Autonomous nanomanufacturing of lead-free metal halide perovskite nanocrystals using a self-driving fluidic lab.

Lead-based metal halide perovskite (MHP) nanocrystals (NCs) have emerged as a promising class of semiconducting nanomaterials for a wide range of optoelectronic and photoelectronic applications. However, the intrinsic lead toxicity of MHP NCs has significantly hampered their large-scale device applications. Copper-base MHP NCs with composition-tunable optical properties have emerged as a prominent lead-free MHP NC candidate. However, comprehensive synthesis space exploration, development, and synthesis science studies of copper-based MHP NCs have been limited by the manual nature of flask-based synthesis and characterization methods. In this study, we present an autonomous approach for the development of lead-free MHP NCs via seamless integration of a modular microfluidic platform with machine learning-assisted NC synthesis modeling and experiment selection to establish a self-driving fluidic lab for accelerated NC synthesis science studies. For the first time, a successful and reproducible in-flow synthesis of Cs3Cu2I5 NCs is presented. Autonomous experimentation is then employed for rapid in-flow synthesis science studies of Cs3Cu2I5 NCs. The autonomously generated experimental NC synthesis dataset is then utilized for fast-tracked synthetic route optimization of high-performing Cs3Cu2I5 NCs.

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.

Exact Gaussian processes for massive datasets via non-stationary sparsity-discovering kernels.

A Gaussian Process (GP) is a prominent mathematical framework for stochastic function approximation in science and engineering applications. Its success is largely attributed to the GP's analytical tractability, robustness, and natural inclusion of uncertainty quantification. Unfortunately, the use of exact GPs is prohibitively expensive for large datasets due to their unfavorable numerical complexity of [Formula: see text] in computation and [Formula: see text] in storage. All existing methods addressing this issue utilize some form of approximation-usually considering subsets of the full dataset or finding representative pseudo-points that render the covariance matrix well-structured and sparse. These approximate methods can lead to inaccuracies in function approximations and often limit the user's flexibility in designing expressive kernels. Instead of inducing sparsity via data-point geometry and structure, we propose to take advantage of naturally-occurring sparsity by allowing the kernel to discover-instead of induce-sparse structure. The premise of this paper is that the data sets and physical processes modeled by GPs often exhibit natural or implicit sparsities, but commonly-used kernels do not allow us to exploit such sparsity. The core concept of exact, and at the same time sparse GPs relies on kernel definitions that provide enough flexibility to learn and encode not only non-zero but also zero covariances. This principle of ultra-flexible, compactly-supported, and non-stationary kernels, combined with HPC and constrained optimization, lets us scale exact GPs well beyond 5 million data points.

Engineering Gelation Kinetics in Living Silk Hydrogels by Differential Dynamic Microscopy Microrheology and Machine Learning.

Microbes embedded in hydrogels comprise one form of living material. Discovering formulations that balance potentially competing for mechanical and biological properties in living hydrogels-for example, gel time of the hydrogel formulation and viability of the embedded organisms-can be challenging. In this study, a pipeline is developed to automate the characterization of the gel time of hydrogel formulations. Using this pipeline, living materials comprised of enzymatically crosslinked silk and embedded E. coli-formulated from within a 4D parameter space-are engineered to gel within a pre-selected timeframe. Gelation time is estimated using a novel adaptation of microrheology analysis using differential dynamic microscopy (DDM). In order to expedite the discovery of gelation regime boundaries, Bayesian machine learning models are deployed with optimal decision-making under uncertainty. The rate of learning is observed to vary between artificial intelligence (AI)-assisted planning and human planning, with the fastest rate occurring during AI-assisted planning following a round of human planning. For a subset of formulations gelling within a targeted timeframe of 5-15 min, fluorophore production within the embedded cells is substantially similar across treatments, evidencing that gel time can be tuned independent of other material properties-at least over a finite range-while maintaining biological activity.

Using simulation to accelerate autonomous experimentation: A case study using mechanics.

Autonomous experimentation (AE) accelerates research by combining automation and machine learning to perform experiments intelligently and rapidly in a sequential fashion. While AE systems are most needed to study properties that cannot be predicted analytically or computationally, even imperfect predictions can in principle be useful. Here, we investigate whether imperfect data from simulation can accelerate AE using a case study on the mechanics of additively manufactured structures. Initially, we study resilience, a property that is well-predicted by finite element analysis (FEA), and find that FEA can be used to build a Bayesian prior and experimental data can be integrated using discrepancy modeling to reduce the number of needed experiments ten-fold. Next, we study toughness, a property not well-predicted by FEA and find that FEA can still improve learning by transforming experimental data and guiding experiment selection. These results highlight multiple ways that simulation can improve AE through transfer learning.

Accelerated AI development for autonomous materials synthesis in flow.

Autonomous robotic experimentation strategies are rapidly rising in use because, without the need for user intervention, they can efficiently and precisely converge onto optimal intrinsic and extrinsic synthesis conditions for a wide range of emerging materials. However, as the material syntheses become more complex, the meta-decisions of artificial intelligence (AI)-guided decision-making algorithms used in autonomous platforms become more important. In this work, a surrogate model is developed using data from over 1000 in-house conducted syntheses of metal halide perovskite quantum dots in a self-driven modular microfluidic material synthesizer. The model is designed to represent the global failure rate, unfeasible regions of the synthesis space, synthesis ground truth, and sampling noise of a real robotic material synthesis system with multiple output parameters (peak emission, emission linewidth, and quantum yield). With this model, over 150 AI-guided decision-making strategies within a single-period horizon reinforcement learning framework are automatically explored across more than 600 000 simulated experiments - the equivalent of 7.5 years of continuous robotic operation and 400 L of reagents - to identify the most effective methods for accelerated materials development with multiple objectives. Specifically, the structure and meta-decisions of an ensemble neural network-based material development strategy are investigated, which offers a favorable technique for intelligently and efficiently navigating a complex material synthesis space with multiple targets. The developed ensemble neural network-based decision-making algorithm enables more efficient material formulation optimization in a no prior information environment than well-established algorithms.

Artificial Chemist: An Autonomous Quantum Dot Synthesis Bot.

The optimal synthesis of advanced nanomaterials with numerous reaction parameters, stages, and routes, poses one of the most complex challenges of modern colloidal science, and current strategies often fail to meet the demands of these combinatorially large systems. In response, an Artificial Chemist is presented: the integration of machine-learning-based experiment selection and high-efficiency autonomous flow chemistry. With the self-driving Artificial Chemist, made-to-measure inorganic perovskite quantum dots (QDs) in flow are autonomously synthesized, and their quantum yield and composition polydispersity at target bandgaps, spanning 1.9 to 2.9 eV, are simultaneously tuned. Utilizing the Artificial Chemist, eleven precision-tailored QD synthesis compositions are obtained without any prior knowledge, within 30 h, using less than 210 mL of total starting QD solutions, and without user selection of experiments. Using the knowledge generated from these studies, the Artificial Chemist is pre-trained to use a new batch of precursors and further accelerate the synthetic path discovery of QD compositions, by at least twofold. The knowledge-transfer strategy further enhances the optoelectronic properties of the in-flow synthesized QDs (within the same resources as the no-prior-knowledge experiments) and mitigates the issues of batch-to-batch precursor variability, resulting in QDs averaging within 1 meV from their target peak emission energy.

A Bayesian experimental autonomous researcher for mechanical design.

While additive manufacturing (AM) has facilitated the production of complex structures, it has also highlighted the immense challenge inherent in identifying the optimum AM structure for a given application. Numerical methods are important tools for optimization, but experiment remains the gold standard for studying nonlinear, but critical, mechanical properties such as toughness. To address the vastness of AM design space and the need for experiment, we develop a Bayesian experimental autonomous researcher (BEAR) that combines Bayesian optimization and high-throughput automated experimentation. In addition to rapidly performing experiments, the BEAR leverages iterative experimentation by selecting experiments based on all available results. Using the BEAR, we explore the toughness of a parametric family of structures and observe an almost 60-fold reduction in the number of experiments needed to identify high-performing structures relative to a grid-based search. These results show the value of machine learning in experimental fields where data are sparse.

Optimization of a novel biophysical model using large scale in vivo antisense hybridization data displays improved prediction capabilities of structurally accessible RNA regions.

Current approaches to design efficient antisense RNAs (asRNAs) rely primarily on a thermodynamic understanding of RNA-RNA interactions. However, these approaches depend on structure predictions and have limited accuracy, arguably due to overlooking important cellular environment factors. In this work, we develop a biophysical model to describe asRNA-RNA hybridization that incorporates in vivo factors using large-scale experimental hybridization data for three model RNAs: a group I intron, CsrB and a tRNA. A unique element of our model is the estimation of the availability of the target region to interact with a given asRNA using a differential entropic consideration of suboptimal structures. We showcase the utility of this model by evaluating its prediction capabilities in four additional RNAs: a group II intron, Spinach II, 2-MS2 binding domain and glgC 5΄ UTR. Additionally, we demonstrate the applicability of this approach to other bacterial species by predicting sRNA-mRNA binding regions in two newly discovered, though uncharacterized, regulatory RNAs.