iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
AI can help predict promising inorganic crystal structures, and robots can test ways to make selected materials—but those are separate steps, not a guarantee of a useful product. Google DeepMind’s GNoME predicted candidate structures and their stability; Lawrence Berkeley National Laboratory’s A-Lab used computational data and robotic experiments to try synthesizing selected powder compounds. In its peer-reviewed study, A-Lab confirmed 36 of 57 target compounds after 17 days of continuous operation. That result demonstrates an automated route from prediction-informed selection to experimental synthesis, not proof that every AI prediction can be made or that the resulting materials work in devices.
What did Google DeepMind’s AI and Berkeley Lab’s robot lab each do?
GNoME, short for Graph Networks for Materials Exploration, is a Google DeepMind system that generates candidate crystal structures and predicts their stability. The A-Lab, an autonomous laboratory at Lawrence Berkeley National Laboratory, is a specialized system for attempting to synthesize selected inorganic powders and examining what it made.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
Inorganic Chemistry | $92.90 | Buy on Amazon |
| 2 |
|
Inorganic Chemistry (5th Edition) | $280.55 | Buy on Amazon |
| 3 |
|
Inorganic Chemistry | $92.99 | Buy on Amazon |
| 4 |
|
Inorganic Chemistry (4th Edition) | $100.60 | Buy on Amazon |
| 5 |
|
Inorganic Chemistry | $94.52 | Buy on Amazon |
The contributions connect through materials data and computational screening, but they are not one pipeline in which every GNoME prediction went straight to a robot. The A-Lab paper says its targets came from the Materials Project and were cross-referenced with an analogous Google DeepMind database. A-Lab combined phase-stability data with machine-learning interpretation, synthesis heuristics learned from scientific literature, robotic handling, and active learning to refine recipes. The A-Lab study in Nature and Google DeepMind’s GNoME announcement describe related but distinct efforts.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsGNoME: computational candidates
Google DeepMind reported that GNoME predicted 2.2 million crystals, including 380,000 it identified as its most stable candidate materials. The company also reported that external researchers had independently created 736 of the predicted structures experimentally. These figures describe GNoME’s broader prediction and validation context; they are not counts of A-Lab targets or A-Lab syntheses.
#1 Best Overall
A-Lab: experimental synthesis attempts
A-Lab selected compounds to attempt to make, prepared and heated precursor powders, then used X-ray diffraction to assess the products. The peer-reviewed paper reports that it synthesized 36 of 57 targets during 17 days of continuous operation. Its authors manually reviewed the diffraction patterns to confirm those 36 target phases. Confirmation of a phase does not, by itself, establish that a sample was pure or suitable for a device.
How did the autonomous laboratory attempt to make materials?
A-Lab was designed for air-stable inorganic powder synthesis, not as a general-purpose chemistry robot. Its workflow linked recipe selection, physical handling, heating, characterization, and follow-up experiments:
Rank #2
- Used Book in Good Condition
- Select a target and propose a recipe. The system used computational phase-stability information, machine-learning interpretation, and synthesis heuristics derived from text-mined research literature to generate initial recipes and propose temperatures.
- Prepare precursor powders. Robots dosed and mixed powder precursors, then moved the mixtures in crucibles to furnaces for heating.
- Characterize the product. After cooling, samples were transferred for grinding and X-ray diffraction. Machine-learning analysis estimated which phases were present and in what fractions; automated Rietveld refinement checked the assessment.
- Adjust recipes when needed. If the target yield was insufficient, active learning used computed reaction energies and observed experimental outcomes to propose follow-up recipes.
This experimental feedback loop matters because computational stability is not a synthesis recipe. A predicted candidate can be thermodynamically promising yet difficult to form under the conditions a laboratory can use.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
What did the experiment establish—and what did it not?
The central result is specific: the Nature paper reports 36 confirmed target phases among 57 attempted compounds, after 17 days of continuous A-Lab operation. The authors’ manual review of diffraction data is the basis for the confirmed count. It does not mean the lab made all targets, or that all 36 products were high-purity samples.
Rank #3
The paper discusses several reasons attempts can fail, including slow reaction kinetics, volatile precursors, amorphization, and inaccuracies in computation. The 17 targets not obtained are part of the result: prediction can narrow the search, but experimental synthesis remains uncertain.
There is also a difference between this peer-reviewed paper’s count and a contemporaneous Nature News account, which described 41 materials. The paper’s later manual X-ray diffraction review confirmed 36 and treated four additional cases as inconclusive. For the study’s confirmed result, the paper’s qualified 36-of-57 figure is the appropriate one to use.
Rank #4
Are the new materials ready for batteries, solar cells, or electronics?
No device performance or commercial readiness was demonstrated by this A-Lab study. Batteries, solar cells, superconductors, and electronics are motivating areas where better materials could matter, not applications proven by the synthesis results. A detected target phase is an experimental materials finding; it does not show that the material delivers useful performance in a working device, can be manufactured economically at scale, or is ready for commercial use.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →The paper describes multigram powder samples as useful for later device-level testing, but it does not report a working-device demonstration. Ekin Dogus Cubuk, who led Google DeepMind’s materials discovery team in London, said, “A lot of the technologies around us, including batteries and solar cells, could really improve with better materials.” That is the motivation for the work, not evidence that these particular products have improved. Nature News, 29 November 2023.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
