Skip to content
The Future Explorerby Benjamin Talin
Enablertechnological

Materials Science

Finding and designing substances with exactly the properties a job needs: stronger, lighter, better at carrying current or surviving heat. Batteries, chips, buildings and power lines are all materials problems first. Computers now suggest hundreds of thousands of possible new materials; one AI model predicted about 380,000 stable crystals. A robot lab ran for 17 days making new compounds, but checking whether they were truly new took two years. Making and proving is the slow part.

Maturity

2/ 5
measured
BasisComputers now propose candidate materials by the hundred thousand. Making them, proving they are new and qualifying them for use have not sped up - the best-known autonomous-lab result was corrected downward in January 2026.SourceMerchant et al., Nature 624:80–85 (29 Nov 2023); Szymanski et al., Nature 624:86–91 (2023), Author Correction (19 Jan 2026); Nature news (12 Dec 2023); Cheetham & Seshadri, Chemistry of Materials (ACS, Apr 2024); White House, Materials Genome Initiative (24 Jun 2011)Vintage2011–2026
ReadinessCapability 3Deployment 2Absorption 2Bound bycapabilityTrendflat

Where it feeds in

4 megatrends in one forceweakest link

Where you have already met this

At Lawrence Berkeley National Laboratory, a robotic lab ran for 17 days - weighing powders, firing furnaces, X-raying the results. The paper said it made 41 new materials from 58 targets. Other chemists questioned whether they were new at all. In January 2026 the paper was corrected. Discovery took seventeen days. Checking it took two years.

If this moved

If qualification caught up with discovery, critical minerals would still be held at 2 by geopolitical alignment, and reshoring by construction technology, geopolitics, hydrogen and managerial capability. The energy transition and the circular economy sit at 1 on consumption norms. Materials alone moves nothing.

What to watch for

A material first found by a machine-learning search receiving a formal industry specification. An autonomous lab making something beyond simple powders - an alloy or a thin film - that passes independent testing.

Not to be confused with

Computational drug discovery. Drug discovery searches for single molecules that bind to a target; materials science searches for solids whose structure gives them useful bulk properties.

What it is

The design and discovery of substances with specified properties - strength, conductivity, weight, thermal behaviour - increasingly with computational search.

How it works

  1. 01Generate candidatesA model proposes new crystal structures by swapping elements and rearranging known ones - millions at a time.
  2. 02Estimate their energyPhysics simulations calculate how much energy each structure has.
  3. 03Keep the stable onesA candidate is kept only if it is lower in energy than the combinations it could fall apart into.
  4. 04Plan the recipeSoftware proposes which starting powders to mix and how to heat them, often by learning from published recipes.
  5. 05Make it and check itRobots weigh and mix the powders, fire them in furnaces, and X-ray the result to see whether the intended structure formed. Proving it is new, and useful, is still work for people.

State of play

  • Google DeepMind's GNoME model predicted 2.2 million new crystal structures, about 380,000 of them stable (Merchant et al., Nature 624, 29 Nov 2023).
  • Of those, 736 had already been made independently by other researchers (Google DeepMind, 29 Nov 2023).
  • The A-Lab robotic platform made 36 of 57 target materials in 17 days of continuous operation, after a correction removed targets whose presence X-ray data could not confirm (Szymanski et al., Nature 624; Author Correction, 19 Jan 2026).
  • The correction also clarified that 'novel' meant new to the prediction platform, not necessarily new to science (Nature, Author Correction, 19 Jan 2026).
  • When the US launched the Materials Genome Initiative, it said a new material could take 20 or more years to go from discovery to a commercial product (White House, 24 Jun 2011).

Why it matters

It sits underneath almost every physical megatrend as an invisible enabler. Batteries, semiconductors, construction and energy transmission are all materials problems before they are anything else.

  • It holds two megatrends at 2: critical minerals, with geopolitical alignment, and reshored manufacturing, with four others.
  • It is the map's cleanest case of a bottleneck that moved rather than cleared. Discovery got fast. Everything after it did not.
  • Announcements describe the fast half. A new battery chemistry on a screen is a long way from a car.

What holds it back

Synthesis

A structure that is stable on paper may be hard or impossible to make, and simulations say little about how to make it.

Proving novelty

Checking whether a material is genuinely new - and really what the X-ray says - takes expert work, as the A-Lab dispute showed.

Qualification

Aerospace, medical and structural uses need years of testing before a new material is allowed in.

Beyond powders

Robotic labs handle simple powders. Alloys, thin films and composites are still made largely by hand.

Terms

Crystal structure
The repeating arrangement of atoms in a solid, which largely decides its properties.
Density functional theory (DFT)
A standard method for calculating a material's energy and electronic structure from physics.
Convex hull
The line of lowest-energy combinations for a set of elements; a structure on it is predicted to be stable.
X-ray diffraction (XRD)
Identifying a crystal structure from the pattern X-rays make when they scatter off it.
Qualification
The testing and certification a material must pass before it can be used in a regulated product.

Steepens

MegatrendWeightRole
Critical Mineral and Resource Scarcitylargeshared weakest link (with 3 others)
Circular Economy and Materials Substitutionlarge
Energy Transition and Electrificationmoderate
Distributed and Reshored Manufacturingmoderateshared weakest link (with 6 others)

The gradient behaves like a sticky rubber band: it is pulled hardest toward the weakest enabler. Marginal links are shown but do not pull.

If this moved one step

Computed from the map as it stands: this score raised from 2 to 3, every other score unchanged. It is arithmetic, not a forecast.

Raised one step on its own, it lifts no megatrend's weakest link; each gradient it feeds still rises a little.

Tied as the weakest link

Here it shares the weakest spot with other enablers. Raising it alone leaves the weakest link where it is and lifts the gradient a little; the enablers it is tied with must move too.

The gradient also rises in

What if it moved?

What would change the gradient

Conditional, not predictive. No dates, no probabilities.

  1. 01Qualification and certification timelines - the actual bottleneck
  2. 02Pilot-scale manufacturing capacity, which is chronically underfunded
  3. 03Substitution pressure from supply concentration in critical minerals

Common misconception

That materials progress is slow by nature. Discovery has accelerated sharply with computational methods; qualification and scale-up have not, and that gap is now the constraint.

Talin, B. (2026). Materials Science. The Future Explorer, dataset v0.4.0 (2026-11-16). https://explorer.talin.digital/e/materials-science/