14 examples of how LLMs can transform materials science and chemistry: a reflection on a large language model hackathon

Digital Discovery, Journal Year: 2023, Volume and Issue: 2(5), P. 1233 - 1250

Published: Jan. 1, 2023

We report the findings of a hackathon focused on exploring diverse applications large language models in molecular and materials science.

Language: Английский

---

Autonomous, multiproperty-driven molecular discovery: From predictions to measurements and back

Science, Journal Year: 2023, Volume and Issue: 382(6677)

Published: Dec. 21, 2023

A closed-loop, autonomous molecular discovery platform driven by integrated machine learning tools was developed to accelerate the design of molecules with desired properties. We demonstrated two case studies on dye-like molecules, targeting absorption wavelength, lipophilicity, and photooxidative stability. In first study, experimentally realized 294 unreported across three automatic iterations design-make-test-analyze cycles while exploring structure-function space four rarely reported scaffolds. each iteration, property prediction models that guided exploration learned structure-property diverse scaffold derivatives, which were multistep syntheses a variety reactions. The second study exploited trained explored chemical previously discover nine top-performing within lightly space.

Language: Английский

---

Self-driving laboratories to autonomously navigate the protein fitness landscape

Nature Chemical Engineering, Journal Year: 2024, Volume and Issue: 1(1), P. 97 - 107

Published: Jan. 11, 2024

Abstract Protein engineering has nearly limitless applications across chemistry, energy and medicine, but creating new proteins with improved or novel functions remains slow, labor-intensive inefficient. Here we present the Self-driving Autonomous Machines for Landscape Exploration (SAMPLE) platform fully autonomous protein engineering. SAMPLE is driven by an intelligent agent that learns sequence–function relationships, designs sends to a automated robotic system experimentally tests designed provides feedback improve agent’s understanding of system. We deploy four agents goal glycoside hydrolase enzymes enhanced thermal tolerance. Despite showing individual differences in their search behavior, all quickly converge on thermostable enzymes. laboratories automate accelerate scientific discovery process hold great potential fields synthetic biology.

Language: Английский

---

Self-Driving Laboratories for Chemistry and Materials Science

Chemical Reviews, Journal Year: 2024, Volume and Issue: 124(16), P. 9633 - 9732

Published: Aug. 13, 2024

Self-driving laboratories (SDLs) promise an accelerated application of the scientific method. Through automation experimental workflows, along with autonomous planning, SDLs hold potential to greatly accelerate research in chemistry and materials discovery. This review provides in-depth analysis state-of-the-art SDL technology, its applications across various disciplines, implications for industry. additionally overview enabling technologies SDLs, including their hardware, software, integration laboratory infrastructure. Most importantly, this explores diverse range domains where have made significant contributions, from drug discovery science genomics chemistry. We provide a comprehensive existing real-world examples different levels automation, challenges limitations associated each domain.

Language: Английский

---

In Pursuit of the Exceptional: Research Directions for Machine Learning in Chemical and Materials Science

Journal of the American Chemical Society, Journal Year: 2023, Volume and Issue: 145(40), P. 21699 - 21716

Published: Sept. 27, 2023

Exceptional molecules and materials with one or more extraordinary properties are both technologically valuable fundamentally interesting, because they often involve new physical phenomena compositions that defy expectations. Historically, exceptionality has been achieved through serendipity, but recently, machine learning (ML) automated experimentation have widely proposed to accelerate target identification synthesis planning. In this Perspective, we argue the data-driven methods commonly used today well-suited for optimization not realization of exceptional molecules. Finding such outliers should be possible using ML, only by shifting away from traditional ML approaches tweak composition, crystal structure, reaction pathway. We highlight case studies high-Tc oxide superconductors superhard demonstrate challenges ML-guided discovery discuss limitations automation task. then provide six recommendations development capable discovery: (i) Avoid tyranny middle focus on extrema; (ii) When data limited, qualitative predictions direction than interpolative accuracy; (iii) Sample what can made how make it defer optimization; (iv) Create room (and look) unexpected while pursuing your goal; (v) Try fill-in-the-blanks input output space; (vi) Do confuse human understanding model interpretability. conclude a description these integrated into workflows, which enable materials.

Language: Английский

---

Machine intelligence-accelerated discovery of all-natural plastic substitutes

Nature Nanotechnology, Journal Year: 2024, Volume and Issue: 19(6), P. 782 - 791

Published: March 18, 2024

Abstract One possible solution against the accumulation of petrochemical plastics in natural environments is to develop biodegradable plastic substitutes using components. However, discovering all-natural alternatives that meet specific properties, such as optical transparency, fire retardancy and mechanical resilience, which have made successful, remains challenging. Current approaches still rely on iterative optimization experiments. Here we show an integrated workflow combines robotics machine learning accelerate discovery with programmable optical, thermal properties. First, automated pipetting robot commanded prepare 286 nanocomposite films various properties train a support-vector classifier. Next, through 14 active loops data augmentation, 135 nanocomposites are fabricated stagewise, establishing artificial neural network prediction model. We demonstrate model can conduct two-way design task: (1) predicting physicochemical from its composition (2) automating inverse fulfils user-specific requirements. By harnessing model’s capabilities, several substitutes, could replace non-biodegradable counterparts exhibiting analogous Our methodology integrates robot-assisted experiments, intelligence simulation tools eco-friendly starting building blocks taken generally-recognized-as-safe database.

Language: Английский

---

Expanding the Horizons of Machine Learning in Nanomaterials to Chiral Nanostructures

Advanced Materials, Journal Year: 2024, Volume and Issue: 36(18)

Published: Jan. 19, 2024

Abstract Machine learning holds significant research potential in the field of nanotechnology, enabling nanomaterial structure and property predictions, facilitating materials design discovery, reducing need for time‐consuming labor‐intensive experiments simulations. In contrast to their achiral counterparts, application machine chiral nanomaterials is still its infancy, with a limited number publications date. This despite great advance development new sustainable high values optical activity, circularly polarized luminescence, enantioselectivity, as well analysis structural chirality by electron microscopy. this review, an methods used studying provided, subsequently offering guidance on adapting extending work nanomaterials. An overview within framework synthesis–structure–property–application relationships presented insights how leverage study these highly complex are provided. Some key recent reviewed discussed Finally, review captures achievements, ongoing challenges, prospective outlook very important field.

Language: Английский

---

3D printing and artificial intelligence tools for droplet microfluidics: Advances in the generation and analysis of emulsions

Applied Physics Reviews, Journal Year: 2025, Volume and Issue: 12(1)

Published: Jan. 21, 2025

Droplet microfluidics has emerged as highly relevant technology in diverse fields such nanomaterials synthesis, photonics, drug delivery, regenerative medicine, food science, cosmetics, and agriculture. While significant progress been made understanding the fundamental mechanisms underlying droplet generation microchannels fabricating devices to produce droplets with varied functionality high throughput, challenges persist along two important directions. On one side, generalization of numerical results obtained by computational fluid dynamics would be deepen comprehension complex physical phenomena microfluidics, well capability predicting device behavior. Conversely, truly three-dimensional architectures enhance microfluidic platforms terms tailoring enhancing flow properties. Recent advancements artificial intelligence (AI) additive manufacturing (AM) promise unequaled opportunities for simulating behavior, precisely tracking individual droplets, exploring innovative designs. This review provides a comprehensive overview recent applying AI AM microfluidics. The basic properties multiphase flows production are discussed, current fabrication methods related introduced, together their applications. Delving into use technologies topics covered include AI-assisted simulations real-time within systems, AM-fabrication systems. synergistic combination is expected active matter expediting transition toward fully digital

Language: Английский

---

Self-driving lab for the photochemical synthesis of plasmonic nanoparticles with targeted structural and optical properties

Nature Communications, Journal Year: 2025, Volume and Issue: 16(1)

Published: Feb. 8, 2025

Many applications of plasmonic nanoparticles require precise control their optical properties that are governed by nanoparticle dimensions, shape, morphology and composition. Finding reaction conditions for the synthesis with targeted characteristics is a time-consuming resource-intensive trial-and-error process, however closed-loop enables accelerated exploration large chemical spaces without human intervention. Here, we introduce Autonomous Fluidic Identification Optimization Nanochemistry (AFION) self-driving lab integrates microfluidic reactor, in-flow spectroscopic characterization, machine learning optimization multidimensional space photochemical nanoparticles. By targeting properties, AFION identifies different types designated shapes, morphologies, compositions. Data analysis provides insight into role type. This work shows an effective platform on-demand The automated challenging task. Here authors integrate fluidic real-time in self-driven properties.

Language: Английский

---

Preventing cation intermixing enables 50% quantum yield in sub-15 nm short-wave infrared-emitting rare-earth based core-shell nanocrystals

Nature Communications, Journal Year: 2023, Volume and Issue: 14(1)

Published: July 25, 2023

Short-wave infrared (SWIR) fluorescence could become the new gold standard in optical imaging for biomedical applications due to important advantages such as lack of autofluorescence, weak photon absorption by blood and tissues, reduced scattering coefficient. Therefore, contrary visible NIR regions, tissues translucent SWIR region. Nevertheless, bright biocompatible probes is a key challenge that must be overcome unlock full potential fluorescence. Although rare-earth-based core-shell nanocrystals appeared promising probes, they suffer from limited photoluminescence quantum yield (PLQY). The control over atomic scale organization complex materials one main barriers limiting their performance. Here, growth either homogeneous (α-NaYF4) or heterogeneous (CaF2) shell domains on optically-active α-NaYF4:Yb:Er (with without Ce3+ co-doping) core reported. can controlled preventing cation intermixing only with dramatic impact PLQY. latter reached 50% at 60 mW/cm2; highest reported PLQY values sub-15 nm nanocrystals. most efficient were utilized vivo above 1450 nm.

Language: Английский

---