← all systems · see it in The System
Materials science (the field)
Materials science is an interdisciplinary field concerned with understanding the relationships between the structure of materials and their properties and using this knowledge to design materials for specific applications.
Properties published scienceour proposal
Sourced fields, and our own readings.
| born at | practice: by the 5th millennium BC (lost-wax casting at Mehrgarh, 6,000 years ago; Balkan smelting late sixth millennium BC); field: 1940s-1955 |
| born precision | millennia for the practice; a decade for the field |
| boundary | Inside: the study of how processing sets a material's structure, how structure sets its properties and how properties set its performance in service; characterization (diffraction, microscopy, spectroscopy, mechanical and thermal tests); the thermodynamics and kinetics that explain it; computation and materials databases; extraction, synthesis and processing; failure analysis. Also booked, as what the field works on: the materials themselves (metals and alloys, ceramics and glasses, polymers, composites, semiconductors, biomaterials, nanomaterials). Outside: the devices and structures made from them once released (engineering's walks), the pure physics and chemistry it draws on, and the people, funders and markets it serves. |
| nested level | A human practice (rung 11) carried by materials scientists, departments, laboratories, societies and databases; embedded in science and engineering and funded by states and industry; the materials it works on sit at rung 4 (solids, crystals, minerals) and up, and the technologies they enable at rung 12. |
| scale | social |
Measures (35) published science
| A-Lab campaign | 17 days (353 experiments) |
| A-Lab synthesis rate | >2 new materials per day |
| A-Lab success rate | 63 (36 of 57) % |
| A-Lab failures from sluggish kinetics | 11 of 17 failed targets (steps under 50 meV per atom) |
| A-Lab re-analysis after correction | 36 of 40 reported successes confirmed (4 inconclusive) |
| Claim to correction | 29 November 2023 to 2026 dates |
| GNoME newly stable materials | 381,000 crystals (of 2.2 million below the previous hull) |
| Stable crystals before GNoME | 48,000 computationally stable materials |
| GNoME hit rate with structure | >80 (from 1) % |
| GNoME structures independently realized | 736 structures |
| Layered materials stable | about 1,000 to about 52,000 materials |
| Metals studied for substitution | 62 metals |
| Metals without adequate substitutes | a dozen metals |
| Energy-intensity reduction achievable | 50-56 (goal 75) % by 2050 with aggressive strategies |
| Nacre toughness over its constituents | three orders of magnitude (energy terms) |
| Nacre-like alumina flexural strength range | 89-800 MPa (toughness 4 to more than 9 MPa m^1/2) |
| Freeze-cast alumina-PMMA | ~225 MPa strength; >30 MPa m^1/2 toughness at 80 vol.% ceramic |
| RAFM Stage 3 total elongation | 49 % |
| RAFM embrittlement windows | 250-350 and 550-650 °C |
| Steel carbon range | 0.01-2.00 wt% C |
| Cast iron carbon range | 2.00-6.67 wt% C |
| Stainless steel chromium | >10 wt% Cr |
| Nanostructure range | 1-100 nm |
| Microstructure range | 100 nm to a few cm length |
| Grain size where Hall-Petch strengthening ends | about 10 nm |
| Grain size of maximum strength (simulation) | about 50 atoms across (grain-boundary thickness 2-6 particles) |
| Age of the Mehrgarh lost-wax amulet | 6,000 years |
| Copper melting in the amulet's casting | above 1,085 °C |
| Lead added by Baluchistan founders | 10-30 wt% Pb |
| Illimani copper emission record | 6,500 years |
| Illimani Cu enrichment factor, late 20th century | 16.4 enrichment factor |
| First materials science department | 1955 year (Northwestern University) |
| Semiconductor market | 530 billion US$ (2021, estimated) |
| Reinforced carbon-carbon service temperature | up to 1510 °C |
| Glass: fracture stress against theoretical bond-breaking stress | about 100 against about 10,000 MPa |
Open contentions (5) open contention
Where sources or readings disagree. Both stay on the map.
- C-413 Where do thoughts come from?
booked: No theory is proven, so the map lists the main ones side by side and doesn't pick one. It shows where thinking is handled in the brain, not where thoughts start.
the other reading: positions listed in the Human walk - C-2391 Are AI-predicted materials really new?
booked: The map keeps the claim and the critique.
the other reading: the authors: 381,000 newly stable materials, an order-of-magnitude expansion (Merchant 2023); 41 novel compounds from 58 targets (A-Lab, as first published); the critics: scant evidence of compounds that are novel, credible and useful; many predicted orderings unlikely (Cheetham and Seshadri 2024); the correction: new to the prediction platform, not necessarily new to science; 36 of 40 confirmed, 4 inconclusive (2026) - C-2392 When did copper smelting begin in the Andes?
booked: The map keeps the competing dates.
the other reading: ~2000 BC, from peat-bog pollution in southern South America; first sophisticated copper metallurgy with the Moche ~200-800 AD (archaeology); molten copper already ~1000 BC (Malpaso bead), with smelting 'very likely' soon after - C-2421 Why does metal get weaker at the smallest grain sizes?
booked: The map keeps the competing mechanisms.
the other reading: mechanism: grain-boundary sliding once only one or two dislocations fit (encyclopedia); mechanism: bond-order-length-strength correlation and a soft quasi-solid phase (Sun 2008, preprint); the Hall-Petch relation itself is an empirical fit whose pile-up mechanism has not been seen directly - C-2427 When did materials science begin?
booked: The map keeps the competing starts.
the other reading: the field proper begins in the 1940s-1950s, when it is recognised as distinct (first department, Northwestern 1955); its intellectual origins are in the Enlightenment, when analysis was applied to ancient phenomenological observations; which smelting came first is itself open: lead smelting may have predated copper smelting (Metallurgy)
Cross-scale patterns seen here (20) our proposal
Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Materials science (the field). Each links to every other system that shows it.
- same role, different builds (structural)
One job, several builds: holding a drink and keeping its CO2 in is done by glass, aluminium alloy and polyethylene, each with its own bonds and structure (encyclopedia's beverage containers); stopping dislocations (strength) is done by grain boundaries (Hall-Petch), alloying pins, work hardening and - same toolkit, independent builds (structural)
Copper metallurgy built more than once from the same parts (ore, fire, charcoal): the Carpatho-Balkan province 'emerged independently from centers of the Proto-Metal area' (Metallurgy encyclopedia page), and the Andes developed large-scale copper smelting with its own furnaces, recorded from ~700-50 - the whole sets the local reference (structural)
A compound's stability is read against the convex hull formed by all competing phases, so the whole set of known materials fixes each one's reference; adding materials moves the hull: 'Given that discovered materials compete for stability, the updated convex hull consists of 381,000 new entries for - trade one property for another (structural)
Strength bought with ductility and toughness. 'As the yield stress is raised, tensile ductility and the ability to blunt a crack generally drops, which gives the well-known strength-ductility trade-off' (Gong 2025); in steel, more carbon gives more hardness and strength with lower ductility and toug - borrowed-then-held (partial)
Reduction of metal first done by geology, later held in the furnace: cold metallurgy with native copper (7th-6th millennia BCE, Anatolia and Iraq) came before smelting (late sixth millennium, Balkans) (Metallurgy encyclopedia page); the Mehrgarh amulet's 'very pure copper, possibly native copper' (T - conservation law plus closure (partial)
Thermodynamics (energy above the hull, phase diagrams) says which phase should form but not whether it will: kinetics closes the prediction. The A-Lab's predicted-stable targets failed on sluggish kinetics; corrosion 'thermodynamically favorable' can still be slow ('Some metals have naturally slow r - lightning (partial)
Brittle fracture as a discharge: ceramics 'invariably display near-zero tensile ductility and low fracture toughness values, which makes them prone to sudden catastrophic failure' (Wat 2019); Griffith's crack runs once the stress passes a level set by the flaw (Fracture mechanics). The released frac - namespace completeness (the map predicts its gaps) (partial)
The map predicts its gaps: the computed hull predicts stable compounds not yet made, each with a predicted structure and energy; GNoME lists 381,000 new stable crystals and reports 736 'independently experimentally realized'. Contested: Cheetham and Seshadri find 'scant evidence' for compounds that - partner-becomes-part (partial)
Separate fields fused into one: materials science formed by 'renaming and/or combining existing metallurgy and ceramics engineering departments; splitting from existing solid state physics research...; pulling in relatively new polymer engineering and polymer science' (encyclopedia), the former depa - runaway from an unstable balance (partial)
Griffith's balance: below a critical stress a flaw is stable, above it the crack 'rapidly propagate[s]' (Fracture mechanics), the energy released feeding further growth. The balance point is named (the Griffith criterion) and the runaway is the fracture. - same shape, opposite sign (partial)
One process, polarity reversed: impressed-current protection of a metal structure, cathodic or anodic, 'Anodic protection impresses anodic current on the structure to be protected (opposite to the cathodic protection)' (Corrosion). Checked and not booked as this shape: Hall-Petch versus inverse Hall - templating (partial)
A module moved unchanged in design: nacre's brick-and-mortar layout copied into alumina with polymer (PMMA) or metallic-glass mortar and still toughening (Wat 2019); the tetrahedron reused as the frame for new 'tetrahedra' for AI research (Zhang 2026). - The carry (partial)
The wait, and a lookahead. In the A-Lab, 'Sluggish reaction kinetics hindered 11 of the 17 failed targets, each containing reaction steps with low driving forces (<50 meV per atom)': the slow step sets whether the target arrives at all; regrinding and reheating to higher temperatures carried two mor - The Payment to Entropy (partial)
Material and thermal forms. Material: smelting pays a toll of metal to the air (Illimani copper enrichment highest, 16.4, in the late 20th century with large-scale production; Eichler 2017) and metal life cycles have 'rates of recycling and loss' (Graedel 2015); corrosion returns metal toward the ox - Umwelt: each system carves its own slice of one shared medium (partial)
Each probe carves its own slice of one sample: diffraction reads long-range order, microscopy the microstructure, spectroscopy composition, thermal analysis transitions; their limits show where one slice is not enough: X-ray diffraction alone left four A-Lab products 'inconclusive', and computed ord - who opens the door (partial)
At the ceramic-metal crossing in nacre-like composites the door is opened by the receiver's own chemistry: metals 'do not typically wet ceramics', and reactive wetting of a zirconium-based metallic glass lets it infiltrate porous alumina (Wat 2019). In the A-Lab, the door is a port built ahead of ti - coordination-before-mapping (surface)
Craft before theory: ancient metallurgy and ceramics were run on 'phenomenological observations' that Enlightenment analysis later explained (encyclopedia); Baluchistan founders found lead's effect on fluidity by looking for improvements, with no model of why. - seed (surface)
Seed check: maybe. Minute additions open large families: 'steels are classified based on 1/10 and 1/100 weight percentages of the carbon and other alloying elements', and semiconductor conductivity is 'very sensitive to the concentration of impurities' (encyclopedia). Combinatorics unpack from small - smooth to sharp in finite time (surface)
In parameter, not time: 'Theoretically, the stress ahead of a sharp crack tip becomes infinite and cannot be used to describe the state around a crack' (Fracture mechanics), which is why the field uses a single parameter (the stress intensity factor) instead. - symmetry breaking (name the cut) (surface)
Cooling cuts a melt two ways: crystals choose orientations (most materials are polycrystalline aggregates of grains 'with different orientations'), while glass 'solidifies into a disordered state upon cooling' and breaks no long-range symmetry (encyclopedia). The before and the cut (cooling) are nam
Valence: what it trades, and with whom (21) published science
Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.
- Engineering design process via sys:sensory-environment · Materials science (the field) even
- Mechanical engineering (the field) via sys:written-record · Materials science (the field) gain
- Human (Homo sapiens) via sys:sensory-environment · Materials science (the field) gain
- Engineering (design) via sys:written-record · Materials science (the field) gain
- Industrial Revolution (the first, c. 1760 to c. 1840) via sys:matter · Materials science (the field) gain
- Mineral via sys:matter · Materials science (the field) gain
- Chemistry (the field: the practice of chemists and the structure of substances and their reactions it works on) via sys:written-record · Materials science (the field) gain
- Crystal via sys:matter · Materials science (the field) gain
- Chemical bond via sys:electromagnetic-field · Materials science (the field) gain
- Machine learning via sys:computer · Materials science (the field) gain
- Cold War via sys:human-society · Materials science (the field) gain
- Economy via sys:human-society · Materials science (the field) gain
- Pollution via sys:natural-environment · Materials science (the field) gain
- Tokamak via sys:matter · Materials science (the field) gain
- Physics (the field) via sys:written-record · Materials science (the field) gain
- Thermodynamics via sys:written-record · Materials science (the field) gain
- Condensed matter physics via sys:written-record · Materials science (the field) gain
- Semiconductor via sys:matter · Materials science (the field) gain
- Crystallography via sys:electromagnetic-field · Materials science (the field) gain
- Nacre (bioinspiration donor) via sys:matter · Materials science (the field) gain
- Steel via sys:matter · Materials science (the field) gain
Partners and parts published science
Built from / builds (8)
- Materials characterization
- Synthesis and processing
- Computational materials science
- The materials paradigm (tetrahedron)
- Materials databases (Materials Project, ICSD, GNoME listings)
- Metallurgy (physical and extractive)
- Failure analysis and forensic materials engineering
- A metal part against the environment's oxidation (corrosion)
Sits inside / holds (3)
- Science (as a human social and cultural system)
- Engineering (design)
- Human society (from local community up to societies)
expressed by (21)
- Generation/Synthesis
- Recombination/Composition
- Scaffolding/Enablement (Meta-Tooling)
- Templating / Standardization
- Niche Construction (alter environment)
- Quality Improvement/Refinement
- Interoperability/Transduction (Bridging)
- Amplification/Scale-Up (Manufacture & Distribution)
- Symbiosis/Co-Creation (Collaborative Specialization)
- Sensing/Scanning
- Search/Foraging
- Mapping/Modeling
- Uncertainty Reduction/Learning
- Active Experimentation / Probing (Optimal Experimental Design)
- Multi-Scale Data Fusion / Triangulation
- Trail-Marking / Externalized Memory (Stigmergy)
- Offline Simulation & Replay-Guided Planning
- Social Learning / Public Information Use
- Physical/Virtual Barrier
- Validation of Effectiveness
- Quality Control & Error Correction
floor filled by (22)
- Sculpture
- Human (Homo sapiens)
- Written record (durable external marks: texts, archives, canon)
- Engineering (design)
- Mineral
- Matter
- Crystal
- Chemical bond
- Machine learning
- Crystallographic defect
- Thermodynamics
- Condensed matter physics
- Crystallography
- Autonomous laboratory
- Natural environment
- Microstructure
- Materials characterization
- Synthesis and processing
- Computational materials science
- The materials paradigm (tetrahedron)
- Materials databases (Materials Project, ICSD, GNoME listings)
- Steel
represents (1)
Kinds (9)
- Computational materials science
- Metallurgy (physical and extractive)
- Ceramic engineering
- Polymer science and engineering
- Composite engineering
- Materials informatics and selection
- Materials physics
- Biomaterials science
- Nanomaterials research
Notes published science
- The oldest known lost-wax casting is a 6,000-year-old copper amulet from Mehrgarh, Pakistan, so corroded it is all oxide, yet light shone on it still shows the ghost of the metal's cooling pattern, telling how it was cast (src:thoury-2016-mehrgarh).
- A glacier in Bolivia keeps a 6,500-year diary of copper smelting in the air: it puts large-scale Andean smelting about 2,700 years ago, and its copper signal is highest of all in the late 20th century (src:eichler-2017-andes).
- Mother-of-pearl is about a thousand times tougher, in energy terms, than the mineral and protein it is made of; engineers copy its brick-and-mortar layout, with a metallic glass as the mortar, to make ceramics that crack less suddenly (src:wat-2019-nacre).
- A robot lab announced 41 new compounds in 17 days; after other chemists checked, the paper was corrected in 2026 to say they were 'new to the prediction platform, not necessarily new to science' (src:szymanski-2023-alab, src:szymanski-2026-alab-correction, src:cheetham-seshadri-2024).
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
a material from its setting to its service (the reader's view of a material: setting and loads, bonding, atomic arrangement and defects, microstructure, properties, performance and failure)
- L1: The setting: the elements on hand, the heat and atmosphere of processing, the loads, temperatures, radiation and chemistry of service
- L2: Atoms and bonding: metallic, ionic, covalent and secondary bonds; composition down to trace elements
- L3: Atomic arrangement: crystal lattice or amorphous network, with point, line and planar defects (vacancies, dislocations, grain boundaries)
- L4: Microstructure: grains, phases, precipitates, fibres and layers (100 nm to a few cm), and in composites the bricks and mortar
- L5: Properties: strength, ductility, toughness, hardness; conductivity, optical and magnetic response; corrosion resistance
- L6: Performance and failure in service: the part works, passivates, wears, corrodes, embrittles, cracks or fatigues
materials science as materials scientists describe it (reader's view: materials in the world, characterization, data, structure, properties and mechanisms, the paradigm and shared record, processing and performance)
- L1: Matter on hand in the world: ores and elements, raw substances, samples and parts, under the conditions of the furnace, the load and the service environment
- L2: Characterization and testing: diffraction (X-ray, electron, neutron), microscopy, spectroscopy and chemical analysis, thermal analysis, mechanical tests
- L3: Data: diffraction patterns, micrographs, spectra, stress-strain curves, computed energies, written into records and databases
- L4: Structure identified across length scales: atomic arrangement and bonding, crystal or amorphous, defects, nanostructure, microstructure, macrostructure
- L5: Properties and their mechanisms: mechanical, electrical, thermal, optical, magnetic, chemical properties explained by thermodynamics (phase diagrams) and kinetics (diffusion), dislocation theory, fracture mechanics, DFT
- L6: The paradigm and the shared record: the processing-structure-properties-performance tetrahedron, the literature and the materials databases, and the community's judgement of claims
- L7: Design, processing and performance: choosing a target and making it (extraction, synthesis, heat treatment, casting, sintering, thin films), so the material performs in service; failures read back
the boardroom (shared floors)
- L4: Structure identified: atomic arrangement, bonding, defects, phases and microstructure, put together from the probes
- L5: Properties and mechanisms: thermodynamics, kinetics, dislocation theory, fracture mechanics and computation explain why the structure gives the properties
- L6: The paradigm and the shared record: the tetrahedron, the literature and databases, and the community's judgement of claims (novelty, credibility, utility)
- L7: Design and making: choosing targets and processing routes so materials perform in service
way in: Biological materials as donors (bioinspiration)
- L1: Natural materials such as nacre
- L2: Studies of their structure and toughening
- L3: Design principles taken across (brick and mortar)
way in: Computation: DFT, molecular dynamics, phase field, finite element, machine-learned models
- L1: Computing power
- L2: Codes at every length scale
- L3: Computed energies and stability against the convex hull
- L4: Candidate lists filtered by models and checked by DFT
way in: Demand: industry, war and funders asking for materials
- L1: Needs of space, defence, energy and industry
- L2: Contracts and programmes (ARPA laboratories)
- L3: A target property set (e.g. a fusion steel)
way in: Diffraction: X-ray, electron and neutron beams scattered by the sample
- L1: A beam meets the sample's atoms
- L2: Diffractometers; for polycrystals the powder method
- L3: The pattern, read by people or by ML models
way in: Failure in service: parts that break, corrode, embrittle
- L1: A component fails in service
- L2: Inspection and forensic examination
- L3: The cause named (a flaw, a mechanism)
- L4: Fracture mechanics as the port: stress intensity and flaw size
way in: Literature and databases: published syntheses, structures and properties
- L1: Papers and databases
- L2: Text mining and database queries (ICSD, Materials Project)
- L3: Analogies to known related materials
way in: Mechanical testing: loading to yield, fracture and fatigue
- L1: Load applied to the specimen
- L2: Tensile, flexural and fracture-toughness tests
- L3: Stress-strain curves and property values (yield, ultimate, elongation, K)
- L4: Property read against structure (yield strength against grain size)
way in: Microscopy: optical and electron microscopes on prepared surfaces and foils
- L1: A prepared surface or thin foil
- L2: Transmission electron microscope; photoluminescence imaging
- L3: Images and mosaics (the Mehrgarh map: 414 tiles, 1.5 gigapixels)
way in: Sight: the materials scientist's eye on samples, micrographs and patterns
- L1: Macrostructure seen with the naked eye
- L4: Single crystals show their atomic structure in their shapes
way in: Spectroscopy and chemical analysis
- L1: Light and the sample's electrons and bonds
- L2: Spectrometers tuned in excitation and emission
- L3: Spectra and composition read-outs (trace elements)
way in: The archaeological and environmental record: old artefacts and ice cores read for past processes
- L1: Corroded artefacts, slags, glacier ice
- L2: Imaging and chemical analysis of the specimen or core
- L3: A dated history of process (lost-wax casting; smelting episodes)
way in: Thermal analysis and calorimetry
- L1: Heat and the sample's phase changes
- L2: Calorimeters and thermal analysers
- L3: Enthalpy changes, transition temperatures
way in: Thought: chemical intuition and analogy
- L1: Contested: where thoughts come from is not settled (positions listed in the Human walk, C-413); the field names 'human chemical intuition'
- L4: Analogy to known related materials
way out: Autonomous synthesis: robots carrying out machine-planned recipes
- L1: Powder products
- L2: Three stations and robot arms
- L7: Targets and recipes chosen by models
way out: Emissions: what processing releases to the air and land
- L1: Metal in the atmosphere, deposited in glacier ice
- L7: Smelting at scale
way out: Processing: extraction, synthesis, casting, heat treatment, sintering, deposition
- L1: The material made
- L2: Furnaces, rolling mills, sintering presses
- L5: A route designed from mechanisms (e.g. a thermomechanical schedule)
- L7: The choice of a processing route
way out: Publishing: papers, databases and corrections
- L1: Journals and databases
- L3: Structures, energies and recipes written out
- L6: A claim the community will judge
way out: Supplying engineering: materials and property data into designs and industry
- L1: Materials in products (beverage containers, wings, wafers)
- L7: Material selection for an application
Sources (25) published science
Merchant A, Batzner S, Schoenholz SS et al. 2023. Scaling deep learning for materials discovery. Nature 624:80. doi:10.1038/s41586-023-06735-9 (PMC10700131) (behind its links)
Dislocation (Wikipedia, via Wikimedia Enterprise) (behind its links)
Zhang S, Yao Y, Long H et al. 2026. Research Paradigm of Materials Science Tetrahedra with Artificial Intelligence. arXiv:2603.13744 (preprint) (behind its links)
Grain boundary strengthening (Wikipedia, via Wikimedia Enterprise) (behind its links)
History of materials science (Wikipedia, via Wikimedia Enterprise) (behind its links)
Fracture mechanics (Wikipedia, via Wikimedia Enterprise) (behind its links)
Corrosion (Wikipedia, via Wikimedia Enterprise) (behind its links)
Engineering design process (Wikipedia, via Wikimedia Enterprise) (behind its links)
Mechanical engineering (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01, revision of 2026-10-30) (behind its links)
Strength of materials (Wikipedia, via Wikimedia Enterprise) (behind its links)
Metallurgy (Wikipedia, via Wikimedia Enterprise) (behind its links)