AlphaFlow: autonomous discovery and optimization of multi-step chemistry using a self-driven fluidic lab guided by reinforcement learning

Nature Communications, Год журнала: 2023, Номер 14(1)

Опубликована: Март 14, 2023

Closed-loop, autonomous experimentation enables accelerated and material-efficient exploration of large reaction spaces without the need for user intervention. However, advanced materials with complex, multi-step processes data sparse environments remains a challenge. In this work, we present AlphaFlow, self-driven fluidic lab capable discovery complex chemistries. AlphaFlow uses reinforcement learning integrated modular microdroplet reactor performing steps variable sequence, phase separation, washing, continuous in-situ spectral monitoring. To demonstrate power toward high dimensionality chemistries, use to discover optimize synthetic routes shell-growth core-shell semiconductor nanoparticles, inspired by colloidal atomic layer deposition (cALD). Without prior knowledge conventional cALD parameters, successfully identified optimized novel route, up 40 that outperformed sequences. Through capabilities closed-loop, learning-guided systems in exploring solving challenges nanoparticle syntheses, while relying solely on in-house generated from miniaturized microfluidic platform. Further application chemistries beyond can lead fundamental generation as well route discoveries optimization.

Язык: Английский

---

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

Journal of the American Chemical Society, Год журнала: 2023, Номер 145(40), С. 21699 - 21716

Опубликована: Сен. 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.

Язык: Английский

---

A dynamic knowledge graph approach to distributed self-driving laboratories

Nature Communications, Год журнала: 2024, Номер 15(1)

Опубликована: Янв. 23, 2024

Abstract The ability to integrate resources and share knowledge across organisations empowers scientists expedite the scientific discovery process. This is especially crucial in addressing emerging global challenges that require solutions. In this work, we develop an architecture for distributed self-driving laboratories within World Avatar project, which seeks create all-encompassing digital twin based on a dynamic graph. We employ ontologies capture data material flows design-make-test-analyse cycles, utilising autonomous agents as executable components carry out experimentation workflow. Data provenance recorded ensure its findability, accessibility, interoperability, reusability. demonstrate practical application of our framework by linking two robots Cambridge Singapore collaborative closed-loop optimisation pharmaceutically-relevant aldol condensation reaction real-time. graph autonomously evolves toward scientist’s research goals, with effectively generating Pareto front cost-yield three days.

Язык: Английский

---

What is missing in autonomous discovery: open challenges for the community

Digital Discovery, Год журнала: 2023, Номер 2(6), С. 1644 - 1659

Опубликована: Янв. 1, 2023

Self-driving labs (SDLs) leverage combinations of artificial intelligence, automation, and advanced computing to accelerate scientific discovery.

Язык: Английский

---

Review of low-cost self-driving laboratories in chemistry and materials science: the “frugal twin” concept

Digital Discovery, Год журнала: 2024, Номер 3(5), С. 842 - 868

Опубликована: Янв. 1, 2024

Low-cost self-driving labs (SDLs) offer faster prototyping, low-risk hands-on experience, and a test bed for sophisticated experimental planning software which helps us develop state-of-the-art SDLs.

Язык: Английский

---

The future of self-driving laboratories: from human in the loop interactive AI to gamification

Digital Discovery, Год журнала: 2024, Номер 3(4), С. 621 - 636

Опубликована: Янв. 1, 2024

Self-driving laboratories (SDLs) are the future for scientific discovery in a world growing with artificial intelligence. The interaction between scientists and automated instrumentation leading conversations about impact of SDLs on research.

Язык: Английский

---

The advancement of artificial intelligence in biomedical research and health innovation: challenges and opportunities in emerging economies

Globalization and Health, Год журнала: 2024, Номер 20(1)

Опубликована: Май 21, 2024

The advancement of artificial intelligence (AI), algorithm optimization and high-throughput experiments has enabled scientists to accelerate the discovery new chemicals materials with unprecedented efficiency, resilience precision. Over recent years, so-called autonomous experimentation (AE) systems are featured as key AI innovation enhance research development (R&D). Also known self-driving laboratories or acceleration platforms, AE digital platforms capable running a large number autonomously. Those rapidly impacting biomedical clinical innovation, in areas such drug discovery, nanomedicine, precision oncology, others. As it is expected that will impact healthcare from local global levels, its implications for science technology emerging economies should be examined. By examining increasing relevance contemporary R&D activities, this article aims explore health highlighting implications, challenges opportunities economies. presents an opportunity stakeholders co-produce knowledge landscape health. However, asymmetries capabilities acknowledged since suffers inadequacies discontinuities resources funding. establishment decentralized infrastructures could support overcome restrictions opens venues more culturally diverse, equitable, trustworthy health-related through meaningful partnerships engagement. Collaborations innovators facilitate anticipation fiscal pressures policies, obsolescence infrastructures, ethical regulatory policy lag, other issues present Global South. Also, improving cultural geographical representativeness contributes foster diffusion acceptance worldwide. Institutional preparedness critical enable navigate coming years.

Язык: Английский

---

Smart Dope: A Self‐Driving Fluidic Lab for Accelerated Development of Doped Perovskite Quantum Dots

Advanced Energy Materials, Год журнала: 2023, Номер 14(1)

Опубликована: Ноя. 12, 2023

Abstract Metal cation‐doped lead halide perovskite (LHP) quantum dots (QDs) with photoluminescence yields (PLQYs) higher than unity, due to cutting phenomena, are an important building block of the next‐generation renewable energy technologies. However, synthetic route exploration and development highest‐performing QDs for device applications remain challenging. In this work, Smart Dope is presented, which a self‐driving fluidic lab (SDFL), accelerated synthesis space autonomous optimization LHP QDs. Specifically, multi‐cation doping CsPbCl 3 using one‐pot high‐temperature chemistry reported. continuously synthesizes multi‐cation‐doped high‐pressure gas‐liquid segmented flow format enable continuous experimentation minimal experimental noise at reaction temperatures up 255°C. offers multiple functionalities, including mechanistic studies through digital twin QD modeling, closed‐loop discovery, on‐demand manufacturing high‐performing Through these developments, autonomously identifies optimal Mn‐Yb co‐doped PLQY 158%, highest reported value class date. illustrates power SDFLs in accelerating discovery emerging advanced materials.

Язык: Английский

---

Engineering a Sustainable Future: Harnessing Automation, Robotics, and Artificial Intelligence with Self-Driving Laboratories

ACS Sustainable Chemistry & Engineering, Год журнала: 2024, Номер 12(34), С. 12695 - 12707

Опубликована: Авг. 6, 2024

The accelerating depletion of natural resources undoubtedly demands a radical reevaluation research practices addressing the escalating climate crisis. From traditional approaches to modern-day advancements, integration automation and artificial intelligence (AI)-guided decision-making has emerged as transformative route in shaping new methodologies. Harnessing robotics high-throughput alongside intelligent experimental design, self-driving laboratories (SDLs) offer an innovative solution expedite chemical/materials timelines while significantly reducing carbon footprint scientific endeavors, which could be utilized not only generate green materials but also make process itself more sustainable. In this Perspective, we examine potential SDLs driving sustainability forward through case studies discovery optimization, thereby paving way for greener efficient future. While hold immense promise, discuss challenges that persist their development deployment, necessitating holistic approach both design implementation.

Язык: Английский

---

Accelerated Multiobjective Calibration of Fused Deposition Modeling 3D Printers Using Multitask Bayesian Optimization and Computer Vision

Advanced Intelligent Systems, Год журнала: 2025, Номер unknown

Опубликована: Фев. 5, 2025

Proper process parameter calibration is critical to the success of fused deposition modeling (FDM) three‐dimensional (3D) printing, but time‐consuming and requires expertise. While existing systems for autonomous have demonstrated in calibrating a single objective, users may need balance multiple conflicting objectives. Herein, an easily deployable, camera‐based system FDM printers that optimizes both part quality completion time presented. Autonomous achieved through novel, multifaceted computer vision characterization multitask learning extension Bayesian optimization. The on four popular filament types using two distinct 3D printers. results show significantly outperforms manufacturer across machine material configurations, achieving average improvement 32.2% 31.2% decrease with respect benchmark.

Язык: Английский

---