, an interdisciplinary initiative in Toulouse-known as the Toulouse Task Force on ISRU-has been launched to bring together expertise in planetary science, materials science, chemical engineering, biotechnology and mechanical engineering needed to address the topic of space manufacturing in an environment with limited resources that must be preserved. Since 2019, four CNRS laboratories (IRAP (Institute of Research in Astrophysics and Planetology), LGC (Chemical Engineering Laboratory), ICA (Clément ADER Institute) and TBI (Toulouse Biotechnology Institute)) and the CNES have joined forces to tackle this interdisciplinary topic with the goal of using sources found on the Moon: regolith and solar energy, to produce oxygen and manufacture spare parts. The Toulouse Task Force is actively engaged in multidisciplinary collaborative projects. It organizes a biennial thematic workshop on ISRU; the previous one was held at the Cité de l'Espace in 2025, and the next one will be held at the European Astronaut Center (ESA-EAC) in Cologne in 2027. This workshop includes topics dedicated to physical properties of regolith, electrochemistry, additive manufacturing, biotechnology and biomining. It brings together participants from European universities such as Open University (UK), Laser Zentrum Hannover Institute - LZH (DE), Norwegian Geotechnical Institute - NGI (NO), University of Glasgow (UK), Hochschule Aalen university (DE), from space agencies such as European Space Agency - ESA ESTEC (NL), ESA EAC (DE), ESA ECSAT (UK), ESRIC (LU), CNES (FR) and from European SMEs. One of the Toulouse Task Force's objectives is to play a leading role in strengthening European collaborations in its expertise domains.

1.3 Potential of the BEST project
The BEST project offers the real opportunity to expand European collaboration with university of Cologne and the ESA-European Astronaut Center (EAC-Cologne), which complements Toulouse Task Force regional collaborations on the topics addressed [1-3] of the production of oxygen and technical parts using a Mare-type lunar regolith simulant, the Pic d'Ysson Basalt, called BPY.
The collaboration with the university of Cologne and ESA/EAC's fits within the thematic priority of sustainable transition, toward manufacturing processes using natural materials, in which wastes are fully recycled. Driven by the particularly harsh conditions of living remotely from Earth where every resource counts, this type of process can then be transferred to Earth, with less stringent constraints, as part of a sustainable development for the mobility in general and the energy sector. The methods developed here to optimize material properties, given the severe resource and energy constraints specific to the lunar environment, are expected to provide insights that can be applied to resource-efficient production on Earth, thereby reinforcing the terrestrial relevance of a research program that is otherwise space-focused. The SLM (selective laser melting) additive manufacturing process, whilst well suited to significantly reducing the environmental impact of ceramic component production, faces major obstacles in producing reliable components. For terrestrial applications, these complex obstacles have received very little scientific attention, with the focus instead on the development of indirect additive manufacturing processes which do not resolve the issues of organic material consumption, the production of reaction by-products resulting from their extraction from the ceramic parts, and the high level of waste heat required for sintering. The sustainability requirements for space applications, as already described, make it essential to address such bottlenecks, which will undoubtedly enhance the ability to adapt space-based solutions for use on Earth.
More specifically, the university of Cologne and ESA/EAC's provide operational expertise regarding regolith simulants, as well as the requirements and specifications for parts in relation to the actual conditions and sustainable constraints of the lunar surface. The ESA/EAC offers a Mare-type regolith simulant, named EAC-1, whose thermo-optical properties can be compared with those of BPY. This collaboration would provide an opportunity to explore and study other simulants, particularly those of the Highlands type located in the southern exploration area of NASA's programs. In addition, ESA/EAC and university of Cologne [4-6] have expertise in material processing methods based on direct energy-material interaction.


1.4 International Impact and Prospects
This collaboration is intended to take the form of a co-supervision and a one-year stay at the university of Cologne and ESA-EAC, alongside regular technical exchanges throughout the project.
This PhD also opens up opportunities for collaborations with Hochschule Aalen university (DE) and Laser Zentrum Hannover Institute - LZH (DE), that are already participating in the ISRU workshop and are involved in lunar additive manufacturing processes [7,8]. These potential collaborations during the doctoral program could contribute to rapid progress in this additive manufacturing process, both in terms of optimizing it for regolith and in understanding energy-matter interactions, as well as in adapting the process for use in space with a solar energy source [4,9].

2. State of the art and objectives
2.1 Global objective
The general objective of this collaboration is to prepare for production activities in space-and more specifically on the Moon-with a focus on repairing and manufacturing parts using only the material available on the Moon: regolith and solar energy in a sustainable approach and by using an additive manufacturing process. One of the project's objectives is also to analyze the potential for applying the results obtained from lunar applications to terrestrial applications.

2.2 State of the art of ceramic process
As part of this collaboration on additive manufacturing processes for ceramic materials such as regolith, the state of the art shows that indirect additive manufacturing processes are currently better understood [10-13].
Nevertheless, these indirect additive manufacturing processes (such as fused filament deposition using filled filaments) require an organic binder-which must be transported from Earth or produced on the Moon. Furthermore, the manufacturing process involves a sequence of three steps: 3D printing, debinding, and sintering. The last two steps involve supplying energy to reach temperatures above 1500°C. On the Moon, the recovery of gases and waste heat must be considered.
More generally, indirect additive manufacturing routes clearly dominate over additively manufactured ceramics. Industrial leaders such as Lithoz (Austria) and 3DCeram Sinto (France) have built mature, ISO-certified production ecosystems around vat photopolymerization (lithography-based ceramic manufacturing, SLA), where a green part is shaped by curing a ceramic-loaded resin and only fully densified in a subsequent debinding-and-sintering step. Binder jetting, pursued by players like ExOne/Desktop Metal and adopted by end-users such as Saint-Gobain, follows the same logic: shape first, densify later. Some examples of successfully 3D-printed objects include ultra-precise sintered cooling devices in aluminum nitride, casting cores used for building single crystal turbine blades for next-gen aircraft engines, and zirconia dental implants.
Another class of processes that requires no organic binder and proceeds in a single processing step, is the direct additive manufacturing processes. It is being developed, primarily using metallic materials. One of the most popular is the Selective Laser Melting (SLM) process, initially presented in a patent in the late 90s [14]. The relatively long time that has elapsed since then has enabled sufficient maturity to transfer knowledge to industry for the production of high value-added parts, particularly in the medical domain (titanium implants at Materialise [15]), space (combustion chamber, injector, etc... at SpaceX [16]), and aeronautics (turbine blades at Siemens [17]).
Moreover, adaptation of this process to ceramic materials such as silicon carbide, alumina or zirconia, is possible [9]. Simplicity of application in a single step, with direct production of objects close to the target dimensions, lack of waste heat, and the fineness of the microstructure obtained by rapid cooling, are all advantages that have led some research teams to adapt the SLM process to eutectic ceramics [10], cermets for the manufacture of injection molds for plastics [11], or hydroxyapatite for the manufacture bone implants [12]. Finally, the application of the SLM process to the exploitation of lunar regolith only began in 2014 with exploratory work by Fateri et al [13], and since then, a growing interest from the ISRU community to Powder Bed Fusion additive manufacturing technologies. Our laboratory, ICA, has also addressed research [1, 4] on selective laser melting (SLM) using lunar regolith simulants. That earlier work demonstrated the feasibility of obtaining satisfactory mechanical properties from Mare regolith simulants processed by SLM, including a simulant formulated and produced in-house under closely controlled conditions (Pic d'Ysson Basalt, BPY).
However, ceramics are poor conductors of heat, with a high melting point and brittle mechanical behavior, resulting in important temperature gradients and so the development of defects such as cracks, warping, and porosities in the final part. Mastering laser-ceramic interaction remains a major challenge for the community and represents a major scientific hurdle to overcome, based on efforts to optimize the printing strategy, particularly heat management, as proposed in this project.


2.3 Scientific objectives
The scientific specific objectives are to develop a direct additive manufacturing process, compatible with a solar-powered energy source, to manufacture and repair parts using only lunar regolith from the Highlands, with a view to establishing a lunar base at the South Pole, where the geology is dominated by highland terrain. Generally speaking, particular attention will be paid to establishing strong links between materials, processes and final properties. The sub-objectives are:
- select, evaluate, and characterize a highlands-type lunar regolith simulant and perform a physicochemical and thermo-optical characterizations across the spectral and temperature ranges of the process.
- propose an optimized version of the selected Highlands simulant in the form of metal-ceramic and/or carbon-metal-ceramic composite feedstocks to fully exploit the manufacturing process and improve the properties of the manufactured parts. A series of physicochemical and thermo-optical characterization tests within the process's spectral and temperature ranges will also be performed on this composite.
- optimize the direct additive manufacturing process (Selective Laser melting (SLM)) using this Highland simulant and composite: managing thermal gradients to limit stress concentrations and cracking, implementing in-process monitoring of the molten pool (thermal and optical instrumentation), and developing an in-situ post-processing protocol (second laser pass without remelting).
- qualify the mechanical and geometric of the produced parts, as well as validate against functional specifications defined with ESA.
- quantify the environmental impact and energy balance of the developed processing route both for spatial and terrestrial applications.
Finally, once the scientific obstacles to direct additive manufacturing have been overcome, the transfer of this technology to Earth will make it possible to manufacture parts with lower energy costs, using natural mineral materials in highly optimized quantities. This approach will be addressed for each objective
The project is organized around five work packages:

WP1 - Feedstock Selection and Preparation of Highlands simulant
PhD work of Julien GRANIER [18] focused exclusively on Mare-type regolith simulants, and established the influence of amorphous phase content and titanium dioxide (TiO) concentration on melting behaviour and resulting microstructure. Given that current lunar base scenarios favour a South Pole location, where the terrain is dominated by Highland-type geology rather than the Mare basalts studied so far, a first objective of this WP is to extend the characterization effort to highland simulants and to identify the resulting differences in thermal, optical and phase-transformation behaviors. This will be addressed using a combination of experimental techniques including Fourier-Transform Infrared spectroscopy (FTIR), Differential Thermal Analysis (DTA), Thermogravimetric Analysis (TGA), Differential Scanning Calorimetry (DSC), laser-flash thermal diffusivity measurements, Scanning and Transmission Electron Microscopy (SEM, TEM) and X-Ray Diffraction (XRD).
A literature review of available Highlands regolith simulants is planned to select a reference candidate. However, LHS-1 (Lunar Highlands Simulant, Exolith Lab / Space Resource Technologies) has been identified as a strong candidate feedstock: it is available in industrial quantities, already referenced and thoroughly characterised in the literature [19] and is increasingly used by the community for both indirect [20] and direct [21] additive manufacturing (respectively, vat-photopolymerization/DLP and Laser Directed Energy Deposition).
The selected simulant will be characterized in terms of its thermal and thermo-optical properties using scientific equipment such as: FTIR spectrometer, Differential Thermal Analysis (DTA), differential scanning calorimetry (DSC), X-ray diffraction analysis (XRD), and scanning electron microscope (SEM).
The deliverable is the result data base of thermal and thermo-optical characterization of the selected Highland simulant, which will be compared to the Mare simulant one. Particular attention will be paid to a detailed scientific analysis of the differences in the behaviour of these two regolith simulants.
This terrestrial simulant can be used as a mineral material.

WP2 - Feedstock Optimization of Highlands simulant
This second work package addresses the topic of optimizing Mare or Highlands regolith for direct additive manufacturing.
One challenge is to take into account the specific characteristics of regolith powders and regolith simulants (large grain size, wide grain size distribution, angular grain shape, etc.) and their impact on properties that are essential to the SLM process (flowability, spreadability, minimum layer thickness, etc.). Accordingly, the properties of the Highland simulant powder will be characterized, whilst also considering previously published data relating to these issues. One objective will be to determine whether the feedstock preparation parameters obtained for the Mare simulant can be applied to the Highland one.
Another known generic limitation of ceramic parts is their brittleness and high sensitivity to defects. To mitigate this, the project will explore metal-ceramic composite feedstocks, in which a ductile metallic phase is combined with the regolith ceramic matrix to preserve mechanical strength while improving toughness. Several teams are currently developing technologies to reduce regolith into oxygen and metals [22,23] and the resulting metallic by-products could be directly incorporated into regolith-based feedstocks, as already envisaged by some groups [24].
In this collaboration, one solution to explore would be to use the metal byproducts of the pyrolysis process used to generate oxygen from regolith [25]. In the cited article, oxygen yield was quantified with simultaneous recovery of metallic by-products. This raw material will then need to be prepared to meet the technical capabilities of the SLM machine: grinding to obtain a powder with controlled particle size and morphology, mixing in measured quantities with the unreduced regolith simulant, etc.
Another solution, as part of the Toulouse Regolith Task Force, is to establish a collaboration with the LGC lab (Laboratoire de Génie Chimique) which addresses research on molten-salt electrolysis of regolith [2]. The carbon electrode used in this process, once coated with electrochemically reduced regolith, could be crushed and blended with untreated regolith prior to SLM processing. During laser processing, the carbon would itself be consumed through carbothermal reduction - converting into CO and CO while contributing additional reduced (metallic) material to the melt pool, effectively coupling feedstock reduction and additive manufacturing into a single value chain.
The expected results are candidate feedstock formulations (regolith + carbon-based/metallic additives) and their characterization in terms of particle size distribution, phase composition, and thermo-optical properties, as in WP1. Moreover, the expected outcome, including terrestrial applications, is to demonstrate the ability to modify the composition of an initial mineral into another mineral. This composition represents a specific natural composition-that is, a mixture of natural minerals-designed to improve the manufacturing process


WP3 - Direct Additive Manufacturing Process Optimization with highlands simulant
A known critical limitation of the SLM process when applied to ceramics is the need to prevent the formation of defects caused by the high thermal gradients generated around the melt pool and also between the different layers of the part being printed. A comparative analysis will be carried out of the behaviour of the Highlands and Mares simulants during the SLM process and in terms of the final properties of the manufactured samples. The results will make it possible both to determine the extent to which the methodologies developed for one simulant can be applied to another, and to identify ways for adjusting the SLM process parameters for better final properties (absence of defects, mechanical properties, surface roughness).
The work package aims to optimize the SLM process as applied to this new Highland' simulant around three items: (1) management of thermal gradients to limit stress concentrations and cracking, (2) implementation of in-process monitoring of the molten pool (thermal and optical instrumentation), and (3) development of an in-situ post-processing protocol (a second laser pass without remelting).
The description of the three items is provided below:
- Thermal management of the SLM process: process parameters must be optimized to mitigate the thermal gradients responsible for warping and crack formation. Recent efforts at DTU [26] point toward maintaining a single, stable melt pool that progressively absorbs powdery regolith, thereby reducing residual porosity - an approach this project intends to build on and adapt to highland-based feedstocks. This work package plans to perform a parametric study of the process (energy density, scanning strategy, etc.) to achieve a single, stable melt. The effects of these parameters will be evaluated in terms of their impact on reducing residual porosity.

- In-process thermal monitoring: This monitoring requires process instrumentation to measure the temperature and behavior of the molten bath in real time, as demonstrated in recent studies [26,27]. Ben Dahou et al. for example successfully reconstructed cooling profiles using a FLIR X6901sc thermal camera combined with optical imaging. This work highlights the need to improve process instrumentation on the SLM 125 HL system (SLM Solutions) available at the ICA laboratory. The measurements obtained in this way could be compared with results from the literature, would allow for the refinement of process parameters and potentially serve as input data for process modeling.

- Post-processing and in-situ heat treatment: the doctoral thesis [18] and the study by Wang et al. [28] showed that annealing samples beyond their crystallization temperature builds a resistant crystalline network and substantially increases mechanical strength. This process currently requires an ex-situ treatment in a furnace, which could instead be replaced by an in-situ heating step performed during the SLM build itself - for instance, through a second, non-melting laser pass over each solidified layer. Very recent work has already demonstrated the feasibility of this in-situ post-treatment approach [29]. This item requires significant work to define and test a protocol for in-situ post-processing using a second laser pass without remelting. The properties of the parts produced in this way will be compared to those of parts annealed in an ex-situ furnace in order to validate this new approach.
This work package also applies to the optimized Highland simulant from work package 2. This second stage of the work package will demonstrate the ability to adapt parameters identified for a given mineral to other minerals. More generally, the workshop will provide a methodology for adapting process parameters to any terrestrial mineral.


WP4 - Validation of parts produced by direct additive manufacturing
The work package, which performs comprehensive microstructural, mechanical and geometric qualification of the manufactured parts, as well as their validation against functional specifications defined with ESA, is an important part of the project.
The three-dimensional geometric validation of the parts is the first step in verifying that their dimensions conform to those of the manufacturing model. We suggest validation using a lab's 3D scanner, the Creaform C-Track 3D scanner. It will provide the comparison between the printed geometries against the original CAD model, to determine whether observed deviations are isotropic or reveal a directional bias indicative of underlying thermal effects during the build.
The next step is the volumetric reconstruction of parts using a tomography method. This method applies to regolith, which is relatively low X-ray attenuation. It makes it suitable for observation by X-ray Computed Tomography [30]. This study shows a non-destructive method to inspect the full volume of a printed part, with particular attention to internal porosity distribution.
At the same time, microstructural analyses will be performed using a scanning electron microscope, accessing a new scale of the matter. The nature of the phases present, as determined by EDS analysis, and their distribution within the samples can be closely examined using the equipment already available at the ICA laboratory, the FEI - Nova NanoSEM 450.
The final step is to characterize the mechanical properties of the parts using samples manufactured at the same time as the part. Compressive, flexural and fracture toughness testing will be performed, complemented systematically by Weibull statistical analysis - essential for a brittle, defect-sensitive ceramic system - to assess the reliability and viability of manufactured parts for their intended function.
In addition to these generic characterizations, the parts will be evaluated against specific functional requirements to be defined jointly with ESA, depending on the target application (e.g. heat exchanger, irrigation pipe, particle filter), ensuring that validation criteria remain grounded in realistic mission use-cases.


WP5 - Analysis of the SLM process environmental impact and terrestrial applications
Environmental assessments will be evaluated using LCAs (ISO 14040-14046) to quantify the environmental impacts of manufacturing using the SLM process. The objectives of this task are to gather the information necessary to compile the inventory, calculate the initial impacts derived from the LCA, and compare scenarios to quantify the differences between impacts in order to identify key points for proposing potential improvements toward a possible eco-design of the process coupled with large-scale implementation.
In this sense, the thesis goals are fully aligned with the efforts undertaken by CNES and ESA through Spaceship FR with other ISRU and life support projects like MELISSA:
- Increased autonomy: Less dependence on supplies shipped from Earth.
- Safety: Reduced risk of supply shortages or pollution from materials not sourced locally.
- Circularity of the space economy: manufacturing of replacement parts for a lunar rover or habitats, 3D-printed from regolith, which is readily available.
The candidate will draw on ESA's expertise in this field, which will facilitate their interdisciplinary development.
Once the spatial context has been established, the process may finally inspire sustainable innovations for Earth, in line with the UN's Sustainable Development Goals, particularly in the construction sector: specifically, the manufacture of plain or composite materials for insulation or temporary structures for a lower environmental and energetic cost.
As part of this technology transfer to Earth-scale applications, the doctoral student will be required to identify potential applications, particularly those based on mineral resources that are widely available on a European or global scale, or those derived from mining operations, for example.

The candidate should hold a Master's degree in materials science, mechanical engineering, chemistry, or related fields. A multidisciplinary background is preferred, with experience in one or more of the following areas being highly desirable:
- Additive Manufacturing, in particular direct processes (DED, SLM)
- In Situ Resource Utilization (ISRU) with regolith
- Mechanical testing of ceramic materials
- XRD, FTIR, DSC, SEM, or related structural characterization
- Redox processes
A B2 level in English (French optional) is required.
Knowledge of the European space sector, particularly in the field of ISRU, is highly valued.
The candidate must demonstrate the ability to communicate effectively, both in writing and orally, to contribute to the dissemination of scientific results through conferences and scientific articles.
Willingness to work as part of a team is appreciated, as the thesis will be conducted across two different institutions and will require frequent discussions to assess the project's progress

Application link : https://edd-projets.utoulouse.fr/

• Anglais
• SEM
• Flash
• Français
• AM
• Chimie