AI In Material Discovery Market

AI In Material Discovery Market Latest Trends

The AI In Material Discovery Market is increasingly characterized by the integration of generative artificial intelligence with established computational materials science. Instead of evaluating only predefined compositions, generative systems create candidate materials based on targeted functional characteristics. Microsoft's MatterGen research introduced a diffusion-based approach for generating inorganic crystal structures with desired properties. This development illustrates the transition from conventional screening toward AI-assisted inverse materials design.

Another significant trend involves scientific foundation models trained on extensive atomic simulation datasets. Researchers develop transferable representations that can support multiple materials prediction tasks without creating separate models for every application. The Open Materials 2024 dataset includes 110 million density functional theory calculations, providing a substantial resource for training and evaluating computational models. Large datasets improve opportunities for testing model performance across different chemical compositions and structural configurations.

Autonomous experimentation is also reshaping materials development workflows. Robotic laboratories integrate algorithms with instruments capable of preparing samples, conducting measurements, and recording results. Active learning systems select subsequent experiments using information collected during earlier testing cycles. This approach allows scientists to concentrate laboratory resources on candidates with promising predicted characteristics.

Industrial adoption increasingly emphasizes compatibility with existing research infrastructure. Materials companies require software that can integrate experimental databases, simulation outputs, laboratory information systems, and scientific collaboration tools. Platforms offering structured data management and explainable predictions are especially relevant to organizations managing complex formulation portfolios. Demand for practical deployment is encouraging technology providers to develop accessible interfaces alongside advanced mathematical models.

Research teams are also expanding the application of AI to sustainable materials engineering. Investigations include recyclable polymers, battery electrode compositions, hydrogen-related materials, and catalytic systems. Computational assessment helps researchers evaluate competing material characteristics before extensive physical testing. Experimental validation remains essential because predicted stability does not establish manufacturability or long-term industrial performance.

AI In Material Discovery Market Dynamics

The AI In Material Discovery Market is influenced by research productivity requirements, advances in computational modeling, and growing demand for specialized material properties. Conventional experimentation can involve sequential synthesis, characterization, and redesign, particularly when materials exhibit complex chemical interactions. AI systems assist scientists by identifying patterns across experimental and simulation datasets. Their effectiveness depends on the quality of available data, the suitability of predictive models, and the reproducibility of laboratory outcomes.

Scientific organizations increasingly evaluate AI systems through prediction accuracy, experimental usefulness, and compatibility with existing development processes. Manufacturers require evidence that algorithmically selected candidates can satisfy production conditions and operational specifications. Computing infrastructure, specialized talent, and laboratory integration also influence deployment decisions. These factors create different adoption conditions across research institutions, technology companies, and industrial material producers.

Driver

Increasing demand for accelerated materials screening and predictive research capabilities.

A principal driver of the AI In Material Discovery Market is the need to evaluate complex materials efficiently before undertaking extensive physical experimentation. Industrial material development frequently requires simultaneous evaluation of electrical, mechanical, thermal, and chemical characteristics. Computational systems help researchers identify promising candidates while rejecting compositions with unfavorable predicted properties. This process is especially relevant when experiments require specialized equipment or demanding preparation procedures.

Machine learning models can identify relationships between chemical composition and material behavior using historical measurements and simulation results. Researchers use these relationships to estimate properties for candidates that have not yet undergone laboratory evaluation. Graph-based architectures are useful when atomic connectivity and local chemical environments influence performance. Materials discovery programs consequently incorporate predictive models into screening workflows alongside established simulation techniques.

Restraint

Limited availability of standardized and experimentally validated materials datasets.

Data quality remains a major restraint for AI-based materials development. Materials datasets frequently originate from different laboratories, instruments, simulation methods, and experimental procedures. Variations in temperature, pressure, specimen preparation, and measurement protocols can produce inconsistent results for related materials. Models trained on inconsistent records may generate predictions that fail when evaluated under unfamiliar conditions.

Another limitation involves the uneven representation of chemical structures and material classes. Frequently studied compounds may have extensive published information, whereas emerging compositions can lack reliable experimental measurements. Proprietary industrial data also restrict access to valuable formulations and processing information. Organizations therefore require data cleaning, metadata standardization, and rigorous validation before relying on model-generated recommendations.

Opportunity

Expansion of AI-assisted discovery for next-generation energy and sustainable industrial materials.

Energy-related materials present substantial opportunities for AI-assisted scientific development. Battery manufacturers investigate electrode materials, electrolytes, separators, and protective coatings to improve electrochemical performance and operational safety. Researchers also evaluate catalysts for hydrogen production, carbon conversion, and industrial chemical processing. AI models support candidate identification and help establish relationships between composition, structure, and functional behavior.

Sustainable materials research creates additional opportunities in polymer design and industrial chemical substitution. Manufacturers seek formulations that satisfy performance requirements while reducing reliance on restricted substances or difficult processing conditions. Machine learning can support the evaluation of alternative ingredients and processing combinations. The ability to connect materials predictions with manufacturing requirements is particularly valuable for translating research discoveries into industrial applications.

Challenge

Converting computationally promising candidates into reproducible and scalable physical materials.

A major challenge concerns the difference between computational predictions and experimental material performance. Algorithms can identify structures with favorable calculated properties without establishing practical synthesis conditions. Some proposed compounds require processing environments that are difficult to reproduce or maintain. Others may exhibit desirable theoretical properties but undergo unexpected transformations during manufacturing.

Experimental characterization must therefore assess structural stability, impurities, defects, and environmental sensitivity. Industrial qualification introduces additional requirements involving batch consistency, production compatibility, and long-term reliability. Researchers must distinguish uncertainty in algorithmic predictions from variability introduced during physical synthesis. Successful deployment depends on continuous interaction between computational scientists, laboratory specialists, and manufacturing engineers.

AI In Material Discovery SWOT Analysis

Strengths

• High-throughput computational screening: AI systems can evaluate extensive candidate libraries using established scientific datasets.

• Multivariable property optimization: Algorithms can investigate relationships involving composition, structure, processing conditions, and performance.

• Integration with physics-based simulations: Hybrid approaches combine established scientific calculations with data-driven predictions.

• Improved experimental prioritization: Model recommendations help researchers select informative tests for laboratory investigation.

Weaknesses

• Sensitivity to training data: Prediction reliability depends on dataset quality and representative chemical information.

• Model interpretability limitations: Complex architectures can make scientific explanations difficult to establish.

• Specialized implementation requirements: Effective deployment requires expertise in chemistry, materials science, software engineering, and machine learning.

• Uneven digital readiness: Research organizations differ in their ability to integrate existing laboratory records and computational systems.

Opportunities

• Advanced battery research: AI supports investigation of solid electrolytes and electrochemical interfaces.

• Circular materials engineering: Computational approaches can assist the design of recyclable and reusable industrial materials.

• Autonomous research facilities: Laboratory robotics create opportunities for integrated experimental optimization.

• Scientific collaboration platforms: Shared datasets and interoperable models can support research across multiple organizations.

Threats

• Intellectual property disputes: Unclear ownership of generated structures and model-derived discoveries may complicate commercialization.

• Cybersecurity exposure: Digitized laboratory systems and proprietary materials datasets require effective information protection.

• Scientific reproducibility risks: Unverified model outputs can reduce confidence in research results.

• Technological obsolescence: Rapid advances in algorithms and computing architectures may require frequent system updates.

AI In Material Discovery Segmentation Analysis

The AI In Material Discovery Market is segmented by component, material type, technology, application, and end use. These classifications represent different aspects of the materials research ecosystem, including software infrastructure, scientific disciplines, modeling techniques, and industrial requirements. Software platforms support computational screening, while hardware enables intensive processing and scientific workloads. Service providers contribute implementation, customization, and technical support.

Material categories determine the physical properties and experimental methods investigated through AI systems. Technology classifications distinguish machine learning, deep learning, generative models, natural language processing, and predictive analytics. Application categories identify the research functions supported by these technologies. End-use segmentation reflects differences between pharmaceutical research, industrial material development, electronics manufacturing, and academic research.

By Component

Software represents the leading component segment, holding 55.0% of the global market in 2025 according to published industry estimates. This category includes predictive modeling platforms, materials simulation tools, data analytics systems, molecular design software, and materials informatics platforms. These applications organize scientific information and enable researchers to investigate candidate compositions using computational techniques. Software accessibility and integration with existing research environments influence purchasing decisions.

Hardware includes high-performance computing systems, graphics processing units, specialized accelerators, data storage servers, and computational modeling workstations. These resources support intensive numerical calculations and training of scientific machine learning models. Services cover consulting and integration, custom model development, data management, simulation and testing, and training and support. Service providers help organizations implement scientific AI while addressing operational requirements.

By Material Type

Chemicals accounted for 30.0% of the global market in 2025 in published industry segmentation. AI applications in chemical materials include catalyst development, specialty ingredient optimization, and molecular property prediction. Polymers represent another important category, particularly for formulation design, durability assessment, and process-property relationships. Metals and alloys require modeling of composition, microstructure, deformation, and temperature-dependent behavior.

Ceramics and composites involve additional challenges related to structural behavior and complex processing conditions. Nanomaterials research investigates particle characteristics, surface properties, and interactions at small length scales. Semiconductor discovery focuses on electronic structures, defect behavior, and functional materials for advanced devices. Biomaterials research incorporates biological compatibility and material interactions with physiological environments, requiring specialized validation methods.

By Technology

Machine learning held a 36.0% global market share in 2025 according to published market estimates. Conventional machine learning methods are applied to experimental datasets, material descriptors, and structured property records. Deep learning supports complex representations of atomic arrangements and molecular relationships. Graph neural networks and equivariant neural architectures are important developments within this category.

Generative AI enables researchers to propose material candidates that satisfy selected structural or functional requirements. Natural language processing supports the extraction of scientific information from technical documents, patents, and laboratory reports. Predictive analytics applies computational models to estimate material behavior under defined conditions. Technology selection depends on research objectives, dataset characteristics, computing resources, and scientific validation requirements.

By Application

Material property prediction represented 28.0% of the global market in 2025 based on published market segmentation. Researchers employ predictive systems to estimate thermal conductivity, electrical characteristics, mechanical properties, and chemical stability. Molecular modeling and simulation support the investigation of structural interactions and energy-related characteristics. These applications provide scientific information before researchers commit to detailed experimental programs.

Drug and pharmaceutical material discovery includes molecular solids, crystalline forms, and formulation-related material characteristics. AI can support investigations of solid-state behavior and pharmaceutical excipient compatibility. Battery and energy storage material discovery includes electrodes, electrolytes, and interface materials. Researchers evaluate candidates against electrical performance, chemical stability, safety requirements, and manufacturing conditions.

By End Use

Pharmaceuticals and biotechnology constitute a leading end-use category in published industry analysis. Research organizations apply AI to molecular systems, solid-state materials, and formulation-related investigations. Chemicals and advanced materials companies focus on catalysts, specialty formulations, and industrial performance requirements. Electronics and semiconductor manufacturers investigate materials with specific electrical and structural properties.

Automotive organizations evaluate battery-related materials, protective coatings, and lightweight structures. Energy and utility applications include storage technologies, electrochemical materials, and functional systems for energy conversion. Aerospace and defense researchers investigate materials capable of operating under demanding thermal and mechanical conditions. Research institutions contribute foundational datasets, new modeling architectures, and experimentally validated scientific discoveries.

Regional Analysis

Regional market development reflects differences in research infrastructure, industrial demand, computing capabilities, and scientific collaboration. North America maintains a leading position, supported by established AI companies and national laboratory resources. Europe benefits from specialized industrial manufacturing and computational engineering expertise. Asia-Pacific combines semiconductor, battery, and advanced materials production ecosystems.

The Middle East and Africa represent developing opportunities for AI-assisted materials research, with activities concentrated in selected industrial and academic centers. Government research strategies, technical education, and investment in digital infrastructure influence adoption. Cross-border partnerships enable institutions to access expertise and computational resources. Regional comparisons must also account for differences in market definitions and reporting methods.

- North America: North America held a 41.6% global market share in 2025 according to published industry estimates. The United States is a major center for AI-based materials research, supported by universities, national laboratories, and technology developers. Canada's research ecosystem contributes expertise in scientific machine learning and advanced computational methods. Industrial applications include electronic materials, specialty chemicals, and advanced engineering.

North American organizations benefit from established access to cloud infrastructure and high-performance computing facilities. Research collaborations support the evaluation of emerging computational architectures and experimental techniques. Companies can integrate materials discovery algorithms into existing digital engineering environments. This strengthens regional capabilities in scientific software development and applied industrial research.

Academic commercialization also influences market activity. Research groups translate computational discoveries into specialized software platforms and laboratory services. Technology transfer arrangements connect university research with industrial materials development. These relationships help create commercially relevant applications from foundational scientific discoveries.

- Europe: Europe represents an established market for AI-enabled materials research across industrial manufacturing, specialty chemicals, and scientific computing. Germany contributes significant expertise in chemical engineering and automotive materials. France supports scientific software development and advanced computational research. Switzerland and the United Kingdom contribute through specialized scientific institutions and technology companies.

European organizations increasingly connect materials informatics with industrial product development. Manufacturers require traceable data to evaluate formulation changes, material substitutions, and technical specifications. Digital engineering platforms support collaboration between laboratory scientists and manufacturing specialists. This connection is important for industries operating under strict technical qualification requirements.

Regional research programs also investigate battery technologies, resource efficiency, and environmentally compatible chemical processes. AI can assist the identification of candidate materials for these objectives. Collaborative research networks support scientific knowledge exchange between industrial and academic participants. A verified standalone European percentage of the global market has not been established from the available comparable data.

- Asia-Pacific: Asia-Pacific is an important market for AI-driven materials development because of its extensive electronics manufacturing and advanced industrial research capabilities. China supports computational materials research across batteries, electronic materials, and chemical engineering. Japan contributes expertise in precision materials, advanced ceramics, and specialized industrial applications. South Korea maintains significant research activity related to semiconductor and battery technologies.

Research organizations in the region increasingly investigate AI-supported prediction and simulation systems. Semiconductor development requires precise control of material composition, electrical behavior, and structural defects. Battery engineering introduces additional requirements involving electrochemical stability and material interfaces. These research priorities create practical applications for predictive computational tools.

India is developing capabilities through academic materials science research and expanding computational infrastructure. Regional collaborations connect materials scientists with software engineering and advanced manufacturing specialists. Adoption also depends on research data access, industrial validation, and laboratory modernization. Asia-Pacific is identified as a high-growth region in published market studies, although a directly comparable regional share has not been verified here.

- Middle East & Africa: The Middle East and Africa market is developing through research programs involving industrial chemicals, energy-related materials, and scientific computing. Selected Gulf countries are expanding their capabilities in advanced technology research and industrial diversification. Saudi Arabia and the United Arab Emirates support academic and industrial activities relevant to materials engineering. These initiatives establish potential applications for AI-assisted scientific investigation.

Energy-related research offers opportunities to investigate catalysts, membranes, protective materials, and chemical separation systems. Advanced computational models can support material selection under demanding operational conditions. Scientific partnerships provide access to specialized expertise and technical infrastructure. Institutional cooperation is particularly relevant where local laboratory capacity remains limited.

African research institutions contribute to computational chemistry and materials science through university programs and international collaborations. Availability of computing systems, laboratory instruments, and specialist personnel influences implementation. Open scientific datasets can support research participation where proprietary resources are difficult to obtain. The region's global market-share percentage cannot be stated reliably without consistent region-specific measurement.

Competitive Landscape

The AI In Material Discovery Market includes established technology companies, scientific software developers, materials informatics specialists, and industrial research organizations. Competition centers on prediction accuracy, data integration, computational efficiency, and scientific applicability. Leading participants include Microsoft, Google, IBM, Schrödinger, Dassault Systèmes, Citrine Informatics, and MaterialsZone. Companies differentiate their capabilities through proprietary algorithms, scientific workflows, and specialized research applications.

Large technology companies contribute computing infrastructure, foundational AI architectures, and research expertise. Scientific software suppliers provide domain-specific tools that integrate molecular modeling, simulation, and experimental information. Specialized materials informatics providers concentrate on enterprise data integration and predictive formulation development. Their competitive positioning depends on delivering solutions that materials scientists can evaluate and incorporate into established workflows.

Market concentration assessments vary depending on how AI-based materials discovery activities are defined. One published competitive analysis identifies the ten largest suppliers as holding a combined 4.89% share under its specific market measurement methodology. That estimate indicates a fragmented supplier environment, but it should not be interpreted as a universal measure across differently defined market studies. Competitive analysis also requires distinguishing general-purpose AI providers from companies supplying dedicated materials discovery solutions.

List of Top AI In Material Discovery Companies

• Google LLC (U.S.)

• Microsoft Corporation (U.S.)

• IBM Corporation (U.S.)

• Schrödinger Inc. (U.S.)

• Dassault Systèmes SE (France)

• BASF SE (Germany)

• Citrine Informatics Inc. (U.S.)

• MaterialsZone Ltd. (Israel)

• Aionics Inc. (U.S.)

• Enthought Inc. (U.S.)

• XtalPi Inc. (China)

• Iktos SA (France)

• Atinary Technologies GmbH (Germany)

• Kebotix Inc. (U.S.)

Leading Companies by Market Share

- Schrödinger Inc. — 0.97%: Identified as the leading company in a published competitor-share analysis using a narrowly defined AI materials discovery market methodology. The company provides physics-based computational modeling and molecular simulation capabilities relevant to materials research. Its platform portfolio supports scientific modeling workflows used in industrial and academic research.

- Dassault Systèmes SE — 0.91%: Ranked second in the same competitive-share analysis. Its BIOVIA solutions support scientific modeling, laboratory information management, and materials-related computational workflows. The company's established engineering software ecosystem supports integration between scientific research and industrial product development.

Investment Trends & Opportunities

Investment activity in the AI In Material Discovery Market increasingly focuses on scientific software infrastructure, autonomous laboratories, and specialized computational models. Technology developers prioritize platforms capable of translating scientific objectives into structured research workflows. Investment decisions also consider integration with existing laboratory equipment and experimental information systems. Companies offering measurable improvements in experimental productivity are positioned to attract industrial interest.

Materials informatics platforms represent an important area of strategic development. Manufacturers require systems that organize proprietary formulations, process information, and experimental measurements. AI models can extract relationships from these datasets and recommend subsequent research activities. Enterprise integration and data security influence commercial adoption.

Autonomous laboratory infrastructure represents another area of investment. Robotic platforms, intelligent experimental scheduling, and automated analytical instruments enable iterative material development. Atinary's Boston laboratory demonstrates commercial interest in connecting algorithmic experiment selection with physical execution. Such facilities can serve research organizations requiring integrated computational and experimental capabilities.

Scientific AI also creates opportunities for partnerships between technology suppliers and industrial materials manufacturers. Software companies contribute computational expertise, while industrial partners provide application requirements and experimental datasets. Collaborative development arrangements help align model design with practical manufacturing considerations. These partnerships support specialized solutions rather than exclusively general-purpose AI platforms.

Investment evaluation increasingly emphasizes scientific validation and reproducibility. Buyers may assess prediction error, experimental success rates, integration requirements, and model reliability across unfamiliar materials. Organizations also consider intellectual property protection when sharing proprietary research data. These requirements encourage providers to establish transparent model evaluation and secure computing environments.

Product Innovation & Development

Product innovation in the AI In Material Discovery Market increasingly emphasizes accessible scientific interfaces and automated research recommendations. Materials informatics platforms are expanding beyond data storage toward decision-support functions. Developers incorporate natural language interactions, predictive modeling, and experiment planning capabilities into integrated environments. These systems aim to make computational analysis accessible to researchers without extensive machine learning programming experience.

Citrine Informatics introduced Catalyst and Apex in July 2026 to support AI-based materials and chemical product development. These capabilities help research teams translate product objectives into suggested experiments and improve predictive modeling workflows. The company reported 550 AI models deployed monthly and 70000 experiment suggestions generated monthly across customer activity at the time of its announcement. These operational figures illustrate the scale at which specialized materials informatics platforms can support research decisions.

MaterialsZone introduced Maven in April 2026 as a conversational interface for its enterprise materials platform. The system enables users to search, visualize, and analyze organizational materials information using natural language interactions. By connecting internal research records with external information, such interfaces can improve access to previously generated scientific knowledge. This approach addresses the challenge of fragmented experimental information across departments.

Product development also includes advances in structure-generation systems and atomic simulation models. Developers are improving methods that incorporate desired properties directly into candidate generation. Scientific evaluation increasingly combines theoretical stability, structural characteristics, and experimental feasibility. Reliable integration between these functions remains an important development objective.

Recent Developments (2023–2026)

• In July 2026, Citrine Informatics launched Catalyst and Apex to improve AI-supported formulation development, predictive modeling, and experimental recommendations for materials and chemical research teams.

• In April 2026, MaterialsZone introduced Maven, a conversational AI interface designed to help industrial researchers retrieve and analyze connected materials datasets through natural language queries.

• In February 2026, Atinary announced its Boston AI-powered laboratory integrating autonomous experimental platforms with robotics and analytical equipment for materials and chemistry research.

• In January 2025, Microsoft introduced MatterGen research demonstrating a generative modeling framework designed to produce inorganic material structures subject to selected property requirements.

• In October 2024, Meta researchers released the Open Materials 2024 dataset and pretrained models to support scientific machine learning across inorganic materials structures and density functional theory calculations.

• In November 2023, Google DeepMind presented GNoME, a graph neural network system that identified 2.2 million candidate crystal structures and expanded computational investigation of stable inorganic compounds.

Scope of the AI In Material Discovery Market Report

The AI In Material Discovery Market Report examines the technologies, materials categories, scientific applications, and industrial research activities shaping AI-assisted materials development. The analysis covers machine learning platforms, computational modeling solutions, specialized research hardware, and related technical services. It evaluates how these technologies support candidate identification, property prediction, simulation, and experimental decision-making. The report also considers the importance of scientific validation and reproducible research outcomes.

Market segmentation covers software, hardware, and services, with detailed consideration of their respective functional roles. Materials analysis includes chemicals, polymers, metals and alloys, ceramics, composites, nanomaterials, semiconductors, and biomaterials. Technology coverage examines machine learning, deep learning, generative AI, natural language processing, and predictive analytics. Application assessment includes material property prediction, molecular modeling, pharmaceutical materials, and battery-related materials research.

Geographic coverage includes North America, Europe, Asia-Pacific, and the Middle East and Africa. The analysis considers differences in research infrastructure, industrial specialization, scientific collaboration, and technology adoption. Competitive assessment addresses established technology companies and specialized materials informatics providers. It also reviews their product capabilities, development priorities, and positioning within the materials research ecosystem.

The report examines investment activity, product innovation, scientific challenges, and emerging development opportunities. Particular attention is given to data availability, computational reliability, experimental validation, and industrial deployment requirements. Recent company activities and technological developments are assessed to identify changes in research capabilities. This coverage provides a structured basis for evaluating AI applications throughout the materials discovery process.

Pasted text(20261010-050757).txtDocument

Worked for 17s

Unique Information

The AI In Material Discovery Market covers artificial intelligence technologies used to identify, evaluate, and optimize materials with targeted physical, chemical, and structural properties. These technologies combine materials informatics, machine learning, molecular modeling, and computational simulation to support scientific discovery. Researchers use AI systems to investigate atomic structures, predict material behavior, and prioritize candidates for experimental testing. Applications include advanced batteries, specialty chemicals, semiconductors, polymers, and engineered materials.

Scientific progress has demonstrated the scale of AI-assisted computational screening. Google DeepMind's GNoME project identified 2.2 million candidate crystal structures, including 380000 structures predicted to be stable. These findings expanded the available pool of computationally identified inorganic materials. However, experimental validation remains necessary before theoretical candidates can be considered suitable for manufacturing or commercial applications.

The USA AI In Material Discovery Market benefits from advanced scientific computing infrastructure, national laboratories, and established technology developers. American research institutions apply machine learning to discover materials for energy storage, electronics, aerospace engineering, and chemical manufacturing. The United States accounted for 31.8% of the global market in 2025, according to published industry estimates. Software platforms represent the dominant technology component within the country's materials discovery ecosystem.

American laboratories are also developing autonomous experimentation systems that combine robotic equipment with computational decision-making. Lawrence Berkeley National Laboratory's A-Lab demonstrated the synthesis of 41 new materials through an autonomous experimental program. Such systems connect predictive algorithms with physical research equipment. This integration supports continuous experimental learning and provides valuable information for improving subsequent scientific predictions.

The European AI In Material Discovery Market benefits from established capabilities in chemicals, automotive engineering, industrial manufacturing, and scientific software. Germany, France, Switzerland, and the United Kingdom are important centers for materials research and computational engineering. Europe accounted for 22.2% of the global market in 2025, according to published market estimates. Industrial adoption is supported by research institutions and manufacturers seeking improved material performance and efficient development workflows.

European technology companies provide molecular simulation software, scientific information management systems, and industrial modeling platforms. Dassault Systèmes supports computational research through its BIOVIA portfolio, while BASF investigates digital technologies for chemical and materials development. European research programs also address sustainable polymers, battery materials, and industrial catalysts. Regulatory requirements concerning chemical safety and environmental performance further encourage structured data management and scientifically validated material selection.

AI In Material Discovery Market Latest Trends

The AI In Material Discovery Market is experiencing increasing adoption of generative models capable of designing candidate materials according to selected functional properties. Conventional computational approaches primarily evaluate structures supplied by researchers. Generative systems instead create possible compositions and atomic configurations using learned patterns and defined scientific constraints. This transition supports exploration of material combinations that may not have appeared in existing experimental databases.

Microsoft Research introduced MatterGen in January 2025 as a generative AI framework for inorganic materials design. The system uses diffusion modeling to generate crystal structures conditioned on desired material characteristics. Such approaches support inverse design, where researchers begin with a performance objective and investigate structures that may satisfy it. Experimental testing remains essential for confirming generated candidates.

Scientific foundation models represent another important technological trend. Developers train models using extensive collections of atomic structures and computational calculations to create reusable scientific representations. Meta's Open Materials 2024 dataset contains 110 million density functional theory calculations, supporting the development and evaluation of machine learning systems for atomic modeling. These resources contribute to research involving structural stability and material energy calculations.

Autonomous laboratories are also becoming increasingly relevant. They connect experiment planning algorithms with robotic sample preparation, analytical equipment, and automated data collection. Active learning methods use previous results to recommend informative subsequent experiments. This creates an iterative research process in which computational recommendations are updated through physical observations.

Materials companies are additionally adopting conversational AI interfaces for scientific data exploration. These tools enable researchers to retrieve formulation records, compare experiments, and investigate technical relationships using natural language. Integration with enterprise databases is important because industrial materials knowledge often exists across separate departments and research systems.

AI In Material Discovery Market Dynamics

The AI In Material Discovery Market is influenced by the need for efficient scientific experimentation, specialized material performance, and increasingly complex research requirements. Materials engineering frequently involves evaluating multiple chemical compositions and processing conditions before identifying suitable candidates. Artificial intelligence helps researchers examine relationships within existing experimental and computational datasets. These capabilities support better selection of research priorities and reduce dependence on completely sequential experimentation.

Technology adoption also depends on data accessibility, scientific reliability, and compatibility with established research environments. Organizations require algorithms that can incorporate laboratory measurements, simulation outputs, and manufacturing specifications. Successful implementation involves coordination between computational scientists, materials engineers, and laboratory specialists. Market development therefore depends on both software improvements and the ability to demonstrate reproducible scientific outcomes.

Driver

Increasing demand for faster materials screening and predictive scientific modeling.

Growing demand for efficient materials research is a major driver of the AI In Material Discovery Market. Traditional discovery programs require extensive synthesis, characterization, and testing activities. AI-based predictive models help scientists evaluate potential material characteristics before conducting physical experiments. This capability is particularly valuable when candidate materials require complex preparation procedures or specialized testing equipment.

Industrial manufacturers increasingly investigate materials that satisfy several performance requirements simultaneously. Battery developers evaluate electrochemical stability, ion transport, and safety characteristics. Semiconductor researchers investigate electrical behavior, crystalline defects, and structural compatibility. AI algorithms assist these activities by identifying relationships between chemical composition and targeted properties.

Scientific simulation also supports research productivity by generating structured information about material behavior. Machine learning models can use these results to identify promising directions for further investigation. Integration between computational modeling and experimental planning improves the organization of materials development workflows.

Restraint

Limited availability of standardized and experimentally validated materials datasets.

Inconsistent scientific information remains a significant restraint on AI adoption in materials discovery. Experimental records originate from different instruments, laboratories, and testing procedures. Variations in sample preparation, environmental conditions, and measurement techniques can affect recorded properties. These differences create difficulties when combining datasets for model training.

Materials databases also contain uneven coverage across chemical compositions and structural families. Well-studied substances may have extensive characterization records, while emerging compounds have limited reliable information. Algorithms trained on narrow datasets can perform poorly when applied to unfamiliar material classes. Consequently, model accuracy depends on careful data selection and scientific validation.

Industrial research organizations also maintain proprietary formulation information that cannot be freely exchanged. Data confidentiality requirements restrict collaborative model development and external benchmarking. Establishing reliable materials informatics infrastructure requires standardized metadata, controlled access, and consistent experimental documentation.

Opportunity

Expanding application of artificial intelligence in sustainable materials and energy technologies.

Sustainable materials engineering presents important opportunities for the AI In Material Discovery Market. Researchers investigate alternative battery materials, recyclable polymers, low-impact chemicals, and advanced catalysts. These applications require simultaneous evaluation of functional properties and environmental considerations. Computational screening helps identify material candidates that satisfy selected engineering requirements.

Energy storage research presents particularly diverse technical challenges. Electrode compositions, electrolytes, protective coatings, and interface materials must operate together under defined conditions. AI models support the investigation of chemical compatibility and possible structural changes. Experimental analysis then determines whether predicted advantages can be demonstrated under realistic operating conditions.

Industrial sustainability programs also encourage chemical substitution and improved resource efficiency. Manufacturers seek alternatives to substances affected by supply limitations or regulatory requirements. Materials informatics systems can organize historical formulations and suggest potential substitutes based on available scientific evidence. These applications create opportunities for specialized predictive software and research services.

Challenge

Translating AI-generated materials predictions into reproducible industrial applications.

A major challenge involves converting computationally promising candidates into experimentally verified materials. Algorithms may predict favorable properties for structures that are difficult to synthesize. Certain compounds require demanding pressure, temperature, or processing conditions. Others may exhibit unexpected behavior when produced outside controlled laboratory environments.

Industrial qualification requires assessments of manufacturing consistency, material degradation, and long-term performance. A candidate showing desirable theoretical properties must also satisfy practical production requirements. Researchers therefore need methods for connecting predicted characteristics with synthesis feasibility and process stability.

Algorithmic uncertainty creates another technical challenge. Models can generate predictions outside the chemical environments represented in their training data. Reliable deployment requires uncertainty estimation, independent testing, and continuous model evaluation. Collaboration between AI developers and experimental scientists remains essential for industrial implementation.

AI In Material Discovery SWOT Analysis

Strengths

• Accelerated computational screening: AI technologies help prioritize materials from extensive candidate libraries.

• Multivariable optimization: Algorithms support evaluation of composition, structure, and processing relationships.

• Scientific data integration: Materials informatics platforms organize experimental and computational information.

• Research automation compatibility: Predictive models can guide robotic experimentation and laboratory workflows.

Weaknesses

• Dependence on data quality: Incomplete experimental records can reduce prediction reliability.

• Model interpretation difficulties: Complex neural networks may produce outputs that require additional scientific explanation.

• Specialized technical requirements: Deployment requires expertise in materials science, computational modeling, and software engineering.

• Integration complexity: Existing laboratory systems may use incompatible information structures and communication protocols.

Opportunities

• Advanced battery materials: AI supports research into electrodes, electrolytes, and electrochemical interfaces.

• Recyclable material development: Computational screening assists investigation of alternative polymers and industrial compounds.

• Autonomous scientific laboratories: Robotics and adaptive algorithms enable integrated experimental optimization.

• Collaborative research platforms: Shared scientific infrastructure can support institutions with limited internal computing capabilities.

Threats

• Intellectual property uncertainty: Ownership disputes may arise from AI-generated discoveries and proprietary model training.

• Cybersecurity risks: Digitized laboratory records and confidential formulations require effective protection.

• Scientific reproducibility concerns: Poorly validated predictions can undermine confidence in computational research.

• Rapid technological replacement: Advances in scientific AI may reduce the useful lifespan of existing software architectures.

AI In Material Discovery Segmentation Analysis

The AI In Material Discovery Market is segmented by component, material type, technology, application, and end use. These classifications reflect differences in technical infrastructure, material characteristics, scientific methods, and industrial requirements. Component segmentation distinguishes software, hardware, and services, while material categories identify the substances investigated. Technology segmentation covers the computational techniques used for discovery and prediction.

Application segmentation explains how artificial intelligence supports different stages of materials research. End-use classifications identify the industries adopting these capabilities for scientific development. The categories overlap operationally because an individual research project may involve several technologies and multiple material characteristics. Separate segmentation dimensions therefore describe different aspects of the same research activity rather than mutually exclusive scientific workflows.

By Component

Software accounted for 55.0% of the global AI In Material Discovery Market in 2025, making it the leading component category in a published market assessment. Software solutions include predictive modeling platforms, materials simulation tools, data analytics systems, molecular design software, and materials informatics platforms. These applications support scientific data organization, computational analysis, and candidate evaluation. Integration with established research workflows influences their practical adoption.

Hardware includes high-performance computing systems, graphics processing units, specialized accelerators, data storage servers, and computational modeling workstations. These systems provide processing capacity for demanding simulations and scientific AI models. Hardware selection depends on computational workload, infrastructure requirements, and data processing needs.

Services include consulting and integration, custom model development, data management services, simulation and testing services, and training and support. Specialized providers help research organizations configure software and implement relevant scientific workflows. Their activities also support ongoing system evaluation and technical maintenance.

By Material Type

Chemicals represented 30.0% of the global market in 2025 under a published materials-based segmentation framework. Chemical materials research involves predicting molecular properties, optimizing formulations, and investigating catalytic performance. Polymers require examination of molecular structure, processing behavior, durability, and physical characteristics. Materials informatics can connect composition records with experimental performance measurements.

Metals and alloys involve relationships between elemental composition, crystalline structure, and mechanical properties. Ceramics and composites require investigation of structural stability and complex processing conditions. AI assists researchers in identifying parameters that influence material behavior.

Nanomaterials present additional challenges involving particle characteristics and surface interactions. Semiconductor materials require precise evaluation of electronic properties and crystalline defects. Biomaterials research involves compatibility with biological systems and specialized application requirements. Each material category requires suitable scientific descriptors and experimental verification methods.

By Technology

Machine learning accounted for 36.0% of the global AI In Material Discovery Market in 2025 according to published industry estimates. Machine learning models use structured scientific information to identify relationships between materials and measured properties. Applications include classification, regression, candidate ranking, and optimization. Model selection depends on available data and the physical characteristics being investigated.

Deep learning supports the analysis of complex atomic and molecular structures. Neural network architectures can learn representations from extensive scientific datasets without relying exclusively on manually constructed descriptors. Graph-based models are particularly relevant when chemical connectivity influences predicted material behavior.

Generative AI supports the creation of candidate compositions and structures. Natural language processing helps extract useful information from scientific literature, patents, and technical documents. Predictive analytics supports evaluation of material behavior using historical records and computational results. These technologies are frequently combined within integrated research environments.

By Application

Material property prediction is the leading application category in published assessments of the AI In Material Discovery Market. A separate industry segmentation estimates the property prediction workflow at 32% of the market in 2025. This application involves estimating thermal, electrical, mechanical, and chemical characteristics. Predictions help researchers decide which candidates require detailed experimental evaluation.

Molecular modeling and simulation investigate structural interactions and energy-related characteristics. Computational systems can assess candidate behavior under defined conditions and generate information for further scientific analysis. These techniques are particularly useful when experiments cannot directly observe atomic-scale processes.

Drug and pharmaceutical material discovery involves crystalline forms, solid-state characteristics, and formulation-related materials. Battery and energy storage material discovery focuses on electrodes, electrolytes, and interface components. Both applications require careful alignment between model predictions and experimentally measured performance.

By End Use

Pharmaceuticals and biotechnology accounted for 26% of the global market in 2025 under a published end-use industry assessment. Organizations in this category investigate molecular solids, formulation materials, and specialized chemical structures. AI supports scientific screening and the interpretation of complex experimental information. Material behavior must still be validated against application-specific requirements.

Chemicals and advanced materials companies use computational tools for formulation development and compound evaluation. Electronics and semiconductor manufacturers investigate specialized materials for electronic devices and manufacturing processes. Automotive companies evaluate battery-related components, coatings, and lightweight structures.

Energy and utilities applications include storage materials, catalysts, and energy conversion systems. Aerospace and defense organizations investigate materials designed for demanding mechanical and thermal environments. Research institutions contribute foundational scientific models and experimental datasets. These end users differ in their qualification procedures, operational constraints, and technical priorities.

Regional Analysis

Regional performance in the AI In Material Discovery Market reflects the distribution of research institutions, industrial manufacturing capabilities, and advanced computing infrastructure. North America leads the market through established scientific software companies and technology research facilities. Europe maintains a strong position through chemical engineering, computational science, and industrial materials development. Asia-Pacific benefits from semiconductor manufacturing, battery research, and expanding scientific computing capabilities.

The Middle East and Africa represent a smaller but developing market supported by selected energy research programs and academic institutions. Government science strategies and international collaborations influence the adoption of advanced computational technologies. Regional market shares vary by industry definition and methodology. The following geographic comparisons use a consistent published 2025 market assessment.

- North America

North America accounted for 41.6% of the global AI In Material Discovery Market in 2025. The region benefits from established technology companies, advanced research universities, and specialized scientific laboratories. The United States provides a substantial share of regional capabilities through its national laboratory system and commercial AI ecosystem. Canada contributes expertise in scientific machine learning and computational research.

Research organizations across North America investigate materials for batteries, semiconductor devices, industrial chemicals, and aerospace systems. These applications require computational tools capable of processing complex scientific information. Established cloud infrastructure and high-performance computing resources support large simulation workloads. Collaboration between academic institutions and private companies facilitates development of specialized research software.

North American organizations are also active in commercializing AI-based scientific research. University laboratories and technology startups develop modeling platforms that address specific material engineering challenges. Partnerships with manufacturers help evaluate technologies against industrial specifications. These relationships connect experimental research with practical product development requirements.

The United States remains the region's principal commercial center for AI materials discovery. Research institutions participate in computational chemistry, materials simulation, and laboratory automation programs. National scientific initiatives support access to technical expertise and specialized instruments. North America's established research ecosystem continues to provide a foundation for technology development and industrial adoption.

- Europe

Europe held 22.2% of the global AI In Material Discovery Market in 2025. The region benefits from expertise in industrial chemicals, specialized manufacturing, and computational engineering. Germany is an important center for automotive materials and advanced chemical research. France contributes scientific software development and computational modeling capabilities.

European manufacturers use materials informatics to support formulation research and product development. Digital systems organize laboratory results and facilitate comparison between different material compositions. Scientific modeling helps assess behavior under specified operating conditions. These capabilities are important for industries requiring controlled technical qualifications and consistent material performance.

European research programs also investigate environmentally compatible materials and resource-efficient industrial processes. Battery materials, catalysts, and alternative chemical formulations receive significant research attention. AI can support the identification of compounds that satisfy selected engineering constraints. Experimental testing provides the evidence required for industrial deployment.

Collaboration between universities, research institutes, and industrial manufacturers supports regional innovation. Organizations exchange technical knowledge through cooperative development programs and scientific partnerships. European technology providers also integrate materials simulation with broader engineering software environments. This improves connectivity between material investigation and downstream product development.

- Asia-Pacific

Asia-Pacific accounted for 27.8% of the global AI In Material Discovery Market in 2025. The region includes major semiconductor, battery, and advanced materials manufacturing economies. China, Japan, South Korea, and India contribute to research activities across several scientific disciplines. Industrial demand for specialized material performance supports the adoption of computational tools.

China maintains significant capabilities in battery materials, computational chemistry, and electronic materials research. Research institutions investigate alternative compositions and performance characteristics using simulation and machine learning. Industrial manufacturers also explore digital systems for formulation management and experimental analysis. The connection between scientific modeling and manufacturing remains an important regional capability.

Japan and South Korea contribute expertise in precision materials, advanced ceramics, semiconductor compounds, and electrochemical technologies. These sectors require detailed understanding of structural properties and manufacturing compatibility. AI-based analysis supports the investigation of candidate materials and process relationships. Technical qualification requirements continue to influence research priorities.

India is expanding its computational materials research through universities, scientific institutions, and technology organizations. Access to advanced computing and collaborative research programs supports scientific development. Asia-Pacific's manufacturing ecosystem provides opportunities to connect laboratory predictions with industrial testing. Such integration is important for translating computational discoveries into practical technologies.

- Middle East & Africa

The Middle East and Africa represented 3.9% of the global AI In Material Discovery Market in 2025. Regional capabilities are concentrated in selected research institutions and industrial development centers. Saudi Arabia and the United Arab Emirates are developing scientific and technological infrastructure relevant to computational materials research. Energy-related applications provide potential demand for specialized materials modeling.

Industrial research opportunities include catalysts, chemical separation materials, protective coatings, and energy storage components. These materials are relevant to demanding operating environments and industrial processing requirements. Artificial intelligence can assist researchers in evaluating candidate compositions before undertaking laboratory tests. Local experimental infrastructure remains important for confirming computational predictions.

Research institutions in the Middle East participate in scientific collaborations involving computational chemistry and advanced materials engineering. International partnerships enable access to technical expertise and specialized scientific methods. Investments in computing infrastructure support development of advanced analytical capabilities. Technology adoption depends on institutional priorities and the availability of qualified personnel.

African universities and scientific organizations contribute through computational materials science and academic research programs. Access to specialized instruments and consistent datasets can influence the scope of laboratory investigations. Collaborative research networks help institutions exchange information and develop technical expertise. The region presents opportunities for gradual adoption of scientific AI tools as infrastructure and research capabilities develop.

Competitive Landscape

The AI In Material Discovery Market includes major technology companies, computational chemistry developers, materials informatics specialists, and industrial research organizations. Competition focuses on predictive accuracy, scientific workflow integration, computing capabilities, and application-specific expertise. Established participants include Microsoft, Google, IBM, Schrödinger, and Dassault Systèmes. Specialized companies such as Citrine Informatics and MaterialsZone provide platforms designed for industrial materials research.

Major technology companies contribute foundational AI research and scalable computational infrastructure. Scientific software developers provide domain-specific simulation tools and analytical capabilities. Materials informatics specialists emphasize structured experimental data and practical formulation optimization. Competitive differentiation increasingly depends on whether these technologies produce useful recommendations that researchers can validate.

The market remains fragmented under a published competitor-share assessment. The ten leading suppliers collectively accounted for 4.89% of the market under that report's defined methodology. The same assessment identified Schrödinger and Dassault Systèmes as leading individual participants. These figures depend on the specific market boundary and should not be generalized to broader scientific software markets.

Technology providers also compete through partnerships with research institutions and industrial manufacturers. Collaboration provides access to experimental expertise and application-specific requirements. Companies capable of connecting algorithms with established scientific practices may offer stronger practical utility. Data protection, transparency, and consistent model performance remain important considerations for enterprise customers.

List of Top AI In Material Discovery Companies

• Google LLC (U.S.)

• Microsoft Corporation (U.S.)

• IBM Corporation (U.S.)

• Schrödinger Inc. (U.S.)

• Dassault Systèmes SE (France)

• BASF SE (Germany)

• Citrine Informatics Inc. (U.S.)

• MaterialsZone Ltd. (Israel)

• Aionics Inc. (U.S.)

• Enthought Inc. (U.S.)

• XtalPi Inc. (China)

• Iktos SA (France)

• Atinary Technologies GmbH (Germany)

• Kebotix Inc. (U.S.)

Leading Companies by Market Share

- Schrödinger Inc. — 0.97%: Schrödinger holds the leading position in a published competitor-share assessment of the AI materials discovery market. The company provides molecular simulation and computational chemistry software used in scientific research. Its technology combines physics-based modeling with advanced computational methods to support materials investigation.

- Dassault Systèmes SE — 0.91%: Dassault Systèmes ranks second under the same market-share assessment. Its BIOVIA portfolio includes scientific modeling and materials-related information management solutions. The company's engineering software capabilities support integration between computational research and industrial product development.

Investment Trends & Opportunities

Investment trends in the AI In Material Discovery Market increasingly emphasize specialized software platforms and integrated scientific research systems. Technology developers are advancing solutions that connect predictive modeling with experimental information management. Industrial manufacturers seek systems capable of supporting practical research decisions. This focus creates opportunities for software companies with strong materials science expertise.

Materials informatics represents an important investment area because industrial organizations maintain extensive proprietary experimental records. These datasets contain information on chemical compositions, process conditions, and measured material properties. AI platforms can organize existing knowledge and identify potentially useful relationships. Secure integration with enterprise information systems is therefore a major technical consideration.

Autonomous laboratories create opportunities for equipment manufacturers and scientific automation providers. Robotic sample handling, analytical instruments, and adaptive experiment planning systems can operate within integrated workflows. Atinary's Boston laboratory demonstrates the development of infrastructure combining physical experimentation with AI-based decision-making. Such facilities support research programs requiring repeated testing and systematic optimization.

Investment opportunities also exist in scientific foundation models and specialized computing infrastructure. Materials prediction systems can require substantial processing capacity and domain-specific training information. Developers must evaluate model performance against experimentally verified results. Commercial adoption depends on scientific reliability as well as computational capabilities.

Industry partnerships provide another pathway for developing specialized AI applications. Technology suppliers collaborate with chemical companies, battery developers, and advanced manufacturers to address particular research requirements. These relationships create opportunities for customized modeling and technical services. Organizations evaluating investments increasingly consider data ownership, reproducibility, integration requirements, and practical research outcomes.

Product Innovation & Development

Product innovation in the AI In Material Discovery Market increasingly focuses on simplifying access to advanced computational capabilities. Software developers are integrating predictive modeling, experiment recommendations, and scientific data management into accessible platforms. These improvements allow researchers to perform selected analytical tasks without building independent machine learning systems. Product development also emphasizes transparency in algorithm-generated recommendations.

Citrine Informatics launched Catalyst and Apex in July 2026 to support materials and chemical product development. Catalyst enables researchers to describe technical objectives and generate suggested experiments through an AI-assisted workflow. Apex strengthens the predictive modeling capabilities supporting those recommendations. The company reported 550 AI models deployed monthly across customer activities when introducing the products.

Conversational scientific interfaces represent another development direction. MaterialsZone introduced Maven in 2026 to help research teams explore connected materials information through natural language queries. The platform combines conversational capabilities with specialized analytical functions. This approach enables scientists to retrieve historical knowledge and investigate technical relationships without relying entirely on manual database navigation.

Developers are also improving algorithms designed to incorporate physical constraints into AI-generated material structures. These techniques seek to produce candidates with scientifically meaningful characteristics rather than unconstrained mathematical outputs. Evaluation systems increasingly incorporate model uncertainty and independent simulation results. Future product development will depend on reliable connections between computational predictions, laboratory verification, and industrial manufacturing requirements.

Recent Developments (2023–2026)

- In September 2026, Citrine Informatics announced a partnership with Forum Rathenau to support Germany's green deep-technology ecosystem through collaboration involving AI-based materials development.

- In July 2026, Citrine Informatics launched Catalyst and Apex, introducing AI-assisted scientific workflows and advanced predictive modeling capabilities for industrial materials and chemical product development.

- In May 2026, MaterialsZone introduced Maven, a conversational AI interface that enables research teams to explore enterprise materials information and analyze connected scientific datasets.

- In February 2026, Atinary announced its AI-powered laboratory in Boston, integrating autonomous experimental platforms with robotic systems and analytical instruments for advanced scientific research.

- In January 2025, Microsoft introduced MatterGen research, demonstrating a diffusion-based generative AI framework designed to create inorganic material structures with selected functional properties.

- In October 2024, Meta researchers released Open Materials 2024 and associated pretrained models to support machine learning research involving atomic structures and computational materials simulations.

Scope of the AI In Material Discovery Market Report

The AI In Material Discovery Market Report examines artificial intelligence technologies used in materials prediction, computational modeling, candidate identification, and experimental optimization. The study covers scientific software, specialized computing infrastructure, and implementation services. It evaluates technological capabilities and their relevance to different materials research activities. The analysis also considers scientific reliability and integration with established development workflows.

The segmentation framework includes component, material type, technology, application, and end use. Component coverage incorporates software, hardware, and services. Material categories include chemicals, polymers, metals and alloys, ceramics, composites, nanomaterials, semiconductors, and biomaterials. Technology coverage examines machine learning, deep learning, generative AI, natural language processing, and predictive analytics.

Application analysis addresses material property prediction, molecular modeling and simulation, pharmaceutical materials, and battery-related discovery. End-use coverage includes pharmaceuticals and biotechnology, chemicals and advanced materials, electronics and semiconductors, automotive, energy and utilities, aerospace and defense, and research institutions. These classifications help distinguish different research requirements and industrial applications.

The regional analysis covers North America, Europe, Asia-Pacific, and the Middle East and Africa. Competitive evaluation includes established technology companies, scientific software suppliers, and specialized materials informatics developers. The report examines market positioning, technological capabilities, and selected company developments. It also considers investment priorities, product innovation, and the operational challenges associated with deploying AI in scientific research.