Autonomous Chemical Experiments: Challenges and Perspectives on Establishing a Self-Driving Lab

Self-Driving Laboratories for Chemistry and Materials Science

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

Learning Force Field Parameters from Differentiable Particle-Field Molecular Dynamics

We develop ∂-HylleraasMD (∂-HyMD), a fully end-to-end differentiable molecular dynamics software based on the Hamiltonian hybrid particle-field formalism, and use it to establish a protocol for automated optimization of force field parameters. ∂-HyMD is templated on the recently released HylleraaasMD software, while using the JAX autodiff framework as the main engine for the differentiable dynamics. ∂-HyMD exploits an embarrassingly parallel optimization algorithm by spawning independent simulations, whose trajectories are simultaneously processed by reverse mode automatic differentiation to calculate the gradient of the loss function, which is in turn used for iterative optimization of the force-field parameters. We show that parallel organization facilitates the convergence of the minimization procedure, avoiding the known memory and numerical stability issues of differentiable molecular dynamics approaches. We showcase the effectiveness of our implementation by producing a library of force field parameters for standard phospholipids, with either zwitterionic or anionic heads and with saturated or unsaturated tails. Compared to the all-atom reference, the force field obtained by ∂-HyMD yields better density profiles than the parameters derived from previously utilized gradient-free optimization procedures. Moreover, ∂-HyMD models can predict with good accuracy properties not included in the learning objective, such as lateral pressure profiles, and are transferable to other systems, including triglycerides.

Expanding chemistry through in vitro and in vivo biocatalysis

Living systems contain a vast network of metabolic reactions, providing a wealth of enzymes and cells as potential biocatalysts for chemical processes. The properties of protein and cell biocatalysts-high selectivity, the ability to control reaction sequence and operation in environmentally benign conditions-offer approaches to produce molecules at high efficiency while lowering the cost and environmental impact of industrial chemistry. Furthermore, biocatalysis offers the opportunity to generate chemical structures and functions that may be inaccessible to chemical synthesis. Here we consider developments in enzymes, biosynthetic pathways and cellular engineering that enable their use in catalysis for new chemistry and beyond.

Machine learning in drug delivery

For decades, drug delivery scientists have been performing trial-and-error experimentation to manually sample parameter spaces and optimize release profiles through rational design. To enable this approach, scientists spend much of their career learning nuanced drug-material interactions that drive system behavior. In relatively simple systems, rational design criteria allow us to fine tune release profiles and enable efficacious therapies. However, as materials and drugs become increasingly sophisticated and their interactions have non-linear and compounding effects, the field is suffering the Curse of Dimensionality which prevents us from comprehending complex structure-function relationships. In the past, we have embraced this complexity by implementing high-throughput screens to increase the probability of finding ideal compositions. However, this brute force method was inefficient and led many to abandon these fishing expeditions. Fortunately, methods in data science including artificial intelligence / machine learning (AI/ML) are providing ideal analytical tools to model this complex data and ascertain quantitative structure-function relationships. In this Oration, I speak to the potential value of data science in drug delivery with particular focus on polymeric delivery systems. Here, I do not suggest that AI/ML will simply replace mechanistic understanding of complex systems. Rather, I propose that AI/ML should be yet another useful tool in the lab to navigate complex parameter spaces. The recent hype around AI/ML is breathtaking and potentially over inflated, but the value of these methods is poised to revolutionize how we perform science. Therefore, I encourage readers to consider adopting these skills and applying data science methods to their own problems. If done successfully, I believe we will all realize a paradigm shift in our approach to drug delivery.

In Situ Quantitative Monitoring of Adsorption from Aqueous Phase by UV-vis Spectroscopy: Implication for Understanding of Heterogeneous Processes

The development of in situ techniques to quantitatively characterize the heterogeneous reactions is essential for understanding physicochemical processes in aqueous phase. In this work, a new approach coupling in situ UV-vis spectroscopy with a two-step algorithm strategy is developed to quantitatively monitor heterogeneous reactions in a compact closed-loop incorporation. The algorithm involves the inverse adding-doubling method for light scattering correction and the multivariate curve resolution-alternating least squares (MCR-ALS) method for spectral deconvolution. Innovatively, theoretical spectral simulations are employed to connect MCR-ALS solutions with chemical molecular structural evolution without prior information for reference spectra. As a model case study, the aqueous adsorption kinetics of bisphenol A onto polyamide microparticles are successfully quantified in a one-step UV-vis spectroscopic measurement. The practical applicability of this approach is confirmed by rapidly screening a superior adsorbent from commercial materials for antibiotic wastewater adsorption treatment. The demonstrated capabilities are expected to extend beyond monitoring adsorption systems to other heterogeneous reactions, significantly advancing UV-vis spectroscopic techniques toward practical integration into automated experimental platforms for probing aqueous chemical processes and beyond.

Donor Preactivation-Based Glycan Assembly: from Manual to Automated Synthesis

Carbohydrates are called the third chain of life. Carbohydrates participate in many important biochemical functions in living species, and the biological information carried by them is several orders of magnitude larger than that of nucleic acids and proteins. However, due to the intrinsic complexity and heterogeneity of carbohydrate structures, furnishing pure and structurally well-defined glycans for functional studies is a formidable task, especially for homogeneous large-size glycans. To address this issue, we have developed a donor preactivation-based one-pot glycosylation strategy enabling multiple sequential glycosylations in a single reaction vessel.The donor preactivation-based one-pot glycosylation refers to the strategy in which the glycosyl donor is activated in the absence of a glycosyl acceptor to generate a reactive intermediate. Subsequently, the glycosyl acceptor with the same anomeric leaving group is added, leading to a glycosyl coupling reaction, which is then iterated to rapidly achieve the desired glycan in the same reactor. The advantages of this strategy include the following: (1) unique chemoselectivity is obtained after preactivation; (2) it is independent of the reactivity of glycosyl donors; (3) multiple-step glycosylations are enabled without the need for intermediate purification; (4) only stoichiometric building blocks are required without complex protecting group manipulations. Using this protocol, a range of glycans including tumor-associated carbohydrate antigens, various glycosaminoglycans, complex N-glycans, and diverse bacterial glycans have been synthesized manually. Gratifyingly, the synthesis of mycobacterial arabinogalactan containing 92 monosaccharide units has been achieved, which created a precedent in the field of polysaccharide synthesis. Recently, the synthesis of a highly branched arabinogalactan from traditional Chinese medicine featuring 140 monosaccharide units has been also accomplished to evaluate its anti-pancreatic-cancer activity. In the spirit of green and sustainable chemistry, this strategy can also be applied to light-driven glycosylation reactions, where either UV or visible light can be used for the activation of glycosyl donors.Automated synthesis is an advanced approach to the construction of complex glycans. Based on the two preactivation modes (general promoter activation mode and light-induced activation mode), a universal and highly efficient automated solution-phase synthesizer was further developed to drive glycan assembly from manual to automated synthesis. Using this synthesizer, a library of oligosaccharides covering various glycoforms and glycosidic linkages was assembled rapidly, either in a general promoter-activation mode or in a light-induced-activation mode. The automated synthesis of a fully protected fondaparinux pentasaccharide was realized on a gram scale. Furthermore, the automated synthesis of large-size polysaccharides was performed, allowing the assembly of arabinans up to an astonishing 1080-mer using the automated multiplicative synthesis strategy, taking glycan synthesis to a new height far beyond the synthesis of nucleic acids (up to 200-mer) and proteins (up to 472-mer).

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

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

Delocalized, asynchronous, closed-loop discovery of organic laser emitters

Contemporary materials discovery requires intricate sequences of synthesis, formulation, and characterization that often span multiple locations with specialized expertise or instrumentation. To accelerate these workflows, we present a cloud-based strategy that enabled delocalized and asynchronous design-make-test-analyze cycles. We showcased this approach through the exploration of molecular gain materials for organic solid-state lasers as a frontier application in molecular optoelectronics. Distributed robotic synthesis and in-line property characterization, orchestrated by a cloud-based artificial intelligence experiment planner, resulted in the discovery of 21 new state-of-the-art materials. Gram-scale synthesis ultimately allowed for the verification of best-in-class stimulated emission in a thin-film device. Demonstrating the asynchronous integration of five laboratories across the globe, this workflow provides a blueprint for delocalizing-and democratizing-scientific discovery.

Toward Self-Driven Autonomous Material and Device Acceleration Platforms (AMADAP) for Emerging Photovoltaics Technologies

ConspectusIn the ever-increasing renewable-energy demand scenario, developing new photovoltaic technologies is important, even in the presence of established terawatt-scale silicon technology. Emerging photovoltaic technologies play a crucial role in diversifying material flows while expanding the photovoltaic product portfolio, thus enhancing security and competitiveness within the solar industry. They also serve as a valuable backup for silicon photovoltaic, providing resilience to the overall energy infrastructure. However, the development of functional solar materials poses intricate multiobjective optimization challenges in a large multidimensional composition and parameter space, in some cases with millions of potential candidates to be explored. Solving it necessitates reproducible, user-independent laboratory work and intelligent preselection of innovative experimental methods.Materials acceleration platforms (MAPs) seamlessly integrate robotic materials synthesis and characterization with AI-driven data analysis and experimental design, positioning them as enabling technologies for the discovery and exploration of new materials. They are proposed to revolutionize materials development away from the Edisonian trial-and-error approaches to ultrashort cycles of experiments with exceptional precision, generating a reliable and highly qualitative data situation that allows training machine learning algorithms with predictive power. MAPs are designed to assist the researcher in multidimensional aspects of materials discovery, such as material synthesis, precursor preparation, sample processing and characterization, and data analysis, and are drawing escalating attention in the field of energy materials. Device acceleration platforms (DAPs), however, are designed to optimize functional films and layer stacks. Unlike MAPs, which focus on material discovery, a central aspect of DAPs is the identification and refinement of ideal processing conditions for a predetermined set of materials. Such platforms prove especially invaluable when dealing with "disordered semiconductors," which depend heavily on the processing parameters that ultimately define the functional properties and functionality of thin film layers. By facilitating the fine-tuning of processing conditions, DAPs contribute significantly to the advancement and optimization of disordered semiconductor devices, such as emerging photovoltaics.In this Account, we review the recent advancements made by our group in automated and autonomous laboratories for advanced material discovery and device optimization with a strong focus on emerging photovoltaics, such as solution-processing perovskite solar cells and organic photovoltaics. We first introduce two MAPs and two DAPs developed in-house: a microwave-assisted high-throughput synthesis platform for the discovery of organic interface materials, a multipurpose robot-based pipetting platform for the synthesis of new semiconductors and the characterization of thin film semiconductor composites, the SPINBOT system, which is a spin-coating DAP with the potential to optimize complex device architectures, and finally, AMANDA, a fully integrated and autonomously operating DAP. Notably, we underscore the utilization of a robot-based high-throughput experimentation technique to address the common optimization challenges encountered in extensive multidimensional composition and parameter spaces pertaining to organic and perovskite photovoltaics materials. Finally, we briefly propose a holistic concept and technology, a self-driven autonomous material and device acceleration platform (AMADAP) laboratory, for autonomous functional solar materials discovery and development. We hope to discover how AMADAP can be further strengthened and universalized with advancing development of hardware and software infrastructures in the future.

AEcroscopy: A Software-Hardware Framework Empowering Microscopy Toward Automated and Autonomous Experimentation

Microscopy has been pivotal in improving the understanding of structure-function relationships at the nanoscale and is by now ubiquitous in most characterization labs. However, traditional microscopy operations are still limited largely by a human-centric click-and-go paradigm utilizing vendor-provided software, which limits the scope, utility, efficiency, effectiveness, and at times reproducibility of microscopy experiments. Here, a coupled software-hardware platform is developed that consists of a software package termed AEcroscopy (short for Automated Experiments in Microscopy), along with a field-programmable-gate-array device with LabView-built customized acquisition scripts, which overcome these limitations and provide the necessary abstractions toward full automation of microscopy platforms. The platform works across multiple vendor devices on scanning probe microscopes and electron microscopes. It enables customized scan trajectories, processing functions that can be triggered locally or remotely on processing servers, user-defined excitation waveforms, standardization of data models, and completely seamless operation through simple Python commands to enable a plethora of microscopy experiments to be performed in a reproducible, automated manner. This platform can be readily coupled with existing machine-learning libraries and simulations, to provide automated decision-making and active theory-experiment optimization to turn microscopes from characterization tools to instruments capable of autonomous model refinement and physics discovery.

A Computational Perspective on the Reactivity of π-spacers in Self-Immolative Elimination Reactions

The controlled release of chemicals, especially in drug delivery, is crucial, often employing "self-immolative" spacers to enhance reliability. These spacers separate the payload from the protecting group, ensuring a more controlled release. Over the years, design rules have been proposed to improve the elimination process's reaction rate by modifying spacers with electron-donating groups or reducing their aromaticity. The spacer design is critical for determining the range of functional groups released during this process. This study explores various strategies from the literature aimed at improving release rates, focusing on the electronic nature of the spacer, its aromaticity, the electronic nature of its substituents, and the leaving groups involved in the elimination reaction. Through computational analysis, I investigate activation free energies by identifying transition states for model reactions. My calculations align qualitatively with experimental results, demonstrating the feasibility and reliability of computationally pre-screening model self-immolative eliminations. This approach allows proposing optimal combinations of spacer and leaving group for achieving the highest possible release rate.

Autonomous closed-loop mechanistic investigation of molecular electrochemistry via automation

Electrochemical research often requires stringent combinations of experimental parameters that are demanding to manually locate. Recent advances in automated instrumentation and machine-learning algorithms unlock the possibility for accelerated studies of electrochemical fundamentals via high-throughput, online decision-making. Here we report an autonomous electrochemical platform that implements an adaptive, closed-loop workflow for mechanistic investigation of molecular electrochemistry. As a proof-of-concept, this platform autonomously identifies and investigates an EC mechanism, an interfacial electron transfer (E step) followed by a solution reaction (C step), for cobalt tetraphenylporphyrin exposed to a library of organohalide electrophiles. The generally applicable workflow accurately discerns the EC mechanism's presence amid negative controls and outliers, adaptively designs desired experimental conditions, and quantitatively extracts kinetic information of the C step spanning over 7 orders of magnitude, from which mechanistic insights into oxidative addition pathways are gained. This work opens opportunities for autonomous mechanistic discoveries in self-driving electrochemistry laboratories without manual intervention.

Revisiting El-Sayed Synthesis: Bayesian Optimization for Revealing New Insights during the Growth of Gold Nanorods

In diverse fields, machine learning (ML) has sparked transformative changes, primarily driven by the wealth of big data. However, an alternative approach seeks to mine insights from "precious data", offering the possibility to reveal missed knowledge and escape potential knowledge traps. In this context, Bayesian optimization (BO) protocols have emerged as crucial tools for optimizing the synthesis and discovery of a broad spectrum of compounds including nanoparticles. In our work, we aimed to go beyond the commonly explored experimental conditions and showcase a workflow capable of unearthing fresh insights, even in well-studied research domains. The growth of AuNRs is a nonequilibrium process that remains poorly understood despite the presence of well-established seeded growth protocols. Traditional research aimed at understanding the mechanism of AuNR growth has primarily relied on altering one reaction condition at a time. While these studies are undeniably valuable, they often fail to capture the synergies between different reaction conditions, thus constraining the depth of insights they can offer. In the present study, we exploit BO, to identify diverse experimental conditions yielding AuNRs with similar spectroscopic characteristics. Notably, we identify viable and accelerated synthesis conditions involving elevated temperatures (36-40 °C) as well as high ascorbic acid concentrations. More importantly, we note that ascorbic acid and temperature can modulate each other's undesirable influences on the growth of AuNRs. Finally, by harnessing the power of interpretable ML algorithms, complemented by our deep chemical understanding, we revisited the established hierarchical relationships among reaction conditions that impact the El-Sayed-based growth of AuNRs.

Accelerated chemical science with AI

In light of the pressing need for practical materials and molecular solutions to renewable energy and health problems, to name just two examples, one wonders how to accelerate research and development in the chemical sciences, so as to address the time it takes to bring materials from initial discovery to commercialization. Artificial intelligence (AI)-based techniques, in particular, are having a transformative and accelerating impact on many if not most, technological domains. To shed light on these questions, the authors and participants gathered in person for the ASLLA Symposium on the theme of 'Accelerated Chemical Science with AI' at Gangneung, Republic of Korea. We present the findings, ideas, comments, and often contentious opinions expressed during four panel discussions related to the respective general topics: 'Data', 'New applications', 'Machine learning algorithms', and 'Education'. All discussions were recorded, transcribed into text using Open AI's Whisper, and summarized using LG AI Research's EXAONE LLM, followed by revision by all authors. For the broader benefit of current researchers, educators in higher education, and academic bodies such as associations, publishers, librarians, and companies, we provide chemistry-specific recommendations and summarize the resulting conclusions.

PolyNC: a natural and chemical language model for the prediction of unified polymer properties

Language models exhibit a profound aptitude for addressing multimodal and multidomain challenges, a competency that eludes the majority of off-the-shelf machine learning models. Consequently, language models hold great potential for comprehending the intricate interplay between material compositions and diverse properties, thereby accelerating material design, particularly in the realm of polymers. While past limitations in polymer data hindered the use of data-intensive language models, the growing availability of standardized polymer data and effective data augmentation techniques now opens doors to previously uncharted territories. Here, we present a revolutionary model to enable rapid and precise prediction of Polymer properties via the power of Natural language and Chemical language (PolyNC). To showcase the efficacy of PolyNC, we have meticulously curated a labeled prompt-structure-property corpus encompassing 22 970 polymer data points on a series of essential polymer properties. Through the use of natural language prompts, PolyNC gains a comprehensive understanding of polymer properties, while employing chemical language (SMILES) to describe polymer structures. In a unified text-to-text manner, PolyNC consistently demonstrates exceptional performance on both regression tasks (such as property prediction) and the classification task (polymer classification). Simultaneous and interactive multitask learning enables PolyNC to holistically grasp the structure-property relationships of polymers. Through a combination of experiments and characterizations, the generalization ability of PolyNC has been demonstrated, with attention analysis further indicating that PolyNC effectively learns structural information about polymers from multimodal inputs. This work provides compelling evidence of the potential for deploying end-to-end language models in polymer research, representing a significant advancement in the AI community's dedicated pursuit of advancing polymer science.

Frontiers in nonviral delivery of small molecule and genetic drugs, driven by polymer chemistry and machine learning for materials informatics

Materials informatics (MI) has immense potential to accelerate the pace of innovation and new product development in biotechnology. Close collaborations between skilled physical and life scientists with data scientists are being established in pursuit of leveraging MI tools in automation and artificial intelligence (AI) to predict material properties in vitro and in vivo. However, the scarcity of large, standardized, and labeled materials data for connecting structure-function relationships represents one of the largest hurdles to overcome. In this Highlight, focus is brought to emerging developments in polymer-based therapeutic delivery platforms, where teams generate large experimental datasets around specific therapeutics and successfully establish a design-to-deployment cycle of specialized nanocarriers. Three select collaborations demonstrate how custom-built polymers protect and deliver small molecules, nucleic acids, and proteins, representing ideal use-cases for machine learning to understand how molecular-level interactions impact drug stabilization and release. We conclude with our perspectives on how MI innovations in automation efficiencies and digitalization of data-coupled with fundamental insight and creativity from the polymer science community-can accelerate translation of more gene therapies into lifesaving medicines.

A Review of Transition Metal Boride, Carbide, Pnictide, and Chalcogenide Water Oxidation Electrocatalysts

Transition metal borides, carbides, pnictides, and chalcogenides (X-ides) have emerged as a class of materials for the oxygen evolution reaction (OER). Because of their high earth abundance, electrical conductivity, and OER performance, these electrocatalysts have the potential to enable the practical application of green energy conversion and storage. Under OER potentials, X-ide electrocatalysts demonstrate various degrees of oxidation resistance due to their differences in chemical composition, crystal structure, and morphology. Depending on their resistance to oxidation, these catalysts will fall into one of three post-OER electrocatalyst categories: fully oxidized oxide/(oxy)hydroxide material, partially oxidized core@shell structure, and unoxidized material. In the past ten years (from 2013 to 2022), over 890 peer-reviewed research papers have focused on X-ide OER electrocatalysts. Previous review papers have provided limited conclusions and have omitted the significance of "catalytically active sites/species/phases" in X-ide OER electrocatalysts. In this review, a comprehensive summary of (i) experimental parameters (e.g., substrates, electrocatalyst loading amounts, geometric overpotentials, Tafel slopes, etc.) and (ii) electrochemical stability tests and post-analyses in X-ide OER electrocatalyst publications from 2013 to 2022 is provided. Both mono and polyanion X-ides are discussed and classified with respect to their material transformation during the OER. Special analytical techniques employed to study X-ide reconstruction are also evaluated. Additionally, future challenges and questions yet to be answered are provided in each section. This review aims to provide researchers with a toolkit to approach X-ide OER electrocatalyst research and to showcase necessary avenues for future investigation.

Chemical Species Ontology for Data Integration and Knowledge Discovery

Web ontologies are important tools in modern scientific research because they provide a standardized way to represent and manage web-scale amounts of complex data. In chemistry, a semantic database for chemical species is indispensable for its ability to interrelate and infer relationships, enabling a more precise analysis and prediction of chemical behavior. This paper presents OntoSpecies, a web ontology designed to represent chemical species and their properties. The ontology serves as a core component of The World Avatar knowledge graph chemistry domain and includes a wide range of identifiers, chemical and physical properties, chemical classifications and applications, and spectral information associated with each species. The ontology includes provenance and attribution metadata, ensuring the reliability and traceability of data. Most of the information about chemical species are sourced from PubChem and ChEBI data on the respective compound Web pages using a software agent, making OntoSpecies a comprehensive semantic database of chemical species able to solve novel types of problems in the field. Access to this reliable source of chemical data is provided through a SPARQL end point. The paper presents example use cases to demonstrate the contribution of OntoSpecies in solving complex tasks that require integrated semantically searchable chemical data. The approach presented in this paper represents a significant advancement in the field of chemical data management, offering a powerful tool for representing, navigating, and analyzing chemical information to support scientific research.

Explainability and human intervention in autonomous scanning probe microscopy

The broad adoption of machine learning (ML)-based autonomous experiments (AEs) in material characterization and synthesis requires strategies development for understanding and intervention in the experimental workflow. Here, we introduce and realize a post-experimental analysis strategy for deep kernel learning-based autonomous scanning probe microscopy. This approach yields real-time and post-experimental indicators for the progression of an active learning process interacting with an experimental system. We further illustrate how this approach can be applied to human-in-the-loop AEs, where human operators make high-level decisions at high latencies setting the policies for AEs, and the ML algorithm performs low-level, fast decisions. The proposed approach is universal and can be extended to other techniques and applications such as combinatorial library analysis.

Revolutionizing drug formulation development: The increasing impact of machine learning

Over the past few years, the adoption of machine learning (ML) techniques has rapidly expanded across many fields of research including formulation science. At the same time, the use of lipid nanoparticles to enable the successful delivery of mRNA vaccines in the recent COVID-19 pandemic demonstrated the impact of formulation science. Yet, the design of advanced pharmaceutical formulations is non-trivial and primarily relies on costly and time-consuming wet-lab experimentation. In 2021, our group published a review article focused on the use of ML as a means to accelerate drug formulation development. Since then, the field has witnessed significant growth and progress, reflected by an increasing number of studies published in this area. This updated review summarizes the current state of ML directed drug formulation development, introduces advanced ML techniques that have been implemented in formulation design and shares the progress on making self-driving laboratories a reality. Furthermore, this review highlights several future applications of ML yet to be fully exploited to advance drug formulation research and development.

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

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

Efficient degassing and ppm-level oxygen monitoring flow chemistry system

Low oxygen levels are critical for a long range of chemical transformations carried out in both flow and batch chemistry. Here, we present an inline continuous flow degassing system based on a gas-permeable membrane inside a vacuum chamber for achieving and monitoring ppm-level oxygen concentrations in solutions. The oxygen presence was monitored with a molecular oxygen probe and a continuously running UV-vis spectrometer. An automated setup for discovering optimal reaction conditions for minimal oxygen presence was devised. The parameters tested were: flow rate, vacuum pressure, solvent back-pressure, tube material, tube length and solvent oxygen solubility. The inline degassing system was proven to be effective in removing up to 99.9% of ambient oxygen from solvents at a flow rate of 300 μl min^-1 and 4 mbar vacuum pressure inside the degassing chamber. Reaching lower oxygen concentrations was limited by gas permeation in the tubing following the degassing unit, which could be addressed by purging large volume flow reactors with an inert gas after degassing or by using tubing with lower gas permeability, such as stainless steel tubing. Among all factors, oxygen solubility in solvents was found to play a significant role in achieving efficient degassing of solvents. The data presented here can be used to choose optimal experimental parameters for oxygen-sensitive reactions in flow chemistry reaction setups. The data were also fitted to an analytically derived model from simple differential equations in physical context of the experiment.

Generative Models as an Emerging Paradigm in the Chemical Sciences

Traditional computational approaches to design chemical species are limited by the need to compute properties for a vast number of candidates, e.g., by discriminative modeling. Therefore, inverse design methods aim to start from the desired property and optimize a corresponding chemical structure. From a machine learning viewpoint, the inverse design problem can be addressed through so-called generative modeling. Mathematically, discriminative models are defined by learning the probability distribution function of properties given the molecular or material structure. In contrast, a generative model seeks to exploit the joint probability of a chemical species with target characteristics. The overarching idea of generative modeling is to implement a system that produces novel compounds that are expected to have a desired set of chemical features, effectively sidestepping issues found in the forward design process. In this contribution, we overview and critically analyze popular generative algorithms like generative adversarial networks, variational autoencoders, flow, and diffusion models. We highlight key differences between each of the models, provide insights into recent success stories, and discuss outstanding challenges for realizing generative modeling discovered solutions in chemical applications.

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

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

Meta optimization based on real-time benchmarking of multiple surrogate models for autonomous flow synthesis

Optimizing a wide range of reaction parameters, steps, and pathways is currently considered one of the most complex and challenging problems in microflow-based organic synthesis. As a novel solution, Bayesian optimization (BO) has been utilized to efficiently guide the optimized conditions of flow reactors; however, the benchmarking process for selecting the optimal model among various surrogate models remains inefficient. In this work, we report meta optimization (MO) by benchmarking multiple surrogate models in real-time without any pre-work, which is realized by evaluating the expected values obtained by the regressor used to build each surrogate model, enabling efficient optimization of reaction conditions. By the comparison of the performance of MO with that of various BOs on four datasets of different flow syntheses, it was verified that MO consistently performs the best-in-class for all emulators developed through machine learning, while the conventional BOs based on surrogate models such as the Gaussian process, random forest, neural network ensemble, and gradient boosting demonstrated varying performances from each emulator, which implies that benchmarking is required.

Enabling Technologies in Carbohydrate Chemistry: Automated Glycan Assembly, Flow Chemistry and Data Science

The synthesis of defined oligosaccharides is a complex task. Several enabling technologies have been introduced in the last two decades to facilitate synthetic access to these valuable biomolecules. In this concept, we describe the technological solutions that have advanced glycochemistry using automated glycan assembly, flow chemistry and data science as examples. We highlight how the synergies between these different technologies can further advance the field, with progress toward the realization of a self-driving lab for glycan synthesis.

Chemical reaction networks and opportunities for machine learning

Chemical reaction networks (CRNs), defined by sets of species and possible reactions between them, are widely used to interrogate chemical systems. To capture increasingly complex phenomena, CRNs can be leveraged alongside data-driven methods and machine learning (ML). In this Perspective, we assess the diverse strategies available for CRN construction and analysis in pursuit of a wide range of scientific goals, discuss ML techniques currently being applied to CRNs and outline future CRN-ML approaches, presenting scientific and technical challenges to overcome.