# Fileset

[Advanced Materials - 2024 - Stier - Materials Acceleration Platforms  MAPs   Accelerating Materials Research and.pdf](https://mdr.nims.go.jp/filesets/1615af58-d7fc-4275-b663-67bb8a83800d/download)

## Creator

[Simon P. Stier](https://orcid.org/0000-0003-0410-3616), Christoph Kreisbeck, Holger Ihssen, [Matthias Albert Popp](https://orcid.org/0000-0001-7620-1245), [Jens Hauch](https://orcid.org/0000-0003-4384-2112), [Kourosh Malek](https://orcid.org/0000-0002-3021-0813), [Marine Reynaud](https://orcid.org/0000-0002-0156-8701), T.P.M. Goumans, [Johan Carlsson](https://orcid.org/0000-0002-1596-0923), Ilian Todorov, [Lukas Gold](https://orcid.org/0000-0001-7444-2969), [Andreas Räder](https://orcid.org/0000-0002-2673-0832), [Wolfgang Wenzel](https://orcid.org/0000-0001-9487-4689), [Shahbaz Tareq Bandesha](https://orcid.org/0000-0002-9365-8703), Philippe Jacques, [Francisco Garcia‐Moreno](https://orcid.org/0000-0001-6114-0240), [Oier Arcelus](https://orcid.org/0000-0001-9075-3820), [Pascal Friederich](https://orcid.org/0000-0003-4465-1465), [Simon Clark](https://orcid.org/0000-0002-8758-6109), [Mario Maglione](https://orcid.org/0000-0002-5747-8879), [Anssi Laukkanen](https://orcid.org/0000-0002-2308-744X), [Ivano Eligio Castelli](https://orcid.org/0000-0001-5880-5045), [Javier Carrasco](https://orcid.org/0000-0003-3117-6933), [Montserrat Casas Cabanas](https://orcid.org/0000-0002-9298-2333), [Helge Sören Stein](https://orcid.org/0000-0002-3461-0232), [Ozlem Ozcan](https://orcid.org/0000-0003-1965-7996), [David Elbert](https://orcid.org/0000-0002-2292-180X), [Karsten Reuter](https://orcid.org/0000-0001-8473-8659), Christoph Scheurer, [Masahiko Demura](https://orcid.org/0000-0002-7308-3041), Sang Soo Han, [Tejs Vegge](https://orcid.org/0000-0002-1484-0284), [Sawako Nakamae](https://orcid.org/0000-0001-9174-3165), [Monica Fabrizio](https://orcid.org/0000-0003-2257-8375), [Mark Kozdras](https://orcid.org/0000-0001-5113-3047)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Materials Acceleration Platforms (MAPs): Accelerating Materials Research and Development to Meet Urgent Societal Challenges](https://mdr.nims.go.jp/datasets/863afaeb-c149-4ff2-8bec-0bb8fd6d6686)

## Fulltext

Materials Acceleration Platforms (MAPs): Accelerating Materials Research and Development to Meet Urgent Societal ChallengesPERSPECTIVEwww.advmat.deMaterials Acceleration Platforms (MAPs): AcceleratingMaterials Research and Development to Meet UrgentSocietal ChallengesSimon P. Stier,* Christoph Kreisbeck, Holger Ihssen, Matthias Albert Popp, Jens Hauch,Kourosh Malek, Marine Reynaud, T.P.M. Goumans, Johan Carlsson, Ilian Todorov,Lukas Gold, Andreas Räder, Wolfgang Wenzel, Shahbaz Tareq Bandesha, Philippe Jacques,Francisco Garcia-Moreno, Oier Arcelus, Pascal Friederich, Simon Clark, Mario Maglione,Anssi Laukkanen, Ivano Eligio Castelli, Javier Carrasco, Montserrat Casas Cabanas,Helge Sören Stein, Ozlem Ozcan, David Elbert, Karsten Reuter, Christoph Scheurer,Masahiko Demura, Sang Soo Han, Tejs Vegge, Sawako Nakamae, Monica Fabrizio,and Mark KozdrasClimate Change and Materials Criticality challenges are driving urgent responses from global governments. These globalresponses drive policy to achieve sustainable, resilient, clean solutions with Advanced Materials (AdMats) for industrialsupply chains and economic prosperity. The research landscape comprising industry, academe, and government identi-fied a critical path to accelerate the Green Transition far beyond slow conventional research through Digital Technologiesthat harness Artificial Intelligence, Smart Automation and High Performance Computing through Materials AccelerationPlatforms, MAPs. In this perspective, following the short paper, a broad overview about the challenges addressed, ex-isting projects and building blocks of MAPs will be provided while concluding with a review of the remaining gaps andmeasures to overcome them.S. P. Stier, M. A. Popp, L. Gold, A. Räder, S. T. BandeshaDepartment Digital TransformationTLZ-RTFraunhofer ISCNeunerplatz 2, 97082 Würzburg, GermanyE-mail: simon.stier@isc.fraunhofer.deC. KreisbeckAixelo Inc.Cambridge, MA 02141, USAH. IhssenHelmholtz AssociationRue du Trône 98, Bruxelles B-1050, BelgiumJ. HauchForschungszentrum Jülich GmbH, Helmholtz-Institut Erlangen-Nürnbergfor Renewable Energy (HI ERN)Institute of Materials for Electronics and Energy Technology (i-MEET)91058 Erlangen, GermanyThe ORCID identification number(s) for the author(s) of this articlecan be found under https://doi.org/10.1002/adma.202407791© 2024 The Author(s). Advanced Materials published by Wiley-VCHGmbH. This is an open access article under the terms of the CreativeCommons Attribution License, which permits use, distribution andreproduction in any medium, provided the original work is properly cited.DOI: 10.1002/adma.202407791K. MalekForschungszentrum Jülich GmbHTheory and Computation of Energy Materials (IEK-13)Institute of Energy and Climate Research (IEK)52428 Jülich, GermanyM. Reynaud, O. Arcelus, J. Carrasco, M. C. CabanasCentro de Investigación Cooperativa de Energías Alternativas(CIC energiGUNE)Basque Research and Technology Alliance (BRTA)Parque Tecnológico de Álava, Albert Einstein 48, Vitoria-Gasteiz 01510,SpainT. GoumansSoftware for Chemistry & Materials BVDe Boelelaan 1083, Amsterdam 1081 HV, The NetherlandsJ. CarlssonDassault Systemes Deutschland GmbH51063 Cologne, GermanyI. TodorovScientific Computing DepartmentScience and Technology Facilities Council, Daresbury LaboratoryWarrington WA4 4AD, UKW. Wenzel, P. FriederichInstitute of Nanotechnology (INT)Karlsruhe Institute of TechnologyHermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen,GermanyAdv. Mater. 2024, 36, 2407791 2407791 (1 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbHhttp://www.advmat.demailto:simon.stier@isc.fraunhofer.dehttps://doi.org/10.1002/adma.202407791http://creativecommons.org/licenses/by/4.0/http://creativecommons.org/licenses/by/4.0/http://crossmark.crossref.org/dialog/?doi=10.1002%2Fadma.202407791&domain=pdf&date_stamp=2024-09-06www.advancedsciencenews.com www.advmat.de1. Motivation and Background1.1. A Global ChallengeGlobally, governments have embraced an indisputable fact: so-ciety will face significant impact from climate change and ourP. JacquesEMIRI AISBLRue de Ransbeek 310, Brussels B-1120, BelgiumF. Garcia-MorenoInstitute of Applied MaterialsHelmholtz-Zentrum Berlin für Materialien und EnergieHahn-Meitner-Platz 1, 14109 Berlin, GermanyS. ClarkSINTEF IndustryNew Energy SolutionsSem Sælands vei 12, Trondheim 7034, NorwayM. MaglioneInstitut de Chimie de la Matière Condensée de Bordeaux (ICMCB)-UMR5026, CNRSUniversité de Bordeaux87 Avenue du Docteur Schweitzer, Pessac F-33608, FranceA. LaukkanenVTT Technical Research Centre of Finland Ltd.Espoo 02044, FinlandI. E. Castelli, T. VeggeDepartment of Energy Conversion and StorageTechnical University of DenmarkKgs. Lyngby DK-2800, DenmarkJ. CarrascoIKERBASQUE - Basque Foundation for SciencePlaza Euskadi 5, Bilbao 48009, SpainH. S. SteinTechnical University of Munich (TUM)Digital CatalysisLichtenbergstr. 4, 85748 Garching b. München, GermanyO. OzcanFederal Institute for Materials Research and Testing (BAM)Unter den Eichen 87, 12205 Berlin, GermanyD. ElbertHopkins Extreme Materials InstituteJohns Hopkins UniversityBaltimore, MD 21218, USAK. Reuter, C. ScheurerFritz-Haber-Institut der Max-Planck-GesellschafFaradayweg 4-6, 14195 Berlin, GermanyM. DemuraNational Institute for Materials Science (NIMS)1-2-1 Sengen, Tsukuba, Ibaraki 305-0044, JapanS. S. HanKorea Institute of Science and Technology (KIST)5 Hwarangno 14-gil, Seongbuk-gu, Seoul 136-791, Republic of KoreaS. NakamaeService de physique de l’état condensé, CEA, CNRSUniversité Paris-SaclayCEA Saclay, Gif-sur-Yvette Cedex 91191, FranceM. FabrizioInstitute of Condensed Matter and Technologies for EnergyNational Research CouncilCorso Stati Uniti, 4 - 35127, Padua, ItalyM. KozdrasCanmet MATERIALSNatural Resources Canada183 Longwood Road South, Hamilton, ON L8P 0A5, Canadagrowing demand for natural resources. This will inevitably callfor new policies and technologies to improve quality of life andmitigate negative impacts globally and locally.[1] In the broadestsense, governments are addressing these challenges through keyinitiatives under the United Nations, specifically the UN Frame-work Convention on Climate Change[2] and the 2030 StrategicAgenda for Sustainable Development[3] with its 17 sustainabledevelopment goals[4] (SDGs).Meanwhile, global energy consumption continues to rise, andrecent gas shortages in Europe have led to an increase in coal us-age in 2022 (IEA press release[5]). This highlights the urgent needto address SDG 7 (clean and affordable energy) by developingsustainable and abundant materials for energy generation andstorage. This will not only help combat the enhanced greenhouseeffect and the strong climate changes (SDG 13) but also affectsdrinking water safety (as highlighted by the IPCC report[6]) andsupport efforts to achieve SDG 6 (clean water and sanitation). In abroader perspective, it is necessary that we learn how to more effi-ciently produce, recycle and reuse materials to make the best useof our limited resources on earth. This requires concerted effortsand partnerships across government, industry, and academia todrive innovation, and ensure sustainable development for the fu-ture generations.These global actions drive regional and national governmentpolicy in both developed and developing nations that must fo-cus on clean and green, as well as sustainability and resilience.These policies co-exist in a framework of an industrial complexand its associated economic prosperity leading to societal chal-lenges such as sustainable land use, efficient mobility and liv-able cities.1.2. Importance of Advanced MaterialsIt is known that climate change impact mitigation strategies, en-ergy resilience and economic prosperity hinge significantly onAdvanced Materials (AdMats). With 20% of the industrial baseand 70% of technical innovations relying on AdMats,[7] close at-tention to materials supply chains and accelerated materials dis-covery and development is needed. Further, as the percentageof the cost of materials technologies, particularly renewables, innew devices and systems increases, intense efforts are needed tofind a lower cost, less toxic, environmentally friendly and moreearth-abundant AdMat solutions. AdMats, alongside the press-ing demand for accelerated advancements in discovery, develop-ment, and commercialization of novel processes, devices, andsystems, are intricately intertwined with the environmental, en-ergy, and resource imperatives set forth by governments world-wide.Aligned with the industrial landscape, they aim to foster sus-tainable practices, facilit ating the judicious utilization of raw ma-terials, ensuring a resilient energy infrastructure, fostering thedevelopment of eco-friendly chemical materials, and enhancinghealthcare solutions for all. Their importance is not based solelyon the value added by the industries directly involved in materialsproduction, but rather on their significance as a cross-sectionaltechnology. Materials research can have a significant and de-cisive impact on raw material sovereignty if we accelerate thedevelopment and scaling up of technologies that require fewerAdv. Mater. 2024, 36, 2407791 2407791 (2 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deFigure 1. Nine innovation markets identified by the Materials 2030 Manifesto for which the development of advanced materials is critical. Reproducedwith permission.[13] Copyright 2022, AMI2030.(elimination) critical raw materials (CRMs) or those that alreadyexist on territory (replacement). New materials are also neededfor new device design for a circular economy and life cycle man-agement that minimizes CRM usage (e.g., longer product lifespan, reuse, remanufacture, dematerialization, from products toservices). Material research can also support the extraction ofCRMs from old excavation ore materials or mine waste using,for example, floatation methods with more specific chemicals.European Strategy: In Europe, the Green Deal IndustrialPlan,[8] supports the transition to climate neutrality by enhanc-ing the competitiveness of Europe’s net-zero industry. Bolsteredby the Digital Transition[9], the Green Deal played a major rolein setting the 2020 Industrial Strategy.[10–12] This strategy is builtupon the strength of a single EU market as its most important as-set that offers resilience through certainty, scale and global accessfor European companies.The Materials 2030 Manifesto: Most recently, the critical un-derpinning of advanced materials on a green transformation, andeconomic security and prosperity was articulated in the AdvancedMaterials 2030 Manifesto.[13] It calls upon systemic collaborationhorizontally across sectors and vertically within the sector sup-ply chains. The systemic collaborative approach will offer faster,scalable and efficient responses to advanced materials challengesto create opportunities for the European society. The Manifestoidentified nine innovation markets (see Figure 1) and exampleapplications critical within the European context. The need andactions necessary to achieve such a systemic approach under theManifesto were elaborated in the Materials 2030 Roadmap,[14]and established a draft set of actions needed for the realizationof the Materials 2030 initiative.[15] A crucial factor in this initia-tive is that it is founded on a public-private partnership modelwhere a competitive advanced materials sector could lead globallyby identifying 1) low-hanging fruit; 2) long-term game changersaccelerated by efficient materials research (technology push); and3) the requirements of the markets for new solutions or products(market pull). This is where the convergence of digital technolo-gies, advanced computation, and automated experimental syn-thesis and characterization moves high-throughput experimenta-tion, processing and analytics to a new level - one seeking to spanthe materials value chain and accelerate the discovery, develop-ment and commercialization of new advanced materials, devicesand systems. The degree and nature of acceleration will vary butrecent examples (see Section 2.3) have confirmed that advancedmaterials development cycles could be reduced from tens of yearsto years, with a commensurate reduction in the development cy-cle costs. These will be explored further in this article.1.3. The History of Accelerated Advanced Materials ResearchHigh-Throughput History: First, AdMats screening usinghigh-throughput theoretical and computational predictive ap-proaches is an effective initial step in the acceleration of the R&Dprocess.[16] There is in fact a rich history in acceleration of re-search through accessible high-throughput research strategiespioneered in Germany in the 1950s.[17] With the push for automa-tion at the turn of the century in Asia and the USA, pioneersdeployed automation through the paradigm of combinatorialAdv. Mater. 2024, 36, 2407791 2407791 (3 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.dematerials science to accelerate research. The realization by theUSA government that AdMats play a critical role in global com-petitiveness then led to the USA’s Materials Genome Initiativein the early 2010s.[18,19] The idea of inverse design within thisinitiative was driven through theoretical predictions that identifyattractive candidate materials for experimental validation.High-Throughput state of the Art: Next, automated, combina-torial and/or high-throughput experiments aim at rapidly andsystematically screening large library of materials. While au-tomatic synthesis of organic molecules has become a stan-dard in pharmaceutical R&D and synthetic and industrialchemistry,[20,21] accelerating experimental exploration of AdMatsin other fields such as inorganic battery materials remain chal-lenging because of the complexity of automating some of thesynthetic routes employed for the preparation of these materials.However, by building on high-throughput experimentation, au-tomated systems can significantly reduce the quantity, and there-fore the cost of pricy input materials and improve statistical sig-nificance, allowing for a larger number of highly efficient exper-iments under comparable budgets. As a result, the costs asso-ciated with materials discovery and development can be signifi-cantly reduced, paving the way for transformative breakthroughsin various fields of research.Combination of High-Throughput with AI: Further, the gen-eralization and systematization of digitalization as well as therecent and rapid advances in Artificial Intelligence (AI) offer aunique opportunity to transform the R&D culture and acceler-ate the development of AdMats. AI has already demonstratedits ability to speed up individual research tasks, such as plan-ning experimental campaigns in a high-dimensional materi-als development workspace, processing large amounts of data(Big Data); and accelerating simulation and modeling methods.However, AI research can go beyond these tasks to include therigorous organization and standardization of R&D data, suchas creating ontologies and enabling semantic querying to in-terlink data repositories with data production and processingpipelines. Overall, AI can significantly enhance the AdMats de-velopment process by enabling more efficient and effective R&Dpractices.From High-Throughput and AI to MAPs: By gathering andarticulating high-throughput computational and experimentalapproaches together with digitalization and AI, so-called Mate-rial Acceleration Platforms (MAPs) represent a disruptive, cross-sectional technology across materials science disciplines that willenable faster and more cost-effective R&D as outlined in the pre-sented paper. Given the significant dependence of society on ad-vanced materials, and the revolution occurring in AI methodsand advanced automation technologies, there is no surprise thatMAPs will play a major role in what Mario Molina, Nobel Laure-ate, considers the “second revolution in science.”[22]2. About MAPs2.1. The Idea of a MAPThe concept of a MAP is to use methods of automation anddigitalization in material research to accelerate innovation (c.f.Figure 2) by orders of magnitudes in order to tackle specificpresent and future challenges of mankind.Figure 2. Automation in production and research. Automation in produc-tion has accelerated the manufacturing of goods and resulted in a massiveincrease in materialistic wealth. Conventional materials research is domi-nated by human-centric processes. High-throughput methods and AI dataanalytics accelerate parts of the scientific workflow but leave scientists in-tuition as the origin of materials discovery and further bottlenecks in theworkflow where manual experimentation is still needed. The synergisticcombination of both key elements results in the idea of a MAP which aimsto accelerate the entire research cycle to yield advanced materials for solv-ing problems and tackling challenges of society.MAPs[22] provide a general framework to jointly work on faster,more efficient development and production of new advanced ma-terials for specific applications and targets, with energy mate-rials being a focal area of acceleration today.[23] MAPs are col-laborative infrastructures integrating automation and digitaliza-tion to accelerate the innovation and development of AdMats.They complement and enhance the productivity of a researchteam by replacing labor-intensive experimentation and process-ing with automated high-throughput laboratory techniques andAI-supported data analysis, simulation/modeling and experi-mental planning. This closes the optimization loops and allowsscientists to focus on rapidly accessed, high-fidelity data. Fur-ther, with the world population rapidly aging, MAPs can alsohelp increase productivity to counterbalance the shrinking ac-tive research workforce. Thus, they will enable more rapid adap-tation to ongoing changes in the environment, geopolitical is-sues, and regulation, and in particular they will help the devel-opment of new materials using, and reusing, locally abundantresources.The MAP concept (c.f. Figure 3) follows the vision of a human-supervised, self-learning autonomous lab which automates theproduction and evaluation (synthesis and characterization) of amaterial or a device in an environment supported by AI-informedexperimental planning and data processing, computation andsimulation that can also be accelerated by AI, and is built ona framework of mechatronics to handle materials processingand analysis.This enables MAPs to provide the following capabilities:1. MAPs automate mundane and repetitive tasks, allowing re-searchers to focus on the underlying scientific questions.Adv. Mater. 2024, 36, 2407791 2407791 (4 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deFigure 3. The MAP concept, integrating automated synthesis, manu-facturing, and characterization with high-performance computing andAI/Machine Learning. Reproduced with permission. Copyright 2024, MarkKozdras.2. MAPs are end-to-end workflows that supersede manual andstepwise approaches.3. MAPs can target (not optimize in a narrow range) materialsand devices in a much larger experimental space that has yetto be imagined.4. MAPs solve multi-dimensional and multi-level problems thatconventional methods cannot handle.5. MAPs are predestined for inverse design and thus, ideal toolsfor application-oriented material development.6. MAPs can be used in a shared way by individuals, educators,trainers, and organizations to streamline their learning anddevelopment initiatives.2.2. Why Will it Be Superior?Closed Loop: Through the use of advanced computational ap-proaches guiding experiments for new materials and devices,which in turn provide feedback in a continuous improvementloop, MAPs target the whole addressable parameter space.[24] Forinstance, a battery MAP simultaneously optimizes the materials(electrodes, electrolytes) and the whole battery itself, on a set ofdefined properties for the end use (e.g., charging and dischargetimes under a set of conditions, capacity and safety amongst oth-ers).FAIR: MAPs will revolutionize data-driven experimentationproducing complete, fully documented, meta-data rich and struc-tured data sets. They are FAIR by design as they will include suffi-cient information about the process or material they encompass.FAIR is an acronym for Findability, Accessibility, Interoperability,and Reuse and refers to the guiding principles for scientific datamanagement and stewardship.[25,26] The FAIR data principles fo-cus on machine-readable and processable metadata on scientificdata to enable the use of computational tools to support humansprocessing data with increasing volume, complexity, and creationspeed. Findable implies that data should be easy to find for hu-mans and computers, which requires a dataset to have a glob-ally unique identifier and to be described by rich metadata, thatallows to filter and distinguish datasets. Accessible means that(meta)data are retrievable by their identifier from a publicly acces-sible source, with optional authentication and authorization, overa standardized communication protocol. FAIR (meta)data needto use a formal, accessible, shared and broadly applicable lan-guage for knowledge representations. It is thereby Interoperable.The reuse of data is the ultimate goal of FAIR. This last princi-ple requires the scope and limitations of the data to be described,which is achieved by including all information needed to repro-duce, reevaluate and repurpose (meta)data in a particular context.In the first instance, MAPs will satisfy the internal integrity ofthe digital process/material and in the second they will provide asufficient level of autogenerated description - metadata, seman-tics and ontology to comply with minimum FAIR standards (e.g.,OntoCommons,[27] DOME 4.0,[28,29] FAIR4RS[30]).Holistic: MAPs provide a holistic approach to developing andproducing new materials for optimum impact. They can acceler-ate materials development across a myriad of industrial sectors,including chemicals, electronics, pharmaceuticals, and energy.MAPs go beyond a single material, optimizing the whole devicefor the intended application. MAPs provide a means to simulta-neously develop and test materials and processes all the way todevices and systems.Collaborative and Cross-Domain: MAPs enable unprece-dented connectivity of material data across multiple industrysectors, leveraging AI-driven transfer learning from energy topharmaceuticals and biomaterials. In addition, designed to facil-itate extensive collaboration, both virtually and physically, MAPsempower researchers and stakeholders alike. Overall, this willfoster new collaborative approaches exploiting synergies previ-ously undiscovered, while advocating for “Safe and Sustainableby Design” approaches. By engaging basic researchers, transla-tors from different disciplines, and industry RD&I, MAPs seam-lessly traverse Technology-Readiness Level scales, mitigating the“valley of death” and bridge knowledge gaps between industries.Decentralized and Modular: MAPs can be implemented inone location, or they can integrate decentralized pieces of equip-ment in different geographic locations through a centralized datarepository, experimental analysis and orchestration. They can beremotely controlled providing access for researchers from all overthe globe that can utilize the data and submit experimental de-signs and questions. As such they make it possible to utilize intel-lectual capacities in countries where experimental infrastructuremay not meet high standards, offering “experiments as a service”to researchers in these regions.2.3. Research Projects and InitiativesThere is already consensus about the opportunities. While thebroad application of MAPs is still outstanding there are alreadya number of platforms that are beyond the concept phase anddemonstrate the potential of these approaches. There has beena growing number of programs and initiatives seek to accelerateAdv. Mater. 2024, 36, 2407791 2407791 (5 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deFigure 4. One of the robotic platforms (located at Fraunhofer ISC) developed in BIG-MAP utilizing a pluggable module system where each module isdedicated to a specific task like dosing, mixing, heating, evaporation and phase separation. A six-axis robot in the center of the platform integrates themodules and handles the material and container transport between them. (Reproduced with permission. Copyright 2024, Fraunhofer ISC.).materials development and technology transition through auto-mated closed-loop discovery systems that blend robotic platformstogether with high-throughput screening (HTS), all orchestratedby AI-based data analytics and computing.[31,32] They have beendeployed in areas such as peptide synthesis,[33] the pharma-ceutical industry,[34,35] and the development of high-efficiencysolar cells.[36] More recently, semi- and fully automated plat-forms have started to target chemical synthesis and fuel pro-duction (from atmospheric feedstocks).[37,38] Furthermore, thereare several projects preparing the background infrastructurethat will be necessary to exploit the full potential of these plat-forms and their output for the broader research community andsociety:BIG-MAP[39]: The Battery Interface Genome - Materials Ac-celeration Platform (BIG-MAP) project is part of the large-scaleand long-term European research initiative BATTERY 2030+.[40]It proposes a radical paradigm shift in battery innovation, whichwill lead to a dramatic speed-up in the battery discovery and in-novation time; reaching a 5–10-fold increase relative to the cur-rent rate of discovery within the next 5–10 years. BIG-MAP re-lies on the development of a unique R&D infrastructure and ac-celerated methodology that unites and integrates insights fromleading experts, competencies and data across the entire battery(discovery) value chain with AI,[41] High Performance Comput-ing (HPC), large-scale and high-throughput characterization, andautonomous synthesis robotics.In short, BIG-MAP aims to reinvent the way we invent bat-teries and to develop core modules and Key Demonstrators ofa MAP specifically designed for accelerated discovery of batterymaterials and interfaces.MAPs@Fraunhofer ISC: In addition to research on batterymaterials (see BIG-MAP), the Fraunhofer Institute for SilicateResearch (ISC) is also active in many other fields such as glass,ceramics, polymers, particles and biomaterials, where it is in-creasingly adopting the MAP concept (c.f. Figure 4).[42] Partic-ularly in the field of biological tissues engineering,[43] signifi-cant increases in quality and throughput have been achievedthrough laboratory automation (c.f. Figure 5) while human effortfor nanoparticle synthesis could be reduced by ≈44 %.[44] Overall,the focus at Fraunhofer ISC lies strongly on modular concepts,both in hardware and software, so that individual modules can beeasily transferred between the different specialized departments.VIPERLAB[45] and AutoPeroSol[46,47]: The EU-funded projectVIPERLAB is offering access to a distributed virtual laboratoryfor Perovskite solar cells. While VIPERLAB is not a MAP, it has astrong focus on data curation and databases for perovskite solarcell data that will create an important data repository that futureMAPs can utilize. The AutoPeroSol project funded by the Ger-man Helmholtz Association has a similar focus on data sharingbetween Helmholtz Centers. A strong collaboration between theprojects has evolved targeting a comprehensive ontology for per-ovskite solar cells which could form an important basis of seman-tic learning for MAPs.Figure 5. Robotic cell for tissue engineering utilizing a two-arm robot andconventional lab equipment. (Reproduced under the terms of the CC-BYlicense.[43] Copyright 2021, Haeusner et al.).Adv. Mater. 2024, 36, 2407791 2407791 (6 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deFigure 6. Line1 is a functional MAP for the development of organic solar cells at the Helmholtz Institute Erlangen–Nuremberg. Reproduced with per-mission. Copyright 2023, HI ERN/Kurt Fuchs.AMANDA/Line1[48]: One of the most complex and advancedMAPs currently operating. AMANDA is the orchestration anddata management interface that is able to manage multiple MAPsat the same time. Line1 is an automated processing line forthe manufacturing and characterization of full organic solarcells (c.f. Figure 6). Line1 is capable of formulating inks, de-positing thin functional films, characterization of films (imag-ing + spectral characterization), evaporating electrodes, perform-ing jV-characterization under illumination, and light aging offull organic photovoltaic (OPV) devices to determine stability.AMANDA integrates Bayesian Optimization and Gaussian Pro-cess regression to perform fast optimization of devices (c. f.Figure 7). AMANDA and Line1 were built with funding by theDFG and the project ELF-PV by the state of Bavaria in coop-eration between the Friedrich–Alexander–University Erlangen–Nuremberg and the Helmholtz Institute Erlangen–Nurembergfor Renewable energies (HI ERN). Over time, AMANDA has,with reference to edisonian high-throughput experimentation,achieved an acceleration starting at 30-fold[49] reaching recentlyover 100-fold.[50,51] Among others this has led to the identificationof ultrastable perovskite compositions in the multi-dimensionalparameter space.[52]MAPz@BAM: Federal Institute for Materials Research andTesting (BAM) started the module development for MAPs in2020 for three application areas: Nano and advanced materials,structural materials for green energy transition and sustainableconstruction materials (c.f. Figure 8). In different MAPs alongthese priority topics, the generated data is used for experimentplanning through sequential learning and artificial intelligence.BAM MAPs follow a circular and multi-stage approach by in-tegrating up-scaling, process development and stability throughproxy experiments and workflows. Through collaborations withnational initiatives MaterialDigital and MatWerk/FAIRMat, con-cepts for data and metadata structures as well as data shar-ing workflows are established. Different modules from differentbranches of BAM are currently being merged in MAPz@BAMFigure 7. Example of an autonomous optimization of a quaternary OPV-mixture performed on AMANDA/Line1. The Bayesian optimizer Phoenics wasused to optimize two objectives (initial efficiency and efficiency after aging test) in a 3D-process parameter space. Six samples were investigated in eachiteration, the optimum was found after nine iterations. (Reproduced with permission.[53] Copyright 2021, Tobias Osterrieder.).Adv. Mater. 2024, 36, 2407791 2407791 (7 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deFigure 8. Nano and advanced materials and glass modules (built in cooperation with Fraunhofer ISC) of MAPz@BAM (Reproduced with permission.Copyright 2024, BAM).(MAP-Zentrum of BAM) that will be operational in the Sum-mer of 2025 in the Adlershof Campus as a core facility for trans-fer projects.ION-SELF: A Spanish project initiated in 2019 for the devel-opment of an autonomous experimental MAP, capable of makingeffective predictions and self-driving high-throughput synthesesand characterizations of battery materials.[54] This innovative au-tonomous lab, which is continuously upgraded, has already per-mitted to successfully explore new families of electrode materi-als in collaboration with several European battery industrial part-ners.Synbio-MAP: An effort in Finland by VTT to develop abiotechnology-based MAP linking synthetic biology to highthroughput screening and testing of materials and precursors tophysics-driven modeling and AI for structure and functionalityprediction and design.[55] The initial focal areas are protein-basedmaterials and bioplastic, especially polyesters.AMDEE[56]: The AI-Driven Integrated and Automated Mate-rials Design for Extreme Environments (AMDEE) project in theUnited States linking robotic, high-throughput characterization,a centralized automation control data hub, and streaming scien-tific data focused on development of refractory multi-principal-element alloys (RMPEAs), light alloys, and ceramics with dy-namic and high-temperature properties needed for impact resis-tance and hypersonic performance (c.f. Figure 9). AMDEE labo-ratories utilize streaming, event-driven integration of data infras-tructure integrates design, decision, and control loops for orches-trated deployment of AI/ML decisions.BasCat/FHI: The Fritz Haber Institute - UniCat BASFJointLab[57] aims to develop accelerated or self-driving laborato-ries for heterogeneous catalysis (often also referred to as digitalcatalysis, c.f. Figure 10). In recent research[58] an active learningloop was established, in which an adaptive design of the experi-ment algorithm queried batches of experimental data. The cata-lysts were synthesized and tested, and on the basis of the gainedperformance data, a model for the explored design space was es-tablished. The design space consisted of promoter species andtheir concentrations to enhance a specific reaction (conversionof propane to propylene). Within less than 100 catalysts thus cre-ated and tested (requiring less than 3 months), a new promotercombination was found that is competitive to the one presentlyused by BASF in their commercial Oleflex process.MARK@KIST: The Korea Institute of Science and Tech-nology (KIST) launched the MARK (Materials Acceleration Re-search K-lab) project in 2022. This initiative aims to accelerate thediscovery of nano-materials for electrochemical water-splittingFigure 9. Left: UR10e robot hovering above the bi-directional conveyance system that delivers samples in the AMDEE project. Six such robots areequipped with vacuum grippers to deliver samples to testing and characterization stations around the laboratory. Right: A smaller, UR3e robot withinthe MAXIMA high-transmission XRD and XRF system receives samples from the conveyance-line robot to complete fully automated sample loading andX-ray characterization. Reproduced with permission. Copyright 2024, David Elbert.Adv. Mater. 2024, 36, 2407791 2407791 (8 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deFigure 10. Concept of accelerated catalysts laboratories at BasCat (left, reproduced with permission.[57] Copyright 2024, Frederik Rüther) and identifiedpromotor system (right, reproduced under the terms of the CC-BY 4.0 license.[58] Copyright 2024, Kunkel et al.).catalysts and display quantum dots (see Figure 11). The projectemphasizes full lab automation, the development of AI modelsfor experimental planning, the creation of an operating system(OS) for orchestration, data generation including negative mate-rials data, and data mining from literature using natural languageprocessing. Recently, KIST emphasized the value of autonomouslaboratories, which offer twofold benefits of enhancing materialdevelopmental efficiency and elucidating novel chemical knowl-edge by analyzing the datasets accumulated form the operationsof AI robotic platforms.[59] Moreover, the OS for MAP will stream-line the management of diverse experiments conducted by mul-tiple users, thereby enhancing the scalability of MAP in materialsdiscovery.[60]2.4. Academic and Industrial Research FrameworksBesides focused research projects there are numerous researchframeworks with a broader view on digital materials research.They aim to join efforts on common cross-domain requirementsand demand from industry (e.g., research infrastructure andstandards).MaterialDigital: This platform is an initiative funded by theGerman federal government to coordinate the activities of a pro-gram to develop and demonstrate tools to enable the seamlessdecentralized exchange of materials data between various stake-holders based on a semantic data exchange.[61] Academic and in-dustrial partners collaborate in the development of a platform fordecentralized data exchange that accommodates both the needsof industry for data provenance, as well as the established FAIRprinciples in academics. Besides data exchange, this also includesthe establishment of decentralized workflow environments fordata processing, modeling, and simulations.FAIRmat: The FAIRmat project[62] aims to create a federatedFAIR data infrastructure for materials data with a central hub, theFAIRmat Portal. FAIRmat aims to advance and develop metadataschemas and ontologies and enables the efficient exchange ofFAIR research data, ensuring that the FAIRmat data infrastruc-ture will advance basic science of condensed-matter and materi-als physics by supporting active researchers, and also be of greatvalue for engineering. It tries to convince scientists to also sharedata they consider useless for their present purpose-oriented re-search, and reaches out within and beyond the community pro-viding advice, training, and user support. FAIRmat is fundedFigure 11. Overview of the MARK project by KIST. The KIST team has recently developed an autonomous experimental platform system, featuring amobile robot and three fully automated modules: a batch-type nanoparticle synthesis module with UV–vis measurement, a washing process module,and electrochemical half-cell testing module (RDE). (reproduced with permission. Copyright 2024, KIST).Adv. Mater. 2024, 36, 2407791 2407791 (9 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deFigure 12. Digital services provided by Materials Data Platform (MDPF) in NIMS, Japan (reproduced with permission. Copyright 2024, NIMS).under the scheme of the German National Research Data Ini-tiative (NFDI).DIADEM: In 2022, the French government has funded €85M an 8-years project for accelerating materials discovery.[63]The goal is to set a nation-wide hub of platforms including high-throughput synthesis, shaping, and characterization of materialsand establishing strong links with a digital platform for materialsand processes design, databases, deep learning, and AI.PSDI: In 2021 the UK government agency Engineering andPhysical Sciences Research Council (EPSRC) funded the Phys-ical Sciences Data Infrastructure project (PSDI).[64] The aim ofPSDI is to enable researchers in the physical sciences to handledata more easily by connecting the different data infrastructuresthey use by improving finding, accessing and combining refer-ence quality data from different commercial and open sources,using AI to explore data and enabling sharing of data, software,and models according to FAIR principles.CAPeX: In 2023, five Danish foundations and the Ministryof Higher Education and Science funded a 45M€ and 13-year“Pioneer Center for Accelerating P2X Materials Discovery.”[65]CAPeX is hosted at the Technical University of Denmark (DTU)and Aalborg University and unites five Danish universities withinternational consortia like the Acceleration Consortium at theUniversity of Toronto, the SUNCAT Center at Stanford Univer-sity, and the European SUNERGY initiative at Utrecht Univer-sity in a trans-disciplinary research center focused on accelerat-ing the discovery of sustainable and scalable materials and elec-trocatalysts for Power-to-X, through the development of an AI-orchestrated Decentralized and AsynchroNous Materials Accel-eration Platform (DANMAP).NIMS-MPDF: The National Institute for Materials Science(NIMS) Japan has established the Materials Data Platform(MDPF) to provide digital services that accelerate materials re-search and development. As illustrated in Figure 12, MDPF’s ser-vices are comprised of materials data resources (Source), a ma-terials model development environment (Dev), and a materialsmodel operation environment (Ops).[66]The materials data resources include MatNavi, one of theworld’s largest materials databases. For example, AtomWork-Adv., the inorganic materials database, contains 379 736 existinginorganic crystal structures collected from academic papers alongwith phase diagrams and properties (as of March 2024).[67,68]PoLyInfo, a polymer database, includes data on 32 924 uniquepolymers, detailing their repeated structures, monomers, andproperties (as of March 2024).[69,70] In addition to the databasescurated from academic papers, NIMS publishes databases de-rived from standardized tests conducted by the institute overmany years, covering mechanical performances such as creep,fatigue, and corrosion.[71–73] Other noteworthy databases includethe Diffusion Database (Kakusan),[74] the Computational PhaseDiagram Database (CPDDB),[75] and the Thermophysical Prop-erty Database.[76] The Research Data Environment (RDE) is a sys-tem developed to digitize all research workflows and capture theresearch data produced. It was launched in January 2023.[77] Byusing templates composed of Python codes for the automatic ex-traction of measurement conditions from instrument data filesand customized forms for entering sample information, researchwork data can be structured and recorded in a reusable format.As of August 1, 2024, there are more than 1000 active templates,3447 users, and 1 158 615 data files.As for the Dev environment, a new AI platform called pinaxis under development and is expected to become operational bythe end of 2025. The Ops environment, known as MInt, is a sci-entific workflow system capable of linking processing, structure,property, and performance on a computer.[78–81] Numerous mod-ules (or models), primarily focused on structural materials, havealready been registered and are being utilized by an industry-academia consortium operated by NIMS. MInt is contributing tosolving the inverse problem of designing materials and processesfrom performance.[82]The MDPF, in collaboration with the Advanced Research In-frastructure for Materials and Nanotechnology (ARIM) and theData Creation and Utilization Type Material Research and De-velopment Project (DxMT), forms the core of the Materials DXPlatform initiatives funded by the Ministry of Education, Cul-ture, Sports, Science and Technology (MEXT), Japan. Further-more, the MDPF services are planned for use in various na-tional projects.Adv. Mater. 2024, 36, 2407791 2407791 (10 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deMGI: The U.S. Materials Genome Initiative (MGI), launchedin 2011, is a multi-agency effort aimed at accelerating the discov-ery, development, and deployment of advanced materials. Theinitiative seeks to transform the materials science field by inte-grating computational tools, experimental methods, and digitaldata management, thereby reducing the time and cost associatedwith material innovation. By fostering collaboration between gov-ernment, industry, and academia, MGI enables the rapid screen-ing of material properties and the optimization of material per-formance for various applications, including energy, transport,and manufacturing. Key federal agencies involved include theDepartment of Energy, the National Science Foundation, and theNational Institute of Standards and Technology (NIST).[19,83]Acceleration Consortium: The Canadian Acceleration Consor-tium is a initiative designed to speed up the discovery, develop-ment, and commercialization of advanced materials in Canada.Launched with a focus on using artificial intelligence (AI),robotics, and high-throughput experimentation, the consortiumaims to bridge the gap between academic research and industrialapplication. By fostering collaboration between universities, re-search institutions, and industry partners, the Acceleration Con-sortium supports the rapid prototyping and scaling of new mate-rials, particularly in areas like clean energy, pharmaceuticals, andadvanced manufacturing.[84]European Materials Modeling Council (EMMC) and EuropeanMaterials Characterization Council (EMCC): EMMC[85] was cre-ated in 2014 and aims to combine materials modeling and dig-italization to accelerate and maintain the development of novelmaterials. They recommend improving communication and col-laboration among stakeholders, identifying hurdles and provid-ing solutions, facilitating integrated modeling and digitalization,connecting academic research with industrial exploitation, sus-taining material modeling digitalization in Europe, informingindustrial awareness, and supporting the software industry. Animportant asset governed by the EMMC is the Elementary Mul-tiperspective Material Ontology (EMMO)[86] building the foun-dation of interoperable materials research. EMCC[87] was estab-lished in 2016 with objectives including involving stakeholders indeveloping characterization tools, identifying requirements, or-ganizing R&I initiatives, creating a nano-characterization frame-work, and linking nanometrology with industry. Both councilsplay crucial roles in advancing materials science and technol-ogy in Europe. Important results of the EMCC are efforts thestandardize characterization (CHADA) and model data (MODA)initiatives.[88]Energy Materials Industrial Research Initiative (EMIRI):EMIRI[89] is an organization that focuses on accelerating thedevelopment and deployment of advanced materials for cleanenergy technologies. EMIRI brings together industry, researchinstitutes, and other stakeholders to collaborate on innovative so-lutions for energy transition. Their primary goal is to support thecommercialization of energy materials by fostering research anddevelopment, knowledge sharing, and market uptake. EMIRIplays a vital role in advancing the transition toward a sustainableand low-carbon energy system by promoting the use of advancedmaterials in areas such as renewable energy, energy storage, andenergy efficiency. EMIRI provides strong support of the MAPapproach and regularly hosts workshops to bring the communitytogether.Figure 13. Layers of a MAP: Core components Hardware, Software, andData in the center surrounded by a community driving the MAPs withstrong embedding the scientific domains, industry applications, and pub-lic funding.3. Building BlocksHere we provide the technical context regarding the necessarybuilding blocks of MAPs to illustrate why they fall outside tra-ditional funding and governance schemas and require an inter-disciplinary, coordinated effort. As displayed in Figure 13, MAPsare more than just technology—they are community-drivenecosystems that combine multiple components in a workingsystem.In essence, MAPs are a new concept or meta-method thatwill fundamentally change how we conduct and record the pro-cess of materials research and transformation. They are tran-siting from traditional costly analog trial-and-error approachesto more agile digital research powered by automation and AI-enhanced decision-making. Being disruptive by nature, MAPsdo not exist in isolation. Instead, they reside in a researchcommunity with democratized access to lab-of-the-future in-frastructure and public bodies to govern FAIR and sustainablescience.MAPs come in different shapes and forms such as locally andfully integrated to run specific materials development programs,distributed between different facilities or even highly decentral-ized entities. The latter are cloud labs that provide democratizedaccess to MAPs and serve a broader academic and industrial com-munity. Regardless of their specific implementation, the samekey components are always involved:A MAP is an ecosystem that goes far beyond its technical com-ponents. Those technical components can be divided into twogroups, hardware and software, while the third non-technicalcomponent is the research community providing the MAP’s con-text and area of impact.The hardware components of a MAP provide a highly inte-grated automation of laboratory experiments. A suitable setup isdesigned and maintained by dedicated mechatronic teams. Thesetup consists of synthesis equipment to produce materials anddevices, characterization equipment to analyze the experimentalAdv. Mater. 2024, 36, 2407791 2407791 (11 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deoutput, and robots to automate and accelerate sample handlingbetween the two. Peripheral components include safety and re-source infrastructure.The software components are the backbone of the MAP thatenables its operation by interfacing instrumentation, controls,simulation and hardware, providing data storage and analyticaltools as well as orchestrating the components of the MAP. Im-portant tasks of the software include the data management andanalyses (involving the development of new analytical tools forautomated high-throughput analyses), and then AI-driven exper-imental planning, employing digital twins of R&D workflowsand experiments by multi-scale modeling (electronic, atomistic,mesoscopic, continuum, and macro-scale modeling).[90]Finally, the ecosystem of a MAP depends on a research com-munity that nourishes and guides the MAP, provides feedbackand leverages the knowledge generated. To date, the feedbackalone is not enough to operate a MAP. Even more, the input pro-vided by human creativity - which machines do not possess - isindispensable and creates the area of impact for a MAP in thefirst place.3.1. HardwareMaterials synthesis and analysis require a wide range of oftenhighly specialized laboratory equipment, encompassing process-ing tools like solid and liquid dispensers, mixers, coaters andovens, as well as characterization instruments such as microbal-ances and thermocouples to sophisticated spectrometers. Hard-ware requirements, particularly for synthesis, vary widely and canbe highly dependent on the scope of application, ranging from ad-ditive manufacturing solutions for structural and functional ma-terials to high-throughput bioreactors. This high level of special-ization of the equipment has led to a fragmented market, char-acterized by numerous small-scale specialized equipment man-ufacturers. Even within a modestly sized MAP, integration ofequipment from a variety of manufacturers is necessary to ad-dress the multifaceted processes inherent in typical materials sci-ence challenges. A common problem when integrating these sys-tems into a MAP is the fact that they are often designed as stan-dalone pieces of equipment meant to be operated by humans, thelack of comprehensive standards (unlike in industrial automa-tion) and the frequent lack of documentation for digital inter-faces that may exist. Consequently, the interconnection of thisequipment is commonly an initial obstacle in the establishmentof a MAP.In MAPs, automated laboratories are typically constituted ofdifferent modules or stations, each of them dedicated to a specifictask or subtask (e.g., materials synthesis, materials characteriza-tions, materials handling and materials processing). The diver-sity of synthesis routes, characterization techniques and materi-als processing approaches represent a large variety of technicalsolutions to be designed and implemented in MAPs. These taskscan be carried out in high-throughput or inline manner, in se-quence or in parallel, rather than in small batches in traditionallabs operated by humans. Though commercial solutions areavailable for high-throughput lab experimentations, but can bevery specifically designed, with limited flexibility for customiza-tions or upgrades as well as with limited options for a facile in-tegration in a larger interconnected system such as a MAP. Forthese reasons, MAPs often require custom design and in-housebuilt modules and stations, which enable greater flexibility in thedesign, integration and possible future upgrades. Robots and au-tomation are of particular importance as an integrative elementof a MAP in order to manipulate and transport samples and con-sumables required to carry out the experiments. In particular,multi-axis systems can flexibly transport materials and samplesbetween different stations/modules of the MAP or, conversely,guide tools and measuring instruments over the materials. Un-like humans, robots do not suffer from the monotonous repeti-tive activity and perform it with constant high speed and preci-sion. In addition to work ethics aspects, this also has a direct pos-itive influence on the reproducibility and data quality of the re-sults.However, the use of robots may require substantial initial in-vestments as well as specialized personnel for designing, main-taining and integrating automated laboratories, as the automa-tion of trivial manual motion sequences is often associated withconsiderable complexity. Automation of laboratory hardware isthus a main enabling technology for MAPs. In recent years thecost of automation has decreased significantly and the range ofsmall- to midscale robots which are available for the integrationin a laboratory environment has increased significantly.Nevertheless, automation in a laboratory environment is stilldifferent from automation in an industrial production environ-ment. In industry robots are made to repetitively perform a sin-gle task over and over. They must be robust as the failure of arobot can stop a production line. They also often need to oper-ate under the control of a factory worker, which requires veryrobust and easy-to-use human-machine interfaces. In industry,there are often highly trained teams of software engineers per-forming the system integration of industrial robots into large pro-duction lines, utilizing large-scale virtual environments or digi-tal twins.In research laboratory environments, the requirements dif-fer significantly. Robots intended for scientific applications mustbe seamlessly integrable, user-friendly, and readily upgradable.They should offer swift reconfigurability to adapt experimentsquickly and carry out complex designs of experiment. In theseenvironments, robots are often operated by highly trained scien-tists and technicians, possessing some programming proficiency.However, such profiles are not widespread, as programming re-mains a rare inclusion in educational training programs for ma-terials science and engineering. Consequently, these experts fre-quently need to change setups and reconfigure systems to inte-grate newly developed information. Indeed, MAPs evolve and im-prove constantly as technology advances and learning in the par-ticular area of application takes place. In addition, robotic equip-ment in MAPs will need to be interfaced with highly specializedanalytical equipment performing measurements and generatingthe data that needs to be integrated into an adaptive AI-driven op-timization workflow. As such MAPs are expected to be constantlyevolving in order to perform the required tasks better and betterover time.In MAPs, the integrated equipment platform provides an in-terface of the software and data infrastructure in the backgroundto the physical system. Robots and automation must handle andmanipulate materials, samples, and consumables. Each of theseAdv. Mater. 2024, 36, 2407791 2407791 (12 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.detasks is usually very application-specific and, therefore, must bereasonably easy to reconfigure. MAPs will also integrate stan-dard lab equipment designed for scientific tasks like scales, spin-coaters, or spectrometers. Their integration is often a particularchallenge and sometimes the robot in a MAP needs to preciselymimic a human operator to execute a task like loading and un-loading a piece of equipment.Safety requirements in laboratories are also quite differentfrom industry. Robots in MAPs are usually smaller and do notpose as much threat to the physical safety of the operator. There-fore, truly collaborative robots are not necessarily a requirement,although the collaborative feature may be of benefit for mak-ing the robots more flexible in application. On the other hand,other safety issues prevail in materials science laboratories. TheMAPs may handle hazardous substances, or materials that areinherently sensitive to oxygen or water. As such the requiredequipment will have to be designed to operate in protected at-mospheres and in glove boxes. They may be exposed to vaporsfrom solvents, other chemicals, high temperatures, intense light,or even radiation in operation. This poses new challenges, whichneed to be kept in mind when selecting or developing such equip-ment. But it should be remembered that this approach removesthe potential for human exposure to chemicals.Many of these challenges faced by hardware in MAPs arehighly domain-specific. In addition to that, it is difficult to foreseeall the future requirements in highly dynamic research areas. Toovercome this problem, MAPs generally need to be flexible andeasy to use, offering remote-controlled interfaces and be easily re-configurable, while remaining resilient to environmental factors.Through well-documented interfaces and user-friendly libraries,they should be programmable by scientists to further evolve intheir application. Additionally, they should support many pro-tocols and interfaces to efficiently deliver data and receive com-mands from the control software environment.3.2. Communication Protocols and InterfacesInterfaces connect the hardware components of a MAP to eachother and to the superordinate IT infrastructure. Protocols imple-mented on top of the interfaces enable communication betweenthe hardware controllers and to the common software overlay tocontrol and reconfigure laboratory instruments while perform-ing optimization loops.To make this possible, peripheral devices and laboratory equip-ment that can be remotely controlled are required. This is farfrom being an established standard as of today. Many lab devicescome with proprietary interfaces which are particularly difficultto automate and specialized software to control the device. Theseinterfaces are often undocumented or suppliers are unwilling toprovide the command sets required for automation. This will re-main a challenge in coming years, however, providers with a flex-ible approach may benefit considerably.To successfully foster the development of MAPs scientificequipment well documented APIs are needed. In general, Eth-ernet should be used as the interface and IP-based protocols asfar as possible. This constellation is extremely widely adoptedand represents the basis for the Internet of Things (IoT) with itsflexibility, expandability, throughput and stability. Where robust-ness/safety, real-time capability and industrial transfer is impor-tant, industrial ethernet/field buses like EtherCAT and ProfiNETshould be used.When integrating legacy systems, it may be necessary to wrapoutdated and proprietary protocols and interfaces into a stan-dardized and self-describing version before integrating them intomodern systems. OPC-UA,[91] an upcoming industry standard,is an excellent option for this purpose as it provides a semantic,self-standardized communication protocol which is supported bya growing number of device manufacturers.For simple setups, OpenAPI[92] conform HTTP REST[93] pro-tocols are another option. While not designed for secured high-performance asynchronous communications like OPC-UA theyare very easy to learn and provide a low-barrier entry.3.3. Software InfrastructureSoftware infrastructure refers to the underlying technical sys-tems that provide the development, deployment, and operationof software applications. These include hardware systems, net-work infrastructure, cloud platforms, databases, operating sys-tems, development tools, and other components that form thebackbone of a software system. The goal of software infrastruc-ture is to create a stable, secure and scalable environment for thedevelopment and deployment of software.Materials discovery programs are a rather complex interplay ofexperimental design, digital experiments, lab experiments, andinsights from existing data. This complexity poses multiple tech-nical design principles on the software architecture.1. Compatibility: Support multiple hardware systems, networkinfrastructure, cloud platforms, databases, operating systems,development tools, and other components required to runmaterials discovery programs.2. Agility: Provide the flexibility to add, substitute or change in-dividual components to address the fact that scientific work-flows are constantly adjusted and differ from program to pro-gram. Further, lab equipment, simulation, and AI technologyare continuously evolving. MAPs need to be able to adapt andadd functionality.3. Accessibility: Provide multi-user access via web interfaces to setup, run, and monitor materials discovery programs.4. Security: Besides data and software hosting in secure on-premises or cloud environments, security extends to safetystandards in running (automated) lab experiments.5. Operational Stability: It is necessary to go beyond just open-source software on GitHub and establish governing processesto maintain the software, release regular updates, ensure qual-ity standards, and invest in training and education.The above-listed criteria are best met by microservice-basedarchitectures that offer modern approaches for workflow or-chestration and automation of software infrastructures. Appli-cations and software modules are encapsulated in containerizedservices[94] as decomposable units to execute them independentlyof each other. In this way, not only data but also complex soft-ware configurations can be persisted, versioned and reproduced.This enables programmers to develop systems independently yetAdv. Mater. 2024, 36, 2407791 2407791 (13 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deensure interoperability between the components via suitable Ap-plication Programming Interfaces (APIs) like REST, defined dataschemas, and standards. The emergence of decoupled stream-ing architectures provides a natural backbone for microservicesas well as seamless connection of data producers, consumers,and processors. Streaming infrastructures provide decoupled au-tomation of data curation, reduction, and analysis; real-time ex-periment monitoring and control; and flexible deployment ofAI/ML to guide autonomous research.[95]To process computing workloads, particularly those relying onmassive quantities of data, high-performance computing (HPC)makes use of large-scale computing resources. Such clusters of-fer a way to manage the demanding computational requirementsfor technologies such as machine learning (ML), AI, and the In-ternet of Things (IoT). Common HPC applications typically in-volve simulation and emulation tasks for a wide range of engi-neering and scientific challenges.3.4. SoftwareR&D applications differ significantly in scope, such as the consid-ered materials classes and intended functionality. This diversityresults in specific implementations and uses of application soft-ware, distinguishing this layer from the backend software infras-tructure. Nonetheless, the fundamental concept of a MAP staysthe same, regardless of whether the scientific research pertainsto next-generation battery materials, polymers, or flexible elec-tronic devices. Nevertheless, the best combination of individualcomponents may vary depending on the particular application ofthe MAP.In general, on top of the software infrastructure, applicationsoftware modules of a MAP need to cover multiple aspects alongthe life cycle or continuum of experimental data, ranging fromabstract modeling to result sharing (c.f. Figure 14). Along this cy-cle, a larger community should be involved, at least through shar-ing of (intermediate) results through a larger knowledge graphbuilt on the FAIR principles or even across the cycling throughdistributed MAPs with shared infrastructure.Application software modules can be implemented as fullysupported commercial packages, open-source algorithms for pre-dicting material properties (such as physical, optical, electronic,and mechanical), or customized implementations that cater tothe specific needs of a material discovery program. The latterincludes digital twins for application testing or project-specificdata analytics and should support the MAP philosophy to movefrom a linear design-make-test approach to a closed feedbackloop with fast iteration cycles accelerated by AI and automation.At the end of each cycle, the generated data trains the predic-tive materials modeling modules allowing the AI to trigger amore informed subsequent cycle. In this way, materials discoveryprograms converge quickly towards the desired target propertyprofile.Each cycle comprises several stages, starting with identifyingthe necessary properties and defining the design space for thematerial. This is followed by predictive materials modeling, plan-ning, and executing experiments, capturing data, analyzing data,and sharing knowledge. The key software components involvedin this process are:Figure 14. Life Cycle of Experimental Data: Steps in the closed-loop mate-rials discovery.1. Planning and Optimization: In the R&D workflow, findingthe optimal design in a high dimensional design space is acommon challenge that occurs at multiple stages. This in-cludes identifying the optimal material composition, processparameters for synthesis and manufacturing, and or deviceengineering. Human experimental planning is performed bydefining a simple matrix of experiments intended to scan theparameter space. The experiments are executed according tothe plan and the limited results are often evaluated after all ex-periments have been completed. The MAP concept replacesthe static and linear experiment planning by an iterative andadaptive, AI-driven research process. The AI-driven researchplanner starts with a few data points to create an initial crudemodel. The model is used to propose the next generation ofexperiments, which are used to improve the model. This pro-cedure is iterated in order to explore the relevant regions of theparameter space and to find an optimum. This iterative pro-cess ensures that experimental resources are spent efficientlyand exploiting existing knowledge and data in the best waypossible. Algorithmic implementations rely on Bayesian opti-mization, active learning, reinforcement learning, and relatedmethods. This technique methodically explores and exploits,as appropriate, with fewer constraints than conventional ex-perimental methods. A major responsibility lies in the handAdv. Mater. 2024, 36, 2407791 2407791 (14 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deof scientists, who need to closely monitor the progress of theMAP to ensure that the model evolves in a physically soundway.2. Experimental Execution: Application software modules act asan interface and driver between the planning and optimiza-tion modules, the laboratory equipment and backend orches-tration system to automate the execution of experiments andresults acquisition. This ensures proper data handling, dataquality, lab safety, and secure connections to the hosting cloudor on-premises system within an Internet of Labs (IoL) con-cept. However, the scientists need to oversee the experimentalexecution with graphical user interface to prevent that the AI-driven research planning is driving the search into directionsinaccessible for the experimental setup.3. Predictive Material Modeling and Digital Twins: Lab experi-ments may be complemented by digital experiments, whichare executed virtually/in silico. Physics-based simulations cancontribute additional information or act as virtual twins todeliver properties with sufficient accuracy in less resource-consuming ways. This can take place at multiple levels; forexample, on the molecular level (chemical, ab-initio), on sin-gle parts (multi-physics), or for full systems (device model-ing). Simulations and inverse-design approaches can be usedfor virtual materials screening to identify the most promis-ing candidates. Data-driven ML models can be trained withphysics-based simulation data to provide surrogate models forfaster exploration or to bridging all length scales, from micro-scopic behavior to macroscopic impact on engineering. It isimportant that scientists critically evaluate the prediction ofdata-driven models to determine their accuracy, generalizabil-ity, and thus application range. Therefore, models should bebuilt as self-contained containers with proper documentationand associated runtime requirements. As for research data,metadata standards based specific ontologies[96,97] should alsoapply to those artefacts.4. Documentation: Data management and documentation of ex-perimental procedures are critical aspects in R&D labs. Elec-tronic lab notebooks (ELNs) and lab informatics systems(LIMS) are widely used by scientists to document their work.However, the data acquisition and storage need to be stan-dardized and put into an ontology to make it useful for anAI-enabled MAP. The challenge is consistently handling boththe raw data together with the context and metadata, which iscritical for ML across all methods used and partners partici-pating in the MAP. This is especially critical in an R&D en-vironment where experimental setups constantly adapt to thelatest insights. Frameworks like OpenSemanticLab[98] repre-sent first approaches for an integrated MAP ELN/LIMS pro-viding a holistic perspective including both human and ma-chine agents.5. Data Analytics and Insights: Data analytics is not a post-processing activity, as commonplace in conventional researchprojects, but a central component of a MAP. Data gener-ated in each generation must be analyzed and integrated intothe next generation of the model in a continuous manner toprogress the MAP-based research. Insight into the problemcan be gained both from the evolution of the experimentallybased model and from refined simulation models. Neverthe-less, vivid statistical visualizations must be provided to the hu-man operator to improve data understanding and to integrateabstraction capabilities.6. Data and Knowledge Sharing: Data, data sets and databasesare assets that may be used to answer various questions. Thedata set from a MAP-based research project may act as thestarting point for another MAP addressing another researchquestion. This requires that databases and software modulesuse standards for easy ingestion and queries of data reposito-ries and ontologies. Besides the software, governance is nec-essary to ensure data quality and support. Suitable technicalimplementations include data lakes and data warehouses withenterprise-level quality and intuitive access through graphi-cal interfaces. Application-dependent modules for knowledgesharing are tightly integrated within the orchestration sys-tem of the backend software infrastructure. In general, MAPsshould create linked and semantically enriched data alignedto common standards (JSON-LD[99]/RDF[100]) directly at thesource through documentation with general and domain-specific ontologies that allow FAIR data exchange via DataSpaces or Data Repositories without any further manualsteps.7. Intuitive software through GUIs: Democratizing access to aMAP requires guaranteeing that the R&D community canreadily interact with its software infrastructure. Thus, aMAP must provide intuitive graphical user interfaces (GUIs)that ensure the diverse range of skill sets of a multi-disciplinary R&D community, which may lack in knowl-edge in software development or scripting, are not a bur-den to its safe and efficient use. Thereby, GUIs can be man-aged and maintained as commercial packages or as freeand open-source software embedded in a community ofdevelopers.In recent years, significant progress has been made in ad-vancing each individual software component, and breakthroughtechnologies are expected to continue. For instance, neural net-works offer enormous potential to speed up computationally in-tensive materials simulation and enable more accurate simula-tions of larger systems. Probabilistic ML and Bayesian NeuralNetworks provide uncertainty estimates that allow researchersto evaluate the trustworthiness of predictions, which is par-ticularly important for exploring novel material ideas outsideof the training data. However, further research is needed tomake ML more reliable and interpretable and automate its in-tegration within the research cycle, running in parallel withphysical experiments. Another important point is to gain moreunderstanding of which components, such as AI-driven opti-mizer, experimental methods, and simulations, should be har-moniously combined to create a MAP tailored to address specificproblems.Due to safety and economic reasons, requirements of softwaredevelopment for large-scale software products directly transferto MAPs. The code steering a multi-million-dollar machine andevaluating the generated data needs to be of high quality, applyingstrict coding standards, and employing automated testing beforeroll-out to the physical system (“code in production” in terms ofsoftware development) in a mode known as Continuous Integra-tion/Continuous Delivery (CI/CD).Adv. Mater. 2024, 36, 2407791 2407791 (15 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.de3.5. CommunityOnly a strong and interdisciplinary community is able to buildand successfully use a MAP, and subsequently, MAPs offer a jointplatform for collaboration to solve social challenges.While operating traditional labs in smaller and often single-disciplinary teams has been a long-standing integral part ofchemistry and material science, MAPs represent a new conceptin academia and industrial research. Therefore, an interdisci-plinary ecosystem that creates the necessary skill set and support-ive environment to develop, maintain, and conduct research onMAPs becomes mandatory. Such a community needs to rest onthe following pillars:1. Governance: A community that provides all aspects to sus-tainably operate MAPs, including establishing standards, in-tegrating new modules, managing automated labs, and ensur-ing quality control for closed and open-source hardware andsoftware (c.p. Free/Libre Open Source Software, FLOSS).2. Training and Education: MAPs represent a fundamentally newapproach to R&D and are more than just a replacement fortraditional tools. As such, the research community needs toget trained on how to adapt R&D programs to take advantageof MAPs while also providing education for developers in theMAP community to ensure the high quality, scalability, andreusability of new software and automation modules.3. Create Awareness: Dissemination of knowledge and successstories is essential to generate awareness and momentum forwidespread technology adoption.4. Technology Transfer: Establishing an ecosystem encompassingstartups and corporations is crucial for the smooth transitionof technology into the industrial sector for broader adoption.This is when the Twin Digital & Green Transition will materi-alize, leading to a significant economic and societal impact.Such ecosystems are routed in an organizational structure thatcoordinates the activities around each MAP. Depending on thesize and scope of the project, governing entities are establishedat the project level, at the scientific level, or as part of internationalgovernment-led initiatives.In recent years, communities in each category have beenformed (see also Section 2.4). For instance, BIG-MAP (BatteryInterface Genome – Materials Acceleration Platform) is part ofthe Battery2030+ initiative, which includes various renowned ex-perts and institutes, and maintains close connections with a di-verse set of key industrial stakeholders in battery research andproduction. Meanwhile, the French initiative DIADEM aims tocreate synergies between material sciences and data sciences.The UK initiative, PSDI (Physical Sciences Data Infrastructure),focuses on integrating experiments and material simulation toestablish a data-centric framework for materials. The German-Canadian Materials Acceleration Center (GC-MAC)[101] is a bina-tional approach to aligning methodologies. The European Materi-als Acceleration Center for Energy (EU-MACE)[102] aims to builda community of MAPs across Europe. Complementing this, theEuropean Materials Informatics Network (EuMINe)[103] focuseson method development and application in the area of materialsinformatics. On the international level, the Materials for EnergyInnovation (M4E) initiative is arguably the most prominent com-munity dedicated to providing education, shared infrastructure,and strong networks around MAPs. This initiative is part of Mis-sion Innovation (MI), a global effort to accelerate progress towardthe Paris Agreement goals and pathways to net zero.To build on the early successes and encourage wider adoption,we need to continue strengthening connections among stake-holders from academia, industry, and the public sector. Ecosys-tems that focus on professional training and serve as breedinggrounds for startups and technology pioneers are necessary. Thelogical next step for MAPs is to bring the new concepts into in-dustrial R&D operations and thus increase the velocity and agilityof materials innovation.4. ImpactCurrent societal challenges (see also Section 1.1) imply an in-creasing complexity of AdMats innovation, while at the sametime require drastically accelerating the R&D & innovation pro-cess. New regulatory frameworks (including initiatives like theGreen Deal) could present a risk to industries ability to meet mar-ket demands due to increased development time and costs. Tra-ditional R&D approaches are often slow, expensive, and labor-intensive. However, with the proliferation of computing re-sources and advancements in data analytics, AI, and lab automa-tion, we have the opportunity to revolutionize materials R&D.MAPs are expected to considerably reduce resources neededfor materials development as compared to traditional R&D ap-proaches by: 1) optimizing the number of experiments throughpredictive screenings to identify the most promising materials, 2)significantly reducing the materials resources needed for individ-ual experimental trials, 3) automating repetitive and low-value-added tasks to enhance experimental reproducibility and liberat-ing researchers for more impactful work, and 4) accelerating theoverall process, directly reducing costs. Hence, MAPs will resultin managing and exploiting knowledge, data, and resources in amore efficient way and a society that succeeds first in fully imple-menting the vision of a MAP will be able to stand out clearly ininternational competition.With all the tools available to de-risk and accelerate disruptivematerials innovation, we propose the following threefold strate-gic goals to be achieved with MAPs:1. Synchronize material innovation and end-product development:Today’s long cycle time for novel materials does not match thecycle time of engineering. By accelerating materials discovery,MAPs enable the industry to respond with greater agility tothe application requirements of multi-material devices.2. Enable innovation that maximizes the use of existing assets: To-day, early-stage discovery focuses primarily on getting mate-rial properties right. Techno-economic and production con-siderations are taken into account only at later stages. By fo-cusing on manufacturability instead of just synthesizability,MAPs will reduce the capex, time, and risk of commercializa-tion of new materials and devices by enabling simultaneousdevelopment and verification of novel process or manufactur-ing methods.3. Turn open innovation into societal impact: Today, material in-novation is at the beginning of the value chain and too of-ten remains isolated from relevant downstream stakeholdersAdv. Mater. 2024, 36, 2407791 2407791 (16 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.delimiting its actual implementation in the real world. By serv-ing as an ecosystem, MAPs will bring together technical, eco-nomic, regulatory, social, and policy as a community to enablethe successful rollout of disruptive and sustainable innovationin materials.The aforementioned aspects have immediate implications forgovernments and their strategies:1. Reduced dependency on raw materials: Better utilization ofmaterials, e.g., by reducing the use of rare metals by 20%,would have a significant impact on raw materials sourcing forconsumer electronics, electric vehicles, solar cells, etc.2. Technology Sovereignty: Winning the race of artificial intelli-gence and accelerated materials innovation is crucial to sus-taining or expanding economic leadership positions.3. The Twin Transition (EU Term for green and digital transi-tion): Materials acceleration platforms are a blueprint for howfast transitioning to AI, modern digital R&D process and ma-terials innovation can fuel each other to reach NetZero, thussynergistically combining digital transformation and sustain-ability.4.1. Technological ScopeSynthesis: High-throughput synthesis of materials requires asmall amount of matter at each processing step when screen-ing complex compositional and structural phase spaces. Thinfilms with in-plane composition or structural gradients are a wayto achieve this requirement. Micro or mili-fluidic tools are an-other way to achieve the same goal for powders. Pixelized anneal-ing/sintering of 2D or 3D materials may provide libraries of dif-ferent structures starting from a monolithic bulk sample. Struc-tural, morphological and functional characterization are then tobe involved for the accelerated processing loop to be efficient.Catalysis: Catalysis is involved in the manufacturing of themajority of our products.[104] By lowering the activation barrier, itcan produce new materials at lower energy costs and/or generateless wasteful byproducts. MAPs can target better catalytic mate-rials and processes, which will lower energy consumption andcan lead to better-performing materials. First impact on catalystdiscovery have already been demonstrated.[105]Biotechnology: Synthetic biology provides the industry withthe ability to develop products and processes that are sustain-able in terms of our use of natural resources. Integration of bi-ology, digitalization and automation for the respective technolo-gies to become mainstream and adoptable as an enabling fieldof technology.[106] MAPs are expected to increase the speed andefficiency of strain engineering 20-fold to speed up biotechnolog-ical applications largely due to enabling high throughput whileminimizing trial-and-error. Current applications exist in chemi-cals and pharmaceuticals, while the scope of applications essen-tially covers the structural and functional field as a whole and israpidly expanding toward topics necessitating sustainable mate-rials and solutions.Bio-Sourced Polymers Hybrids and Composites: Molecularchemistry is already using high-throughput platforms to producebuilding blocks for polymers materials.[143,144] The next step us-ing MAPs will be to achieve polymers of advanced functionalitieseither intrinsically or through the inclusion of inorganic parti-cles. Additive manufacturing is a route toward accurate controlof the microstructure in such composites. MAPS are of interestfor the acceleration of monomer discovery (including bio-sourcedones) and for the full automatization of the processing of hybrids.Designing process twins, manufacturing optimization and func-tional testing all require AI-based tools. Such advanced polymerscan also be produced in view of easier recycling at end of life.Health, Pharma and Bio-Medicine: Drug discovery and de-velopment is thus a long, costly, and high-risk process with ex-cessive failure rates (>90%), especially in the final stages. Theadoption of MAPs in the pharmaceutical industry can offer hugeproductivity gains by completing the already well establishedhigh-throughput approaches with AI-based data analytics andexperimental planning. MAPs result in higher throughput andgreater efficiency, while maintaining strict performance criteria,enabling multisite and cross-department collaboration throughaccess to real-time data.Energy Materials: The demand for sustainable energy produc-tion and storage solutions is exponentially increasing. Energy hasbeen identified as a strategic sector by the EU.[107] Developingnovel, innovative solutions for sustainable energy production andstorage requires developing new materials. Facing the currentchallenges requires that research and development of new en-ergy materials should be drastically accelerated. Several regional,national and international initiatives have been initiated to createMAPs dedicated to the development of new materials for batter-ies, photovoltaics, and more.Photovoltaics: Photovoltaic technologies play a crucial role inadvancing renewable energy technologies and addressing cli-mate change. Acceleration in this field is necessary to furtherimprove the efficiency and cost-effectiveness of solar panels, de-velop new materials with enhanced properties, and optimizethe integration of photovoltaic devices. By applying MAPs ap-proaches, especially multi-target optimizations like high effi-ciency in combination with flexibility or absence of critical rawmaterials, can be addressed.Metamaterials: Metamaterials derive their properties fromtheir artificially designed structure rather than from their chem-ical composition and naturally occurring atomistic structures.Because of the large investigation space which comprises alsomesoscopic and macroscopic structuring in addition to chemicaland atomistic structure, it is necessary to accelerate the designand fabrication of metamaterials through MAPs. MAPs can notonly speed-up the discovery but also open up new fields for theuse of metamaterials, where their properties change during op-erational conditions. Possible applications are electrochemistry,membranes, catalysis, and in addition more conventional use-cases such as structural, acoustic and photonic materials.High Entropy Alloys and Oxides: In some cases, like corro-sion resistance or substitution of critical raw metals, high en-tropy alloys resulting from the mixing of more than five in-dividual metals are among the solutions. Screening the wholephase diagram in such complex cases is out of reach usingthe standard approach. Automatized thermodynamic modeling(e.g., CALPHAD), robotized casting, additive manufacturing andhigh-throughput structural and functional probing are then used.This is to be extended to non-metallic inorganic materials inwhich more than four chemical elements are mixed over the fullAdv. Mater. 2024, 36, 2407791 2407791 (17 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.decompositional space. MAPs would be a way to bridge the gapbetween different material classes opening the way to advancedfunctionalities (electromechanical, optoelectronic, multiferroic,etc.)Materials for Additive Manufacturing and the Relation to DigitalTwins: Additive manufacturing largely emerged in the process-ing of powder metal parts with geometries impossible to fabricateby conventional manufacturing methods. It is now extending todiverse materials classes: inorganic glasses, polymers, hybridsmade of inorganic inclusions in a polymer matrix. Novel mate-rials require additional control of thermodynamical and thermo-chemical parameters. Such additional parameters have to be in-cluded in the digital twins of the targeted materials and of themanufacturing processes. Microstructure modeling, in situ mon-itoring of the processing parameters and post-processing func-tional checks will produce a very large amount of data, the han-dling of which will require MAPs tools. Beyond the real spacecontrol of the 3D architectures, additive manufacturing can adda fourth dimension which is the well anticipated reaction to agiven stress (mechanical, electrical, magnetic, optical). Scanningsuch 4D spaces makes the use of MAPs even more relevant.Per- and Polyfluoroalkyl Substances (PFAS): Due to health andenvironmental implications, the use of per- and polyfluoroalkylsubstances (PFAS) in various industry sectors like aerospace, au-tomotive and consumer products is subject to regulations thataim to limit or prohibit its use. This confronts affected companieswith the challenge of quickly finding alternatives for the compre-hensive property profile (thermal and chemical resistance, lubric-ity and processability) of PFAS. Of particular significance is therelevance to the battery and H2/CO2 electrolyzers using PFASmembranes as well as consumer goods such as cooking utensils.MAPs can provide decisive support here by optimizing suitablereplacement candidates in the relevant property profiles and test-ing of results directly at the device level.Recycling Processes (Polymers, Precious Metals): In order topromote a circular economy where materials are efficientlyreused, thereby reducing the reliance on virgin resources andminimizing waste generation, MAPs can support a process forboth maximizing recycled material use while keeping the re-quired product properties. MAPs can further target the recycling(detection, sorting, extraction, and refining) process itself to toler-ate a broader input material stream and to provide higher gradeson the output side.4.2. Use of MAPs in IndustryThe 150-year-old materials industry faces radical upheavals frommultiple directions. There is an increasing urge for breakthroughinnovation to solve sustainability needs. Faster product develop-ment is needed to respond to ever-accelerating market trends.Supply chain challenges and scientific complexity are adding ad-ditional pressure on the industry.All taken together, the margins in the industry aredecreasing[108,109] due to increased supply chain and productioncosts which diminishes the resources available for research anddevelopment. When the industry favors incremental develop-ment of existing products with a predictable outcome ratherthan development of next-generation products that require inno-vative and disruptive technologies, technological and economicprogress are stifled.The introduction of MAPs in industrial research and develop-ment is a very promising strategic response to promote and ac-celerate innovation. Almost all major global players have alreadydeveloped AI strategies to increase their organizational agility, sothey should also consider to make use of the full MAP approach.The major impact of introducing MAPs for industrial researchlies in the reduction of resource consumption, shortening of thedevelopment cycles and cost, potentially more optimized materi-als for the target application and most importantly a more pre-dictable outcome of the research process. The reduction in re-source consumption is achieved by using the adaptive design ofexperiments, which systematically use the most relevant pointsin a large parameter space. This reduces the number of experi-ments and the amount of materials used. This also reduces thetime needed for the research campaign, which shortens the timeto market. The systematic and adaptive search of the parameterspace also has the advantage to be more likely to really find theoptimal material combination in the search space, which is supe-rior. This makes the MAP approach more predictable and reliableto find a solution to the problem compared to the traditional Edis-onian trial-and-error approach. Another advantage of the MAPapproach lies in the fact that the AI model can be directly usedor easily converted into a digital twin of the product. This digitaltwin can be used both in the upscaling toward production and forthe tracking of the product during its lifetime. Having a recordof the history of the product from production and usage is sim-plifying the recycling of the product. The lifetime information ofthe digital twin is invaluable in ultimately being used in the de-velopment of the next generation of products.This makes the MAP approach very appealing to the industryand they have been quick to embrace this methodology. Collabo-ration across companies may be less attractive due to IP issues,but large corporations have R&D centers distributed across theglobe. Connecting these R&D centers into a common, internalR&D MAP is providing a significant opportunity. The companiesparticipating in the Workshop “Re-inventing Materials ResearchWorkshop” organized by EMIRI[110] and the Accelerate Confer-ence series[111] presented various versions of how they are usingMAPs in their research and development. These insights and co-operation announcements[112] in the field indicate that the MAPconcept is neither science fiction nor science only, and it has al-ready gained attention and traction in commercial applications.Multiple proof points have demonstrated that digital tech-nology, artificial intelligence, and closed-loop self-learning sys-tems will fundamentally change chemical R&D. For instance, re-cently published success stories claim a reduction of experimen-tal workloads in general by up to 90%[113] and an acceleration ofthe development of new rare earth-free permanent magnets by afactor of 200[114] The natural next step is to support the industryto move toward larger adoption to make the concept of MAPs thenew standard in a modernized R&D process.The MAP concept may also open up space for new businessmodels. The main application in industry will be the internalMAP, but there are also possibilities to extend the offering of re-search services between companies and universities. Outsourc-ing research to Contract Research Organizations (CROs) hasbeen done for a long time, in particular in the pharmaceuticalAdv. Mater. 2024, 36, 2407791 2407791 (18 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.deindustry, but it is much more complicated to establish a con-nection between universities, research institutes and companies.There are usually hurdles to be overcome in terms of how the in-formation exchange between the service provider and the buyeris done and the IP rights. Making the lab “MAP ready,” establish-ing the MAP communication protocols and establishing a marketplace for research services will decrease this barrier to entry intothe services market and helps to align to common initiatives likeDataspaces and Marketplaces.Besides the chemical industry,[32,115,116] the most prominentexample for which we see signals for an emerging adoption isthe pharmaceutical industry[117] where drug discovery is findingsolutions amid strict requirements, harsh regulations, and highcosts.[118] Cumulative investments in AI companies for drug de-velopment surpassed $60B at the beginning of 2023.[119]The benefits of MAPs in the industry can be summarized as:1. R&D resources are used more effectively due to accelerateddevelopmental cycles (from set-up to development and pro-duction).2. More predictable outcome of research projects.3. Higher Internal Rate of Return (IRR) of innovation.4. Creation of virtual twins of the products.5. Products are simultaneously developed to optimize capital ex-penditure (CAPEX) and operational expenditure (OPEX) inmanufacturing.6. New business models, e.g. (cloud) lab as a service.7. Shorter time to market including production and testing leadtimes.5. GapsWhile initial implementations of MAPs clearly demonstratetheir potential, they also reveal the hurdles that still preventwidespread establishment across the research community andthe steps that need to be taken to overcome them, both on the po-litical/organizational and technical level. In general, challengesare beyond the individual researcher, individual lab, individualcountry and need coordinated regional, national and interna-tional research strategy.Despite the first successes of such platforms, there are stillgaps in the realization of their utility and maturity for accelera-tion and efficiency gains in the process from materials discoveryto device integration.[105,120] A major bottleneck is still their lim-itation in technological scope. There is limited interconnectivitywith modeling, computational, or experimental modules beyondthose readily integrated into the “closed-loop” hardware platform.Therefore, the “closed-loop” future labs concept has found mod-est traction in the field of electrochemical energy conversionand storage, encompassing hydrogen fuel cells and electrolysiscells.5.1. Compatible Infrastructure, Data Structures and Data SharingWhile the “open” idea is already widely established in software,this is not the case to the same extent in hardware. However,OpenHardware[121] is a prerequisite for assembling, optimiz-ing and maintaining complex MAPs from different hardwaremodules. This includes not only construction and circuit di-agrams but also device drivers and communication protocols.With OPC-UA LADS,[122] the OPC foundation has been laidfor at least the latter, which must now be implemented in thefield.Nevertheless, a robust cloud-native software infrastructure in-cluding cloud management and CI/CD pipelines to deploy appsand models must be set up on the basis of an open hardwareplatform to enable secure access to the system components anddata. Up to now, this has mainly been implemented by theneed for local access to the relevant terminals or, more indi-rectly, through shared data repositories. However, in order torealize shared and distributed MAPs, it is essential to also im-plement web-based access which, depending on the user role,enables, for example, the sending of synthesis requests, theviewing of log data and the retrieval of analysis results, forexample.Within this scope, data sharing has its own shortcomings andarising opportunities. Storage and retrieval of the integrated datafrom various sources, enable an effective development of self-serve online analytical tools for automated data analysis, im-proving the ability to connect to the analytical tools and the re-sponsiveness to semantic and integrated queries, and the dataaccess performance, all features that lie beyond the capabil-ities of current disjoint, heterogenous and often transaction-oriented databases and data infrastructures in the materialsdomain. Robust data governance and data management pro-tocols in an interoperable ecosystem - including the manage-ment of data quality, data lineage and tracking, and data se-curity - involve data standardization and metadata manage-ment at its core to provide context to the data and thus en-able materials scientists to locate the data they need mostefficiently.To accelerate research, production and innovation, we there-fore need to semanticise these domains (Industry Commons[123])with common data schemas and vocabulary (ontologies) and cre-ate documentation and metadata standards in the huge effort ofknowledge digitalization. The semantization and standardizationwill offer a pathway to interoperability across domains, while doc-umenting and standardizing will offer a pathway to validationacross complex workflows from raw materials to advanced ma-terials solutions.While general ontologies for the material domain (e.g.,EMMO,[86] MSEO[124] and the PMD ontologies[125]) as fundamen-tal basis for FAIR data management are established, gaps stillexist in standardization and development of domain-specific on-tologies, guiding meta-data management and harmonizing ap-proaches across different self-driving labs. Applying FAIR guide-lines to material R&D and manufacturing[126] is challenging dueto the complex and highly dimensional parameter space andthe lack of standardization in fabrication and characterizationtechniques.[127] To enforce FAIRness, ontologies provide seman-tic context to data and make the interpretations unambiguous butit is essential to build in intuitive software tools that are useable byany researcher. This will, in addition to ontologies, require anno-tated data schemas[128] to provide dedicated shapes for the com-mon patterns in research data. Finally, this will enable not onlysharing data, but generating knowledge within and across the do-main communities.Adv. Mater. 2024, 36, 2407791 2407791 (19 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.de5.2. International, Decentralized and Asynchronous MAPStructuresGiven the scale and complexity of the challenges ahead, we can-not achieve these goals in isolation. International partnershipsand collaborations are essential to accelerating materials discov-ery and promoting economic prosperity. By joining forces withlike-minded partners around the globe, we can strengthen ourresearch capabilities, share expertise and resources, and achieveour common goals more effectively. Therefore, we must prior-itize building strong, collaborative relationships across bordersand work together toward a more sustainable and prosperous fu-ture for all.There is a great need to orchestrate research campaigns be-yond a single lab, research organization and across the globewith dozens of partners complementing and replicating capa-bilities to build truly global platforms. However, there are ad-ditional challenges to create a distributed and asynchronousMAP. The requirements for standardization of communicationprotocols and ontologies become crucial for distributed MAPs,which include multiple labs and may stretch across borders andtime zones. The data and methodologies from different part-ners need to be integrated together. The standardization initia-tives mentioned above are vital and the next step is to implementthese standards into broadly accepted orchestrator and tenantsoftware.5.2.1. Standardized Software for Distributed MAPsThe research group building the MAP will need to set up anorchestrator server as described in Section 3.3. Writing an or-chestrator software from scratch is a large task, in particular if itneeds to handle multiple tenants, integrate various types of datafrom different methods and adhere to defined ontologies. An ad-ditional challenge for the orchestrator for a distributed MAP is tohandle the waiting time for tenants operated in the asynchronousmode. Creating such an orchestrator for a distributed MAP isa significant hurdle to set up the MAP. While there are manygeneric workflow environments and orchestrators[129] there are afew attempts to create or adapt such orchestrator software specif-ically for MAPs. On the pure data side, the globus project[130]is pushing to enable semantic data transfer at great speedswhereas software like FINALES,[131,132] and OpenSemantiLab-MAP[133] which were developed and demonstrated withingthe BIG-MAP project, emphasizes flexibility and ontologylinkage.There are similar requirements for standardized softwareagents, which can easily connect lab equipment or simula-tion software to the MAP. These agents should be able to re-ceive requests for measurements, to translate the meta datainto parameters for the setup of the experiment, execute mea-surements, gather data and finally report result data backto the orchestrator according to the communication proto-col and ontology of the MAP. Open-source software is flex-ible and may easily be adopted to the specific problem tobe solved by the MAP. Commercial software solutions mayprovide more reliability, which is required in industrial MAPapplications.5.2.2. Make existing labs “MAP ready”Integrating standardized software agents in existing labs to con-nect to the local infrastructure and automate the task executionwill decrease the effort to connect the lab to a MAP. The con-nection to a MAP would then be limited to configuring the ten-ant software according to the problem addressed by the MAP.It would in particular make sense to install agent software atlarge-scale facilities to make them “MAP ready” as they may servemany MAP initiatives.5.3. Education and TrainingThe MAP approach, which is still evolving, is multidisciplinaryand collaborative. A full understanding of its impact by the com-munity is yet to come. As such, increased awareness, trainingand implementation is needed to build a critical mass to achieveacceleration toward new materials and devices. It integrates mul-tiple domains together with their knowledge, tools and workflowsto a holistic system.As a cross domain approach, MAPs require a unique skillsetnot present in conventional research institutions. This includesIT experts in the area of core software and engineers in the areaof automation, but specific domain knowledge is also required inalmost all places. Therefore, also due to the tense labor market sit-uation, a continuous (re)training of domain researchers towarddigitization and automation competencies is necessary.By implementing modular MAPs with encapsulated build-ing blocks, the need to understand the overall system can beeliminated and specialization can take place instead. Best prac-tices can be adopted from bioinformatics, where such a trans-formation has already taken place. Possible methodologies in-clude an international Open MAP Academy construct, support-ing the development of trans-disciplinary competencies, e.g. ma-chine learning, autonomous robotics, orchestration, FAIR dataand the continuous improvement/evolution to build the bridgesbetween domains of Materials Science, Informatics and Process-ing/engineering. Other possibilities include the establishmentof international postgraduate degree programmes (MSci, bina-tional/cotutelle PhDs) or an international school, dedicated tostudents and experienced scientists including lectures and sev-eral weeks long stay at MAP platforms for training. A model forthat could be the successful HERCULES school[134] dedicated totraining at large-scale neutrons and synchrotron facilities and theSummer School for MAPs for energy materials[135] hosted by theGerman-Canadian Materials Acceleration Center (GC-MAC).[101]5.4. IP, OwnershipSimilar to large software projects and AI-generated results, the is-sue of intellectual property is complex for MAPs. Assuming thefinal result is an AI model trained with the data generated by var-ious actors at a MAP, its intellectual ownership cannot be deter-mined in a lump sum. Not only the model developer, but also thesystem developer, the operator and the data generators are candi-dates for (partial) ownership. The research questions posed to aMAP and its associated community may also be subject to confi-dentiality, especially if they originate in industry. In addition, theAdv. Mater. 2024, 36, 2407791 2407791 (20 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.decross-border movement of data, information and goods is the rulein distributed, multinational MAPs, requiring compliance with awide range of export control regulations.MAPs would therefore need to be established not only as phys-ical facilities and software services, but also as a legal frameworkdefining access and usage rights. This legal framework shouldprecisely define the conditions under which the MAP can be usedand who is entitled to the rights to the result. Resulting open datamust be free to use, but this does not mean that it must be freeto access since there are costs to creating, maintaining the neces-sary infrastructure. A hybrid model is conceivable, as is alreadyestablished for many software systems: If the user agrees to thedisclosure (e.g., CC BY) or only partial ownership of his results,he can benefit from favorable conditions that are possibly com-pensated by public funding. If the user wants to place the resultsunder a proprietary license, this is also possible, although with-out preferential conditions (e.g., higher price). This provides theopportunity to link MAP with the digital contract and data man-agement of Data Space initiatives (e.g., IDS[136]) and offer bothopen and proprietary data in common marketplace, as well asthe experimental and software service of MAPs themselves. MAPproviders that can guarantee confidential handling of sensitivedata, for example through a sealed-off environment, can in turnbe attractive service providers for industry.5.5. Research StrategyCurrent Funding: The current funding landscape in materi-als science focuses primarily on specific materials (such as hy-drogen) or industrial applications. The methodology and specif-ically the meta-method of the research infrastructure (hardwareand software), is insufficiently taken into account making sus-tainable investments in MAPs that can be applied to wholeclasses of materials difficult. With a few exceptions (BIG-MAP)no specific funding/call for funding to build full MAPs areavailable, therefore, MAPs are usually built from parts of bud-gets of a project focused on a particular application or AdvMat.In addition, there’s no funding to maintain a MAP (pay dedi-cated engineers, technicians) beyond the duration of the creatingproject.Necessary Funding: Funding programs are needed that explic-itly address the meta-method level in a pan-European scope. Em-phasis should be placed on FAIR and, where practical, OPEN re-search results that provide the greatest openness of the researchresults of public funding, not only with regard to OpenData andOpenSoftware, but also to OpenHardware in order to maximizethe flexibility of the emerging MAP and the innovation speed ofits components. In parallel, support should be provided to com-mercial enterprises, especially startups, that develop innovativebuilding blocks for MAPs in compliance with open standards.In this regard MAPs are shared infrastructure with contribu-tions from various members of a community. Established exam-ples of such an infrastructure are synchrotrons like CERN whereinvestments are in the order of billions of Euros. Their main pur-pose is to understand the structure of materials, which could beimportant to improve materials modeling. Building MAPs ac-counts for only a few percent of synchrotron investment, but theyare accelerating the discovery of many new, more powerful ma-terials by understanding the behavior of materials in a device ora system resulting from complex processing. MAPs and their AIsystem rely on good materials modeling, so the synchrotron andMAP system complement each other.AI starts with quality data, thus, a major initial effort needsto focus on AI-aware (data) governance measures. This meansimproving secured cloud AI/data infrastructures, new harmo-nization activities, including ontologized data infrastructuresand common data management frameworks. The inclusion ofpromising MAPs and self-driving labs in general needs to becontinuously supported by high-quality, AI-ready data from con-ventional labs. Connection to conventional labs could guaran-tee the long-term availability of MAPs and complement conven-tional labs, with decentralized data hubs aligned with nationaland European activities such as PMD,[61] NFDI,[62] DIADEM,[63]EUDAT[145] and EOSC.[139]European and National Funding Strategies: Funding shouldbe carried out strategically and take place at different levels.The first goal should be to create a long-term European ecosys-tem of RTOs, universities and industry that supports the generaldevelopment, takes care of standardization issues, and exchangesinformation. This ecosystem should also organize the differentresearch infrastructures and take care of training. Projects shouldbe announced for lower technology readiness level (TRL) to im-prove the infrastructures and to develop new standards, also in-cluding software and ontologized data infrastructure. Support-ing actions would be making MAP building part of postdoc-toral assistants/young excellent scientists funding programs likeERC[137] grants would shape a new generation of experts dedi-cated to the challenges in this field. Pushing promising MAPsto the European Strategic Forum on Research Infrastructure(ESFRI[138,139]) Roadmap could ensure their long-term availabilityand complement to conventional labs that should become MAP-ready and virtually connected to a decentralized MAP.Other projects with higher TRL should have the goal of inte-grating new materials into applications as quickly as possible, inparticular sectors where new materials are urgently needed (e.g.,hydrogen catalysis, batteries, and more, c.f. Section 4.1) to ful-fill the second goal of the funding: The transfer of knowledge toindustry and the use of MAPs by industry. To achieve this, weneed infrastructures focused on specific application areas, datamanagement systems, and we need to train enough experts inmodeling, robotics, materials science, and industrial automationto accelerate the innovation chain.Efforts in the EU and other regions should be flanked by anacceleration consortium similar to the already established Cana-dian initiative[84] receiving long-term funding from the CanadaFirst Research Excellence Fund (CFREF). Those consortia shouldaim to link projects that want to make use of the MAP method-ology, with national initiatives, academia, industry and politicalstakeholders. A European acceleration consortium could take theSET Plan[107] and its implementation groups as a blueprint andbe composed of experts from academia (RTOs and universities),industry, member states, and European Commission represen-tatives supporting material research projects, e.g., shaped by theupcoming Innovative Advanced Materials for Europe (IAM4EU)Partnership,[140] with a strong connection to the TechnologyCouncil for advanced materials that the European CommissionAdv. Mater. 2024, 36, 2407791 2407791 (21 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dewww.advancedsciencenews.com www.advmat.dewill set up in the frame of the policy communication on AdvancedMaterials for Industrial Leadership.[141,142]At the outset, funding schemas, data and AI infrastructuresneed to be developed and established with clear differentiations,cohesiveness, and complementarity to other similar activities inCanada, the US and the UK. This ecosystem should develop andorganize access to the various research infrastructures and en-sure continuous expansion.6. ConclusionThe advanced materials sector, particularly in the field of emerg-ing energy technologies, is facing a major challenge: the paceof development is trailing behind commercialization targetsand urgent societal needs. The main cause of this situation isthat stakeholders with complementary capabilities still operatelargely in isolation, with a lack of coordination between theirefforts. This situation has given rise to a new breed of futurelabs: material-acceleration platforms (MAPs) that combine arti-ficial intelligence (AI) with automated experimental hardware todrive the autonomous discovery and development of new ma-terials. Worldwide, a number of multidisciplinary MAP initia-tives involving governments, academia and industry have re-cently emerged.MAPs have emerged as indispensable tools for expediting ma-terials and manufacturing process development. By leveragingadvanced technologies and automation, MAPs offer immensepotential to revolutionize various industries. One of their mostsignificant advantages is the reduction of research and develop-ment (R&D) time, enabling faster innovation and quicker time-to-market for new materials.An important consideration in the MAP landscape is the dis-tinction between open and closed MAPs, the first being focussedon shared access, the latter on protected intellectual propertywith the primary aim of financial profit. With firstly MAPs ofboth kinds growing, we are at a pivotal moment that will shapethe future of MAPs. Open MAPs promote collaboration, knowl-edge sharing, and collective growth, fostering a collaborativeecosystem that benefits all participants. Conversely, closed IP pro-tected MAPs limit access to valuable information and hinder theprogress of the industry as a whole.In order to unleash the full potential of MAPs, key initiativesneed to be launched to overcome the remaining obstacles andspread skills and know-how that will promote game-changingtechnology that benefits all of society.To shape funding opportunities and pave the way for MAPdevelopment, several key aspects must be addressed. First, gov-ernments should focus on supporting the development of self-driving lab platforms that facilitate methodology developmentrather than solely focusing on materials solutions. This approachencourages a foundation of technological advancement that canbenefit a wide range of industries.Furthermore, governments should incentivize industry andacademia to develop internal MAP systems through both tech-nical and financial support. By fostering partnerships and pro-viding resources, governments can accelerate the adoption andimplementation of MAPs in various sectors, driving innovationand economic growth.Lastly, it is crucial to establish education programs that bridgethe gap between data and materials science. The interface be-tween these two fields presents unique challenges and opportu-nities. By nurturing a skilled workforce with expertise in both ar-eas, we can fully exploit the potential of MAPs and maximize theirimpact on materials development and manufacturing.AcknowledgementsThere was no individual funding for the preparation of this perspectivepaper. However, the authors received funding from various research pro-grams for their general work on Materials Acceleration Platforms and re-lated topics. For a non-exhaustive list, please refer to Sections 2.3 and 2.4.Open access funding enabled and organized by Projekt DEAL.Conflict of InterestThe authors declare no conflict of interest.Keywordsadvanced materials, artificial intelligence, autonomous labs, materials ac-celeration platforms, societal challengesReceived: May 31, 2024Revised: August 9, 2024Published online: September 6, 2024[1] S. Stier, C. Kreisbeck, H. Ihssen, M. A. Popp, J. Hauch, K. Malek,M. Reynaud, J. Carlsson, L. Gold, F. Goumans, I. Todorov, A.Räder, S. T. Bandesha, W. Wenzel, P. Jacques, O. Arcelus, F. Garcia-Moreno, P. Friederich, M. Maglione, S. Clark, A. Laukkanen, M. C.Cabanas, J. Carrasco, I. E. Castelli, H. S. Stein, T. Vegge, S. Nakamae,M. Fabrizio, M. Kozdras, Zenodo 2023, https://doi.org/10.5281/zenodo.8012140.[2] United Nations Framework Convention on Climate Change, UN-FCCC, https://unfccc.int/ (accessed: July 2024).[3] United Nations, Transforming our world: the 2030 Agenda for Sustain-able Development, https://sdgs.un.org/2030agenda (accessed: July2024).[4] United Nations, Sustainable Development - THE 17 GOALS, https://sdgs.un.org/goals (accessed: July 2024).[5] IEA, Global electricity demand growth is slowing, weighed downby economic weakness and high prices - News - IEA, 2022,https://www.iea.org/news/global-electricity-demand-growth-is-slowing-weighed-down-by-economic-weakness-and-high-prices(accessed: July 2024).[6] H. Lee, K. Calvin, D. Dasgupta, G. Krinner, A. Mukherji, P. W.Thorne, C. Trisos, J. Romero, P. Aldunce, K. Barrett, G. Blanco, W.W. Cheung, S. Connors, F. Denton, A. Diongue-Niang, D. Dodman,M. Garschagen, O. Geden, B. Hayward, C. Jones, F. Jotzo, T.Krug, R. Lasco, Y.-Y. Lee, V. Masson-Delmotte, M. Meinshausen, K.Mintenbeck, A. Mokssit, F. E. Otto, M. Pathak, et al., IPCC, 2023: Cli-mate Change 2023: Synthesis Report. Contribution of Working GroupsI, II and III to the Sixth Assessment Report of the IntergovernmentalPanel on Climate Change [Core Writing Team, H. Lee and J. Romero(eds.)]. IPCC, Geneva, Switzerland, https://doi.org/10.59327/IPCC/AR6-9789291691647.001.[7] Joint Research Centre European Commission, Materials for Energy,SETIS magazine 2015, https://op.europa.eu/s/zMVE (accessed:July 2024).Adv. Mater. 2024, 36, 2407791 2407791 (22 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dehttps://doi.org/10.5281/zenodo.8012140https://doi.org/10.5281/zenodo.8012140https://unfccc.int/https://sdgs.un.org/2030agendahttps://sdgs.un.org/goalshttps://sdgs.un.org/goalshttps://www.iea.org/news/global-electricity-demand-growth-is-slowing-weighed-down-by-economic-weakness-and-high-priceshttps://www.iea.org/news/global-electricity-demand-growth-is-slowing-weighed-down-by-economic-weakness-and-high-priceshttps://doi.org/10.59327/IPCC/AR6-9789291691647.001https://doi.org/10.59327/IPCC/AR6-9789291691647.001https://op.europa.eu/s/zMVEwww.advancedsciencenews.com www.advmat.de[8] Secretariat-General European Commission, The European GreenDeal, 2019, https://eur-lex.europa.eu/legalcontent/EN/ALL/?uri=COM:2019:640:FIN (accessed: July 2024).[9] European Commission, A Europe fit for the digital age, 2024,https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit-digital-age_en (accessed: July 2024).[10] Secretariat-General European Commission, A New Industrial Strat-egy for Europe, 2020, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52020DC0102 (accessed: July 2024).[11] Secretariat-General European Commission, Updating the 2020New Industrial Strategy: Building a stronger Single Market for Eu-rope’s recovery, 2021, https://commission.europa.eu/document/download/9ab0244c-6ca3-4b11-bef9-422c7eb34f39en?filename=communication-industrial-strategy-update-2020_en.pdf (ac-cessed: July 2024).[12] Secretariat-General European Commission, European industrialstrategy, https://single-market-economy.ec.europa.eu/industry/strategy_en (accessed: July 2024).[13] AMi2030, MATERIALS 2030 MANIFESTO: Systemic Approachof Advanced Materials for Prosperity – A 2030 Perspective, 2022,https://research-and-innovation.ec.europa.eu/system/files/2022-02/advanced-materials-2030-manifesto.pdf (accessed: July 2024).[14] AMi2030, The Materials 2030 roadmap, https://www.ami2030.eu/roadmap/ (accessed: July 2024).[15] AMi2030, Advanced Materials 2030 Initiative, https://www.ami2030.eu/ (accessed: July 2024).[16] A. Benayad, D. Diddens, A. Heuer, A. N. Krishnamoorthy, M. Maiti,F. Le Cras, M. Legallais, F. Rahmanian, Y. Shin, H. Stein, M. Winter,C. Wölke, P. Yan, I. Cekic-Laskovic, Adv. Energy Mater. 2022, 12,17.[17] R. Thun, Rev. Sci. Instrum. 1959, 30, 399.[18] NSTC Subcommittee on the Materials Genome Initiative, MaterialsGenome Initiative Strategic Plan, 2021, https://www.mgi.gov/sites/default/files/documents/MGI-2021-Strategic-Plan.pdf (accessed:July 2024).[19] MGI, Materials Genome Initiative, https://www.mgi.gov/ (accessed:July 2024).[20] hte, the high throughput experimentation company, https://www.hte-company.com/en/ (accessed: July 2024).[21] Wildcat Discovery Technologies, US-made Cathodes High-Performance Producer, https://www.wildcatdiscovery.com/ (ac-cessed: July 2024).[22] A. Aspuru-Guzik, K. Persson, Materials Acceleration Platform: Ac-celerating Advanced Energy Materials Discovery by Integrating High-Throughput Methods and Artificial Intelligence. Mission Innovation:Innovation Challenge 6, 2018, https://dash.harvard.edu/handle/1/35164974 (accessed: July 2024).[23] G. A. Ozin, J. Y. Y. Loh, Energy Materials Discovery: Enabling a Sus-tainable Future, Royal Society of Chemistry, Cambridge, 2022, https://doi.org/10.1039/9781839163838.[24] S. Divilov, H. Eckert, D. Hicks, C. Oses, C. Toher, R. Friedrich, M.Esters, M. J. Mehl, A. C. Zettel, Y. Lederer, E. Zurek, J.-P. Maria, D. W.Brenner, X. Campilongo, S. Filipovíc, W. G. Fahrenholtz, C. J. Ryan,C. M. DeSalle, R. J. Crealese, D. E. Wolfe, A. Calzolari, S. Curtarolo,Nature 2024, 625, 66.[25] GO FAIR, FAIR Principles, https://www.go-fair.org/fair-principles/(accessed: July 2024).[26] M. D. Wilkinson, M. Dumontier, I. J. J. Aalbersberg, G. Appleton, M.Axton, A. Baak, N. Blomberg, J.-W. Boiten, L. B. da Silva Santos, P. E.Bourne, J. Bouwman, A. J. Brookes, T. Clark, M. Crosas, I. Dillo, O.Dumon, S. Edmunds, C. T. Evelo, R. Finkers, A. Gonzalez-Beltran,A. J. G. Gray, P. Groth, C. Goble, J. S. Grethe, J. Heringa, P. A. C. ’tHoen, R. Hooft, T. Kuhn, R. Kok, J. Kok, et al., Scientific Data 2016,3, 160018.[27] G. Goldbeck, A. Simperler, N. Adamovic, S.-E. Björling, V. Cantrill,S. Chiacchiera, T. Exner, I. Furxhi, D. Gao, M. Karray, D. Kiritsis, J.Lomax, I. Lynch, N. Matentzoglu, M. Noeske, F. Piroi, M. Poveda-Villalón, A. Sarkar, K. Vladislavleva, Zenodo 2023, https://doi.org/10.5281/zenodo.10057816.[28] E. Ghedini, A. Hashibon, J. Friis, DOME 4.0: Deliverable D3.1 -“Semantic data exchange ontology”, 2022, https://dome40.eu/sites/default/files/DOME%204.0%20D3.1%20Semantic%20data%20exchange%20ontology%202022.02.28%20PU.pdf (accessed:July 2024).[29] GitHub Contributors, DOME-4-0/data-set-ontology: DOME EMMOcompliant Dataset Ontology, https://github.com/DOME-4-0/data-set-ontology (accessed: July 2024).[30] M. Barker, N. P. Chue Hong, D. S. Katz, A.-L. Lamprecht, C.Martinez-Ortiz, F. Psomopoulos, J. Harrow, L. J. Castro, M.Gruenpeter, P. A. Martinez, T. Honeyman, Scientific Data 2022, 9,622.[31] D. P. Tabor, L. M. Roch, S. K. Saikin, C. Kreisbeck, D. Sheberla,J. H. Montoya, S. Dwaraknath, M. Aykol, C. Ortiz, H. Tribukait,C. Amador-Bedolla, C. J. Brabec, B. Maruyama, K. A. Persson, A.Aspuru-Guzik, Nat. Rev. Mater. 2018, 3, 5.[32] R. Mullin, The lab of the future is now: Recent demonstrations of AI-directed automation may herald a new world for drug and materials dis-covery, 2021, https://cen.acs.org/business/informatics/lab-future-ai-automated-synthesis/99/i11 (accessed: July 2024).[33] R. B. Merrifield, J. M. Stewart, N. Jernberg, Anal. Chem. 1966, 38,1905.[34] H. Winicov, J. Schainbaum, J. Buckley, G. Longino, J. Hill, C. E.Berkoff, Anal. Chim. Acta 1978, 103, 469.[35] J. S. Lindsey, Chemom. Intell. Lab. Syst. 1992, 17, 15.[36] Fraunhofer ISE, Center for High Efficiency Solar Cells,https://www.ise.fraunhofer.de/en/rd-infrastructure/center/center-for-high-efficiency-solar-cells.html (accessed: July 2024).[37] S. Back, A. Aspuru-Guzik, M. Ceriotti, G. Gryn’ova, B. Grzybowski,G. H. Gu, J. Hein, K. Hippalgaonkar, R. Hormázabal, Y. Jung, S. Kim,W. Y. Kim, S. M. Moosavi, J. Noh, C. Park, J. Schrier, P. Schwaller, K.Tsuda, T. Vegge, O. A. von Lilienfeld, A. Walsh, Digital Discov. 2024,3, 23.[38] M. S. Strano, S. M. Walsh, Robotic Systems and Autonomous Plat-forms: Advances in Materials and Manufacturing, Woodhead Pub-lishing in materials. Woodhead Publishing an imprint of Elsevier,Duxford, United Kingdom, 2019, https://doi.org/10.1016/C2016-0-01129-4.[39] BIG-MAP, Battery Interface Genome - Materials Acceleration Platform,https://www.big-map.eu/ (accessed: July 2024).[40] Battery 2030+, Sustainable batteries of the future, https://battery2030.eu/ (accessed: July 2024).[41] A. Bhowmik, M. Berecibar, M. Casas-Cabanas, G. Csanyi, R.Dominko, K. Hermansson, M. R. Palacin, H. S. Stein, T. Vegge, Adv.Energy Mater. 2022, 12, 17.[42] Fraunhofer ISC, Materials Acceleration Platforms (MAPs), https://www.isc.fraunhofer.de/en/fields-of-activity/applications/digital-transformation/maps-materials-acceleration-platforms.html(accessed: July 2024).[43] S. Haeusner, L. Herbst, P. Bittorf, T. Schwarz, C. Henze, M.Mauermann, J. Ochs, R. Schmitt, U. Blache, A. Wixmerten, S. Miot,I. Martin, O. Pullig, Front. Med. 2021, 8, 712917.[44] S. Dembski, T. Schwarz, M. Oppmann, S. T. Bandesha, J. Schmid,S. Wenderoth, K. Mandel, J. Hansmann, Sci. Rep. 2023, 13, 11440.[45] Viperlab, Fully connected virtual and physical perovskite photovoltaicslab, https://www.viperlab.eu/ (accessed: July 2024).[46] HMC, AutoPeroSol: Towards automatic data managementand a common ontology for perovskite solar cell device data,https://helmholtz-metadaten.de/en/inf-projects/autoperosol-Adv. Mater. 2024, 36, 2407791 2407791 (23 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dehttps://eur-lex.europa.eu/legalcontent/EN/ALL/?uri=COM:2019:640:FINhttps://eur-lex.europa.eu/legalcontent/EN/ALL/?uri=COM:2019:640:FINhttps://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit-digital-age_enhttps://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit-digital-age_enhttps://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52020DC0102https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52020DC0102https://commission.europa.eu/document/download/9ab0244c-6ca3-4b11-bef9-422c7eb34f39en?filename=communication-industrial-strategy-update-2020_en.pdfhttps://commission.europa.eu/document/download/9ab0244c-6ca3-4b11-bef9-422c7eb34f39en?filename=communication-industrial-strategy-update-2020_en.pdfhttps://commission.europa.eu/document/download/9ab0244c-6ca3-4b11-bef9-422c7eb34f39en?filename=communication-industrial-strategy-update-2020_en.pdfhttps://single-market-economy.ec.europa.eu/industry/strategy_enhttps://single-market-economy.ec.europa.eu/industry/strategy_enhttps://research-and-innovation.ec.europa.eu/system/files/2022-02/advanced-materials-2030-manifesto.pdfhttps://research-and-innovation.ec.europa.eu/system/files/2022-02/advanced-materials-2030-manifesto.pdfhttps://www.ami2030.eu/roadmap/https://www.ami2030.eu/roadmap/https://www.ami2030.eu/https://www.ami2030.eu/https://www.mgi.gov/sites/default/files/documents/MGI-2021-Strategic-Plan.pdfhttps://www.mgi.gov/sites/default/files/documents/MGI-2021-Strategic-Plan.pdfhttps://www.mgi.gov/https://www.hte-company.com/en/https://www.hte-company.com/en/https://www.wildcatdiscovery.com/https://dash.harvard.edu/handle/1/35164974https://dash.harvard.edu/handle/1/35164974https://doi.org/10.1039/9781839163838https://doi.org/10.1039/9781839163838https://www.go-fair.org/fair-principles/https://doi.org/10.5281/zenodo.10057816https://doi.org/10.5281/zenodo.10057816https://dome40.eu/sites/default/files/DOME%204.0%20D3.1%20Semantic%20data%20exchange%20ontology%202022.02.28%20PU.pdfhttps://dome40.eu/sites/default/files/DOME%204.0%20D3.1%20Semantic%20data%20exchange%20ontology%202022.02.28%20PU.pdfhttps://dome40.eu/sites/default/files/DOME%204.0%20D3.1%20Semantic%20data%20exchange%20ontology%202022.02.28%20PU.pdfhttps://github.com/DOME-4-0/data-set-ontologyhttps://github.com/DOME-4-0/data-set-ontologyhttps://cen.acs.org/business/informatics/lab-future-ai-automated-synthesis/99/i11https://cen.acs.org/business/informatics/lab-future-ai-automated-synthesis/99/i11https://www.ise.fraunhofer.de/en/rd-infrastructure/center/center-for-high-efficiency-solar-cells.htmlhttps://www.ise.fraunhofer.de/en/rd-infrastructure/center/center-for-high-efficiency-solar-cells.htmlhttps://doi.org/10.1016/C2016-0-01129-4https://doi.org/10.1016/C2016-0-01129-4https://www.big-map.eu/https://battery2030.eu/https://battery2030.eu/https://www.isc.fraunhofer.de/en/fields-of-activity/applications/digital-transformation/maps-materials-acceleration-platforms.htmlhttps://www.isc.fraunhofer.de/en/fields-of-activity/applications/digital-transformation/maps-materials-acceleration-platforms.htmlhttps://www.isc.fraunhofer.de/en/fields-of-activity/applications/digital-transformation/maps-materials-acceleration-platforms.htmlhttps://www.viperlab.eu/www.advancedsciencenews.com www.advmat.detowards-automatic-data-management-and-a-common-ontology-for-perovskite-solar-cell-device-data (accessed: July 2024).[47] T. Unold, Faraday Discuss. 2022, 239, 235.[48] J. Wagner, C. G. Berger, X. Du, T. Stubhan, J. A. Hauch, C. J. Brabec,J. Mater. Sci. 2021, 56, 16422.[49] S. Langner, F. Häse, J. D. Perea, T. Stubhan, J. Hauch, L. M. Roch,T. Heumueller, A. Aspuru-Guzik, C. J. Brabec, Adv. Mater. (DeerfieldBeach, Fla.) 2020, 32, e1907801.[50] T. Osterrieder, F. Schmitt, L. Lüer, J. Wagner, T. Heumüller, J. Hauch,C. J. Brabec, Energy Environ. Sci. 2023, 16, 3984.[51] J. Zhang, B. Liu, Z. Liu, J. Wu, S. Arnold, H. Shi, T. Osterrieder, J. A.Hauch, Z. Wu, J. Luo, J. Wagner, C. G. Berger, T. Stubhan, F. Schmitt,K. Zhang, M. Sytnyk, T. Heumueller, C. M. Sutter-Fella, I. M. Peters,Y. Zhao, C. J. Brabec, Adv. Energy Mater. 2023, 13, 48.[52] Y. Zhao, T. Heumueller, J. Zhang, J. Luo, O. Kasian, S. Langner, C.Kupfer, B. Liu, Y. Zhong, J. Elia, A. Osvet, J. Wu, C. Liu, Z. Wan, C.Jia, N. Li, J. Hauch, C. J. Brabec, Nat. Energy 2022, 7, 144.[53] T. Osterrieder, Investigation and Optimization of quaternary OrganicSolar Cells utilizing HTE and Bayesian Optimizer in a semiautonomouslaboratory approach [Unpublished master’s thesis], Ph.D. Thesis, FAUErlangen, 2021.[54] J. Carrasco, O. Lakuntza, Artificial intelligence for automatedand autonomous discovery of better battery materials, 2021,https://cicenergigune.com/en/blog/artificial-intelligence-discovery-battery-materials (accessed: July 2024).[55] SynBio Powerhouse, Synthetic Biology Ecosystem, https://www.synbio.fi/ (accessed: July 2024).[56] CAIMEE, AI-Driven Integrated and Automated Materials Design forExtreme Environments (AMDEE), https://hemi.jhu.edu/caimee/research/ai-driven-integrated-and-automated-materials-design-for-extreme-environments-amdee/ (accessed: July 2024).[57] TU Berlin, BasCat - UniCat BASF JointLab, https://www.tu.berlin/en/bascat (accessed: July 2024).[58] C. Kunkel, F. Rüther, F. Felsen, C. W. P. Pare, A. Terzi, R. Baumgarten,E. Gioria, R. Naumann d’Alnoncourt, C. Scheurer, F. Rosowski, K.Reuter, ACS Catal. 2024, 14, 9008.[59] H. J. Yoo, N. Kim, H. Lee, D. Kim, L. T. C. Ow, H. Nam, C. Kim,S. Y. Lee, K.-Y. Lee, D. Kim, S. S. Han, Adv. Funct. Mater. 2024, 34,27.[60] S. S. Han, H. J. Yoo, K.-Y. Lee, D. Kim, OCTOPUS: OperationControl System for Task Optimization and Job Parallelization viaa User-Optimal Scheduler, 2024, https://doi.org/10.21203/rs.3.rs-4254570/v1.[61] MaterialDigital, The Material Digitalization Platform, https://www.materialdigital.de/ (accessed: July 2024).[62] FAIRmat, FAIRmat: is building a FAIR data infrastructure forcondensed-matter physics and the chemical physics of solids, https://www.fairmat-nfdi.eu/fairmat (accessed: July 2024).[63] ,DIADEM: an exploratory Priority Research Programme and Infrastruc-ture linking materials and AI, https://www.pepr-diadem.fr/en/ (ac-cessed: August 2024).[64] EPSRC, Physical Sciences Data Infrastructure (PSDI) Phase 1Pilot, https://gow.epsrc.ukri.org/NGBOViewGrant.aspx?GrantRef=EP/W032252/1 (accessed: July 2024).[65] CAPeX, 08.08.2024, https://capex.dtu.dk/.[66] NIMS-MDPF: Materials Data Platform in National Institute for Mate-rials Science, 07.08.2024, https://dice.nims.go.jp/.[67] Y. Xu, M. Yamazaki, P. Villars, Jpn. J. Appl. Phys. 2011, 50, 11RH02.[68] Atomwork-adv: Inorganic materials database, 07.08.2024, https://atomwork-adv.nims.go.jp/.[69] M. Ishii, T. Ito, H. Sado, I. Kuwajima, Sci. Technol. Adv. Mater.: Meth.2024, 4, 1.[70] PoLyInfo: Polymer database, 08.08.2024, https://polymer.nims.go.jp/.[71] K. Sawada, K. Kimura, F. Abe, Y. Taniuchi, K. Sekido, T. Nojima, T.Ohba, H. Kushima, H. Miyazaki, H. Hongo, T. Watanabe, Sci. Tech-nol. Adv. Mater. 2019, 20, 1131.[72] Y. Furuya, H. Nishikawa, H. Hirukawa, N. Nagashima, E. Takeuchi,Sci. Technol. Adv. Mater. 2019, 20, 1055.[73] Kinzoku: Metallic material database, 08.08.2024, https://metallicmaterials.nims.go.jp/.[74] Kakusan: Diffusion database, 08.08.2024, https://diffusion.nims.go.jp/.[75] CPDDB: Computational Phase Diagram Database, 08.08.2024, https://cpddb.nims.go.jp/.[76] Thermophysical: Thermophysical property database, 08.08.2024,https://thermophys.nims.go.jp/.[77] RDE: Research Data Express, 08.08.2024, https://rde.nims.go.jp/.[78] M. Demura, T. Koseki, Mater. Trans. 2020, 61, 2041.[79] S. Minamoto, T. Kadohira, K. Ito, M. Watanabe, Mater. Trans. 2020,61, 2067.[80] M. Demura, ISIJ International 2024, 64, 503.[81] Mint: Materials integration by network technology, 08.08.2024,https://www.mintsys.jp/.[82] V. Nandal, S. Dieb, D. S. Bulgarevich, T. Osada, T. Koyama, S.Minamoto, M. Demura, Sci. Rep. 2023, 13, 12660.[83] MGI: Materials Genome Initiative, https://www.nist.gov/mgi (ac-cessed: July 2024).[84] University of Toronto and Acceleration Consortium, Accelerating thediscovery of materials and molecules needed for a sustainable future,https://acceleration.utoronto.ca/ (accessed: July 2024).[85] EMMC ASBL, The European Materials Modelling Council,https://emmc.eu/ (accessed: July 2024).[86] GitHub Contributors - emmo-repo/EMMO, Elementary Multiper-spective Material Ontology (EMMO), https://github.com/emmo-repo/EMMO (accessed: July 2024).[87] EMCC, The European Materials Characterisation Council, http://characterisation.eu/ (accessed: July 2024).[88] M. T. Horsch, B. Schembera, H. A. Preisig, in 2023 3rd InternationalConference on Applied Artificial Intelligence (ICAPAI), IEEE, 2023, pp.1–6, https://doi.org/10.1109/ICAPAI58366.2023.10193944.[89] EMIRI, The Energy Materials Industrial Research Initiative, https://emiri.eu/ (accessed: July 2024).[90] European Commission, Directorate-General for Research, Innova-tion and A. Baas, What makes a material function? – Let me computethe ways – Modelling in H2020 LEIT-NMBP programme materials andnanotechnology projects – Sixth version, Publications Office of the Eu-ropean Union, 2017, https://doi.org/10.2777/417118.[91] OPC Foundation, Unified Architecture, https://opcfoundation.org/about/opc-technologies/opc-ua/ (accessed: July 2024).[92] OpenAPI Initiative, The world’s most widely used API description stan-dard, https://www.openapis.org/ (accessed: July 2024).[93] R. T. Fielding, Architectural Styles and the Design of Network-based Software Architectures, Dissertation, University of Califor-nia, Irvine, 2000, https://ics.uci.edu/⟨fielding/pubs/dissertation/fielding_dissertation.pdf (accessed: July 2024).[94] C. Pahl, IEEE Cloud Comput. 2015, 2, 24.[95] M. Eminizer, S. Tabrisky, A. Sharifzadeh, C. DiMarco, J. M.Diamond, K. T. Ramesh, T. C. Hufnagel, T. M. McQueen, D. Elbert,J. Open Source Softw. 2023, 8, 4896.[96] M. Dreger, M. J. Eslamibidgoli, M. H. Eikerling, K. Malek, J. Mater.Inform. 2023, 3, 1.[97] S. Clark, F. L. Bleken, S. Stier, E. Flores, C. W. Andersen, M. Marcinek,A. Szczesna-Chrzan, M. Gaberscek, M. R. Palacin, M. Uhrin, J. Friis,Adv. Energy Mater. 2022, 12, 2102702.[98] GitHub Contributors - OpenSemanticLab, Web-based semanticdata platform and workflow environment, https://github.com/OpenSemanticLab (accessed: July 2024).Adv. Mater. 2024, 36, 2407791 2407791 (24 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dehttps://helmholtz-metadaten.de/en/inf-projects/autoperosol-towards-automatic-data-management-and-a-common-ontology-for-perovskite-solar-cell-device-datahttps://helmholtz-metadaten.de/en/inf-projects/autoperosol-towards-automatic-data-management-and-a-common-ontology-for-perovskite-solar-cell-device-datahttps://helmholtz-metadaten.de/en/inf-projects/autoperosol-towards-automatic-data-management-and-a-common-ontology-for-perovskite-solar-cell-device-datahttps://helmholtz-metadaten.de/en/inf-projects/autoperosol-towards-automatic-data-management-and-a-common-ontology-for-perovskite-solar-cell-device-datahttps://helmholtz-metadaten.de/en/inf-projects/autoperosol-towards-automatic-data-management-and-a-common-ontology-for-perovskite-solar-cell-device-datahttps://cicenergigune.com/en/blog/artificial-intelligence-discovery-battery-materialshttps://cicenergigune.com/en/blog/artificial-intelligence-discovery-battery-materialshttps://www.synbio.fi/https://www.synbio.fi/https://hemi.jhu.edu/caimee/research/ai-driven-integrated-and-automated-materials-design-for-extreme-environments-amdee/https://hemi.jhu.edu/caimee/research/ai-driven-integrated-and-automated-materials-design-for-extreme-environments-amdee/https://hemi.jhu.edu/caimee/research/ai-driven-integrated-and-automated-materials-design-for-extreme-environments-amdee/https://www.tu.berlin/en/bascathttps://www.tu.berlin/en/bascathttps://doi.org/10.21203/rs.3.rs-4254570/v1https://doi.org/10.21203/rs.3.rs-4254570/v1https://www.materialdigital.de/https://www.materialdigital.de/https://www.fairmat-nfdi.eu/fairmathttps://www.fairmat-nfdi.eu/fairmathttps://www.pepr-diadem.fr/en/https://gow.epsrc.ukri.org/NGBOViewGrant.aspx?GrantRef=EP/W032252/1https://gow.epsrc.ukri.org/NGBOViewGrant.aspx?GrantRef=EP/W032252/1https://capex.dtu.dk/https://dice.nims.go.jp/https://atomwork-adv.nims.go.jp/https://atomwork-adv.nims.go.jp/https://polymer.nims.go.jp/https://polymer.nims.go.jp/https://metallicmaterials.nims.go.jp/https://metallicmaterials.nims.go.jp/https://diffusion.nims.go.jp/https://diffusion.nims.go.jp/https://cpddb.nims.go.jp/https://cpddb.nims.go.jp/https://thermophys.nims.go.jp/https://rde.nims.go.jp/https://www.mintsys.jp/https://www.nist.gov/mgihttps://acceleration.utoronto.ca/https://emmc.eu/https://github.com/emmo-repo/EMMOhttps://github.com/emmo-repo/EMMOhttp://characterisation.eu/http://characterisation.eu/https://doi.org/10.1109/ICAPAI58366.2023.10193944https://emiri.eu/https://emiri.eu/https://doi.org/10.2777/417118https://opcfoundation.org/about/opc-technologies/opc-ua/https://opcfoundation.org/about/opc-technologies/opc-ua/https://www.openapis.org/https://ics.uci.edu/~fielding/pubs/dissertation/fielding_dissertation.pdfhttps://ics.uci.edu/~fielding/pubs/dissertation/fielding_dissertation.pdfhttps://github.com/OpenSemanticLabhttps://github.com/OpenSemanticLabwww.advancedsciencenews.com www.advmat.de[99] W3C, JSON-LD 1.1, 2020, https://www.w3.org/TR/2020/REC-json-ld11-20200716/ (accessed: July 2024).[100] W3C, RDF 1.1 Concepts and Abstract Syntax, 2014, https://www.w3.org/TR/2014/REC-rdf11-concepts-20140225/ (accessed: July2024).[101] GCMAC, Knowledge & Infrastructure to Accelerate Energy MaterialsPipeline, https://gcmac.ca/ (accessed: July 2024).[102] COST Association, CA22123 - European Materials Acceleration Cen-ter for Energy (EU-MACE), https://www.cost.eu/actions/CA22123/(accessed: July 2024).[103] COST Association, CA22143 - European Materials Informatics Net-work (EuMINe), https://www.cost.eu/actions/CA22143/ (accessed:July 2024).[104] C. R. Catlow, M. Davidson, C. Hardacre, G. J. Hutchings,Philos. Trans. Ser. A, Math., Phys., Eng. Sci. 2016, 374,2061.[105] B. Burger, P. M. Maffettone, V. V. Gusev, C. M. Aitchison, Y. Bai, X.Wang, X. Li, B. M. Alston, B. Li, R. Clowes, N. Rankin, B. Harris, R.S. Sprick, A. I. Cooper, Nature 2020, 583, 237.[106] J. Sarkar, A. Kumar, Biotechnol. J. 2021, 16, 2000288.[107] European Commission, Directorate-General for Research and In-novation, Strategic Energy Technology (SET) Plan, 2023, https://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX:52023DC0634(accessed: July 2024).[108] A. Gocke, A. Rothman, F. Flock, H. Schönberger, A.Gandhi, J. Hu, Facing Unprecedented Pressures, 2023, https://www.bcg.com/publications/2023/resilience-chemicals-industry(accessed: July 2024).[109] I. Herrmann, P. Suresh, S. Struwe, The Chemical Industry OutlookFor 2023 And Beyond, 2023, https://www.oliverwyman.com/our-expertise/insights/2023/jan/chemical-industry-outlook-for-2023-and-beyond.html (accessed: July 2024).[110] EMIRI, Re-inventing materials research: Towards accelerated develop-ment by closely orchestrating ai, data, robotics, and high-performancecomputing, 2023, https://emiri.eu/2022/07/04/digital-revolution-in-materials-discovery/ (accessed: July 2024).[111] Accelerate conference, 08.08.2024, https://www.accelerate24.ca/.[112] Hitachi high-tech and hitachi have started collaborating on digitaltransformation of materials development with itri, contributing to im-provements of taiwan’s industrial standards, 08.08.2024, https://www.hitachi-hightech.com/global/en/news/nr20240520.html.[113] Cambridge Enterprise, Intellegens – cambridge enterprise,09.07.2023, https://www.enterprise.cam.ac.uk/portfolio-companies/intellegensaml/.[114] Materials Nexus, Uses AI-Based Materials Discovery Platform to CreateMagNex, a Novel Rare Earth-Free Permanent Magnet, Materials Nexus11.06.2024.[115] A. D. Little, M. Samoylova, S. Wilmet, D. Witthaut, M. Kolk,Digital technologies for sustainability in the European chemi-cal industry, 2023, https://www.adlittle.com/en/insights/report/digitaltechnologies-sustainability-european-chemical-industry (ac-cessed: July 2024).[116] Cefic, New Report Offers a Blueprint for Overcoming Bottlenecks andUnlocking the Full Potential of Digitalisation in the Chemical Industry,2023, https://cefic.org/media-corner/newsroom/new-report-offers-a-blueprint-for-overcoming-bottlenecks-and-unlocking-the-full-potential-of-digitalisation-in-the-chemical-industry/ (ac-cessed: July 2024).[117] Analytics Insight, How Robots And Robotics are Powering theDrug Discovery Processes?, 2020, https://www.analyticsinsight.net/cryptocurrency-analytics-insight/crypto-gurus-have-their-eyes-on-memecoins-this-q3-200x-is-only-the-start (accessed:July 2024).[118] O. J. Wouters, M. McKee, J. Luyten, JAMA 2020, 323, 844.[119] Deep Pharma Intelligence, AI for Drug Discovery Q1 2023,2023, https://www.deep-pharma.tech/ai-in-dd-q1-2023-subscribe(accessed: July 2024).[120] X. Li, P. M. Maffettone, Y. Che, T. Liu, L. Chen, A. I. Cooper, Chem.Sci. 2021, 12, 10742.[121] OSHWA, Open Source Hardware Definition (English), 2012, https://www.oshwa.org/definition/ (accessed: July 2024).[122] OPC Foundation, LADS - Laboratory and Analytical Device Standard,https://opcfoundation.org/markets-collaboration/lads/ (accessed:July 2024).[123] Industry Commons, Building Systems for Industry Collaborations,https://industrycommons.net/ (accessed: July 2024).[124] GitHub Contributors - Mat-O-Lab/MSEO, Repository of Material Sci-ence and Engineering Ontology MSEO, https://github.com/Mat-O-Lab/MSEO (accessed: July 2024).[125] GitHub Contributors - materialdigital/core-ontology, The PMD CoreOntology (PMDco) official release repository, https://github.com/materialdigital/core-ontology (accessed: July 2024).[126] S. P. Stier, X. Xu, L. Gold, M. Möckel, Energy Technology 2024, https://doi.org/10.1002/ente.202301305.[127] J. Francisco Morgado, S. Chiacchiera, Y. Le Franc, M. Poveda-Villalón, H. Karray, M. Gall, Y. Nahshon, D. Helm, A. Simperler,G. Goldbeck, Zenodo 2023, https://doi.org/10.5281/zenodo.10402214.[128] Simon Stier, Zenodo, 2024, https://doi.org/10.5281/ZENODO.11401726.[129] GitHub Contributors - meirwah/awesome-workflow-engines,A curated list of awesome open source workflow engines, https://github.com/meirwah/awesome-workflow-engines (accessed:July 2024).[130] The University of Chicago, Globus: Research IT. Reimagined, https://www.globus.org/ (accessed: July 2024).[131] M. Vogler, J. Busk, H. Hajiyani, P. B. Jørgensen, N. Safaei, I. E.Castelli, F. F. Ramirez, J. Carlsson, G. Pizzi, S. Clark, F. Hanke, A.Bhowmik, H. S. Stein, Matter 2023, 6, 2647.[132] M. Vogler, S. Steensen, F. Ramirez, L. Merker, J. Busk, J. M. Carlsson,L. Rieger, B. Zhang, F. Liot, G. Pizzi, F. Hanke, E. Flores, H. Hajiyani,S. Fuchs, A. Sanin, M. Gaberscek, I. Castelli, S. Clark, T. Vegge, A.Bhowmik, H. S. Stein, ChemRxiv 2024, https://doi.org/10.26434/chemrxiv-2024-vfq1n.[133] S. P. Stier, M. A. Popp, Zenodo 2024, https://doi.org/10.5281/zenodo.11653807.[134] HERCULES - LPSC, Hercules School, https://hercules-school.eu/(accessed: July 2024).[135] GCMAC, GCMAC Summer School 2023, 2023, https://gcmac.de/event/summer-school-2023/ (accessed: July 2024).[136] International Data Spaces, The future of the data econ-omy is here, https://internationaldataspaces.org/ (accessed:July 2024).[137] ERC, European Research Council: Apply for a grant, https://erc.europa.eu/apply-grant (accessed: July 2024).[138] ESFRI, European Strategy Forum on Research Infrastructures, https://www.esfri.eu/ (accessed: July 2024).[139] ESFRI-EOSC Task Force, G. Rossi, Zenodo 2024, https://doi.org/10.5281/zenodo.10980285.[140] Cefic, IAM4EU in the spotlight: Pioneering Advanced Materials for Eu-rope’s Future, https://cefic.org/media-corner/newsroom/iam4eu-in-the-spotlight-pioneering-advanced-materials-for-europes-future/ (accessed: July 2024).[141] European Commission, Directorate-General for Research,and Innovation, Advanced Materials for Industrial Leadership,https://research-and-innovation.ec.europa.eu/document/download/0fcf06ea-c242-44a6-b2cb-daed39584996_en?filename=com_2024_98_1_en_act_part1.pdf (accessed: July 2024).Adv. Mater. 2024, 36, 2407791 2407791 (25 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dehttps://www.w3.org/TR/2020/REC-json-ld11-20200716/https://www.w3.org/TR/2020/REC-json-ld11-20200716/https://www.w3.org/TR/2014/REC-rdf11-concepts-20140225/https://www.w3.org/TR/2014/REC-rdf11-concepts-20140225/https://gcmac.ca/https://www.cost.eu/actions/CA22123/https://www.cost.eu/actions/CA22143/https://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX:52023DC0634https://eur-lex.europa.eu/legal-content/EN/ALL/?uri=CELEX:52023DC0634https://www.bcg.com/publications/2023/resilience-chemicals-industryhttps://www.bcg.com/publications/2023/resilience-chemicals-industryhttps://www.oliverwyman.com/our-expertise/insights/2023/jan/chemical-industry-outlook-for-2023-and-beyond.htmlhttps://www.oliverwyman.com/our-expertise/insights/2023/jan/chemical-industry-outlook-for-2023-and-beyond.htmlhttps://www.oliverwyman.com/our-expertise/insights/2023/jan/chemical-industry-outlook-for-2023-and-beyond.htmlhttps://emiri.eu/2022/07/04/digital-revolution-in-materials-discovery/https://emiri.eu/2022/07/04/digital-revolution-in-materials-discovery/https://www.accelerate24.ca/https://www.hitachi-hightech.com/global/en/news/nr20240520.htmlhttps://www.hitachi-hightech.com/global/en/news/nr20240520.htmlhttps://www.enterprise.cam.ac.uk/portfolio-companies/intellegensaml/https://www.enterprise.cam.ac.uk/portfolio-companies/intellegensaml/https://www.adlittle.com/en/insights/report/digitaltechnologies-sustainability-european-chemical-industryhttps://www.adlittle.com/en/insights/report/digitaltechnologies-sustainability-european-chemical-industryhttps://cefic.org/media-corner/newsroom/new-report-offers-a-blueprint-for-overcoming-bottlenecks-and-unlocking-the-full-potential-of-digitalisation-in-the-chemical-industry/https://cefic.org/media-corner/newsroom/new-report-offers-a-blueprint-for-overcoming-bottlenecks-and-unlocking-the-full-potential-of-digitalisation-in-the-chemical-industry/https://cefic.org/media-corner/newsroom/new-report-offers-a-blueprint-for-overcoming-bottlenecks-and-unlocking-the-full-potential-of-digitalisation-in-the-chemical-industry/https://www.analyticsinsight.net/cryptocurrency-analytics-insight/crypto-gurus-have-their-eyes-on-memecoins-this-q3-200x-is-only-the-starthttps://www.analyticsinsight.net/cryptocurrency-analytics-insight/crypto-gurus-have-their-eyes-on-memecoins-this-q3-200x-is-only-the-starthttps://www.analyticsinsight.net/cryptocurrency-analytics-insight/crypto-gurus-have-their-eyes-on-memecoins-this-q3-200x-is-only-the-starthttps://www.deep-pharma.tech/ai-in-dd-q1-2023-subscribehttps://www.oshwa.org/definition/https://www.oshwa.org/definition/https://opcfoundation.org/markets-collaboration/lads/https://industrycommons.net/https://github.com/Mat-O-Lab/MSEOhttps://github.com/Mat-O-Lab/MSEOhttps://github.com/materialdigital/core-ontologyhttps://github.com/materialdigital/core-ontologyhttps://doi.org/10.1002/ente.202301305https://doi.org/10.1002/ente.202301305https://doi.org/10.5281/zenodo.10402214https://doi.org/10.5281/zenodo.10402214https://doi.org/10.5281/ZENODO.11401726https://doi.org/10.5281/ZENODO.11401726https://github.com/meirwah/awesome-workflow-engineshttps://github.com/meirwah/awesome-workflow-engineshttps://www.globus.org/https://www.globus.org/https://doi.org/10.26434/chemrxiv-2024-vfq1nhttps://doi.org/10.26434/chemrxiv-2024-vfq1nhttps://doi.org/10.5281/zenodo.11653807https://doi.org/10.5281/zenodo.11653807https://hercules-school.eu/https://gcmac.de/event/summer-school-2023/https://gcmac.de/event/summer-school-2023/https://internationaldataspaces.org/https://erc.europa.eu/apply-granthttps://erc.europa.eu/apply-granthttps://www.esfri.eu/https://www.esfri.eu/https://doi.org/10.5281/zenodo.10980285https://doi.org/10.5281/zenodo.10980285https://cefic.org/media-corner/newsroom/iam4eu-in-the-spotlight-pioneering-advanced-materials-for-europes-future/https://cefic.org/media-corner/newsroom/iam4eu-in-the-spotlight-pioneering-advanced-materials-for-europes-future/https://cefic.org/media-corner/newsroom/iam4eu-in-the-spotlight-pioneering-advanced-materials-for-europes-future/https://research-and-innovation.ec.europa.eu/document/download/0fcf06ea-c242-44a6-b2cb-daed39584996_en?filename=com_2024_98_1_en_act_part1.pdfhttps://research-and-innovation.ec.europa.eu/document/download/0fcf06ea-c242-44a6-b2cb-daed39584996_en?filename=com_2024_98_1_en_act_part1.pdfhttps://research-and-innovation.ec.europa.eu/document/download/0fcf06ea-c242-44a6-b2cb-daed39584996_en?filename=com_2024_98_1_en_act_part1.pdfwww.advancedsciencenews.com www.advmat.de[142] European Commission, Directorate-General for Research, andInnovation, Communication on advanced materials for industrialleadership, Publications Office of the European Union, 2024, https://doi.org/10.2777/401624.[143] F. Strieth-Kalthoff, H. Hao, V. Rathore, J. Derasp, T. Gaudin, N. H.Angello, M. Seifrid, E. Trushina, M. Guy, J. Liu, X. Tang, M. Mamada,W. Wang, T. Tsagaantsooj, C. Lavigne, R. Pollice, T. C. Wu, K. Hotta,L. Bodo, … A. Aspuru-Guzik, Science 2024, 384, https://doi.org/10.1126/science.adk9227.[144] T. Chen, Z. Pang, S. He, Y. Li, S. Shrestha, J. M. Little, H. Yang, T.-C.Chung, J. Sun, H. C. Whitley, I.-C. Lee, T. J. Woehl, T. Li, L. Hu, P.-Y.Chen, Nature Nanotechnol. 2024, 19, 782.[145] EUDAT, The EUDAT Collaborative Data Infrastructure, (accessed:September 2024), https://www.eudat.eu/.Adv. Mater. 2024, 36, 2407791 2407791 (26 of 26) © 2024 The Author(s). Advanced Materials published by Wiley-VCH GmbH 15214095, 2024, 45, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adma.202407791 by National Institute For, Wiley Online Library on [10/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmat.dehttps://doi.org/10.2777/401624https://doi.org/10.2777/401624https://doi.org/10.1126/science.adk9227https://doi.org/10.1126/science.adk9227https://www.eudat.eu/ Materials Acceleration Platforms (MAPs): Accelerating Materials Research and Development to Meet Urgent Societal Challenges 1. Motivation and Background 1.1. A Global Challenge 1.2. Importance of Advanced Materials 1.3. The History of Accelerated Advanced Materials Research 2. About MAPs 2.1. The Idea of a MAP 2.2. Why Will it Be Superior? 2.3. Research Projects and Initiatives 2.4. Academic and Industrial Research Frameworks 3. Building Blocks 3.1. Hardware 3.2. Communication Protocols and Interfaces 3.3. Software Infrastructure 3.4. Software 3.5. Community 4. Impact 4.1. Technological Scope 4.2. Use of MAPs in Industry 5. Gaps 5.1. Compatible Infrastructure, Data Structures and Data Sharing 5.2. International, Decentralized and Asynchronous MAP Structures 5.2.1. Standardized Software for Distributed MAPs 5.2.2. Make existing labs 9040˝MAP ready9040˛ 5.3. Education and Training 5.4. IP, Ownership 5.5. Research Strategy 6. Conclusion Acknowledgements Conflict of Interest Keywords