A GPT-4 Reticular Chemist for Guiding MOF Discovery

Metal-Organic Frameworks for Sustainable Crop Disease Management: Current Applications, Mechanistic Insights, and Future Challenges

Efficient management of crop diseases and yield enhancement are essential for addressing the increasing food demands due to global population growth. Metal-organic frameworks (MOFs), which have rapidly evolved throughout the 21st century, are notable for their vast surface area, porosity, and adaptability, establishing them as highly effective vehicles for controlled drug delivery. This review methodically categorizes common MOFs employed in crop disease management and details their effectiveness against various pathogens. Additionally, by critically evaluating existing research, it outlines strategic approaches for the design of drug-delivery MOFs and explains the mechanisms through which MOFs enhance disease resistance. Finally, this paper identifies the current challenges in MOF research for crop disease management and suggests directions for future research. Through this in-depth review, the paper seeks to enrich the understanding of MOFs applications in crop disease management and offers valuable insights for researchers and practitioners.

Single and Multi-Hop Question-Answering Datasets for Reticular Chemistry with GPT-4-Turbo

The rapid advancement in artificial intelligence and natural language processing has led to the development of large-scale datasets aimed at benchmarking the performance of machine learning models. Herein, we introduce "RetChemQA", a comprehensive benchmark dataset designed to evaluate the capabilities of such models in the domain of reticular chemistry. This dataset includes both single-hop and multi-hop question-answer pairs, encompassing approximately 45,000 question and answers (Q&As) for each type. The questions have been extracted from an extensive corpus of literature containing about 2,530 research papers from publishers including NAS, ACS, RSC, Elsevier, and Nature Publishing Group, among others. The dataset has been generated using OpenAI's GPT-4 Turbo, a cutting-edge model known for its exceptional language understanding and generation capabilities. In addition to the Q&A dataset, we also release a dataset of synthesis conditions extracted from the corpus of literature used in this study. The aim of RetChemQA is to provide a robust platform for the development and evaluation of advanced machine learning algorithms, particularly for the reticular chemistry community. The dataset is structured to reflect the complexities and nuances of real-world scientific discourse, thereby enabling nuanced performance assessments across a variety of tasks.

Automated electrosynthesis reaction mining with multimodal large language models (MLLMs)

Leveraging the chemical data available in legacy formats such as publications and patents is a significant challenge for the community. Automated reaction mining offers a promising solution to unleash this knowledge into a learnable digital form and therefore help expedite materials and reaction discovery. However, existing reaction mining toolkits are limited to single input modalities (text or images) and cannot effectively integrate heterogeneous data that is scattered across text, tables, and figures. In this work, we go beyond single input modalities and explore multimodal large language models (MLLMs) for the analysis of diverse data inputs for automated electrosynthesis reaction mining. We compiled a test dataset of 65 articles (MERMES-T24 set) and employed it to benchmark five prominent MLLMs against two critical tasks: (i) reaction diagram parsing and (ii) resolving cross-modality data interdependencies. The frontrunner MLLM achieved ≥96% accuracy in both tasks, with the strategic integration of single-shot visual prompts and image pre-processing techniques. We integrate this capability into a toolkit named MERMES (multimodal reaction mining pipeline for electrosynthesis). Our toolkit functions as an end-to-end MLLM-powered pipeline that integrates article retrieval, information extraction and multimodal analysis for streamlining and automating knowledge extraction. This work lays the groundwork for the increased utilization of MLLMs to accelerate the digitization of chemistry knowledge for data-driven research.

Switching from Molecules to Functional Materials: Breakthroughs in Photochromism With MOFs

Photochromic materials with properties that can be dynamically tailored as a function of external stimuli are a rapidly expanding field driven by applications in areas ranging from molecular computing, nanotechnology, or photopharmacology to programable heterogeneous catalysis. Challenges arise, however, when translating the rapid, solution-like response of stimuli-responsive moieties to solid-state materials due to the intermolecular interactions imposed through close molecular packing in bulk solids. As a result, the integration of photochromic compounds into synthetically programable porous matrices, such as metal-organic frameworks (MOFs), has come to the forefront as an emerging strategy for photochromic material development. This review highlights how the core principles of reticular chemistry (on the example of MOFs) play a critical role in the photochromic material performance, surpassing the limitations previously observed in solution or solid state. The symbiotic relationship between photoresponsive compounds and porous frameworks with a focus on how reticular synthesis creates avenues toward tailorable photoisomerization kinetics, directional energy and charge transfer, switchable gas sorption, and synergistic chromophore communication is discussed. This review not only focuses on the recent cutting-edge advancements in photochromic material development, but also highlights novel, vital-to-pursue pathways for multifaceted functional materials in the realms of energy, technology, and biomedicine.

Accelerating the Discovery of Efficient High-Entropy Alloy Electrocatalysts: High-Throughput Experimentation and Data-Driven Strategies

High-entropy alloys (HEAs) present both significant potential and challenges for developing efficient electrocatalysts due to their diverse combinations and compositions. Here, we propose a procedural approach that combines high-throughput experimentation with data-driven strategies to accelerate the discovery of efficient HEA electrocatalysts for the hydrogen evolution reaction (HER). This enables the rapid preparation of HEA arrays with various element combinations and composition ratios within a model system. The intrinsic activity of the HEA arrays is swiftly screened using scanning electrochemical cell microscopy (SECCM), providing precise composition-activity data sets for the HEA system. An ensemble machine learning (EML) model is then used to predict the activity database for the composition subspace of the system. Based on these database results, two groups of promising catalysts are recommended and validated through actual electrocatalytic evaluations. This procedural approach, which combines high-throughput experimentation with data-driven strategies, provides a new pathway to accelerate the discovery of efficient HEA electrocatalysts.

Accelerating the Discovery of Oxygen Reduction Electrocatalysts: High-Throughput Screening of Element Combinations in Pt-Based High-Entropy Alloys

The vast number of element combinations and the explosive growth of composition space pose significant challenges to the development of high-entropy alloys (HEAs). Here, we propose a procedural research method aimed at accelerating the discovery of efficient electrocatalysts for oxygen reduction reaction (ORR) based on Pt-based quinary HEAs. The method begins with an element library provided by a large language model (LLM), combined with microscale precursor printing and pulse high-temperature synthesis techniques to prepare multi-element combination HEA array in one step. Through high-throughput measurement using scanning electrochemical cell microscopy (SECCM), precise identification of highly active HEA element combinations and exploration of composition space for a specific combination are achieved. Advantageous element combinations are further validated in practical electrocatalytic evaluations. The contributions of individual element sites and the synergistic effects among elements of such HEAs in enhancing reaction activity are elucidated via density functional theory (DFT) calculations. This method integrates high-throughput experiments, practical catalyst validation, and DFT calculations, providing a new pathway for accelerating the discovery of efficient multi-element materials in the field of energy catalysis.

Introducing Noble Gas as Space Holder under High Pressure to Design Porous Titanium Carbides with Open Metal Sites for Hydrogen Storage at Near-Ambient Conditions

It has been a long-standing challenge to develop high-performance solid-state hydrogen storage materials operated under near-ambient conditions. In this work, we propose a new strategy of using noble gases for space holding to design porous titanium carbides with abundant open metal sites for hydrogen storage. By using machine learning and graph theory-assisted universal structure searching methods, we obtain 28 porous titanium carbides from three precursors (TiC dimer, C atom, and Kr atom) under 30 GPa of pressure. The stability and hydrogen storage performance of the resulting structures are further assessed and validated through density function theory and grand canonical Monte Carlo simulations with a DFT-fitted force field. Finally, p-TiC2 is identified as a promising quasi-molecular hydrogen storage material with capacity of 4.0 wt % and 106.0 g/L at 230 K and 16 bar.

Machine Learning Optimizing Enzyme/ZIF Biocomposites for Enhanced Encapsulation Efficiency and Bioactivity

In this study, we present the first example of using a machine learning (ML)-assisted design strategy to optimize the synthesis formulation of enzyme/ZIFs (zeolitic imidazolate framework) for enhanced performance. Glucose oxidase (GOx) and horseradish peroxidase (HRP) were chosen as model enzymes, while Zn(eIM)2 (eIM = 2-ethylimidazolate) was selected as the model ZIF to test our ML-assisted workflow paradigm. Through an iterative ML-driven training-design-synthesis-measurement workflow, we efficiently discovered GOx/ZIF (G151) and HRP/ZIF (H150) with their overall performance index (OPI) values (OPI represents the product of encapsulation efficiency (E in %), retained enzymatic activity (A in %), and thermal stability (T in %)) at least 1.3 times higher than those in systematic seed data studies. Furthermore, advanced statistical methods derived from the trained random forest model qualitatively and quantitatively reveal the relationship among synthesis, structure, and performance in the enzyme/ZIF system, offering valuable guidance for future studies on enzyme/ZIFs. Overall, our proposed ML-assisted design strategy holds promise for accelerating the development of enzyme/ZIFs and other enzyme immobilization systems for biocatalysis applications and beyond, including drug delivery and sensing, among others.

A Pillar[5]arene-Containing Metal-Organic Framework for Rapid and Highly Capable Adsorption of a Mustard Gas Simulant

Mustard gas causes irreversible damage upon inhalation or contact with the human body. Consequently, the development of adsorbents for effective interception of mustard gas at low concentrations and high flow rates is an urgent necessity. Here we report a stable porous pillar[5]arene-containing metal-organic framework (MOF) based on zirconium (EtP5-Zr-scu), demonstrating that embedding pillar[5]arene units in MOFs can provide specific binding sites for efficient adsorption of a mustard gas simulant, 2-chloroethyl ethyl sulfide (CEES). EtP5-Zr-scu achieves a higher capacity and more rapid adsorption compared to its counterpart without embedded pillar[5]arene units (H4tcpt-Zr-scu) and perethylated pillar[5]arene (EtP5) alone. Single crystal X-ray diffraction and solid-state nuclear magnetic resonance reveal that the enhanced performance of EtP5-Zr-scu is derived from the host-guest complexation between CEES and pillar[5]arene moieties. Moreover, breakthrough experiments confirmed that the interception performance of EtP5-Zr-scu against CEES (800 ppm, 120 mL/min) was significantly improved (566 min/g) compared with H4tcpt-Zr-scu (353 min/g) and EtP5 (0.873 min/g), attributed to the integration of open channels with specific recognition sites. This work marks a significant advancement in the development of macrocycle-incorporated crystalline framework materials with recognition sites for the efficient capture of guest molecules.

Automation and machine learning augmented by large language models in a catalysis study

Recent advancements in artificial intelligence and automation are transforming catalyst discovery and design from traditional trial-and-error manual mode into intelligent, high-throughput digital methodologies. This transformation is driven by four key components, including high-throughput information extraction, automated robotic experimentation, real-time feedback for iterative optimization, and interpretable machine learning for generating new knowledge. These innovations have given rise to the development of self-driving labs and significantly accelerated materials research. Over the past two years, the emergence of large language models (LLMs) has added a new dimension to this field, providing unprecedented flexibility in information integration, decision-making, and interacting with human researchers. This review explores how LLMs are reshaping catalyst design, heralding a revolutionary change in the fields.

Large-Language-Model-Based AI Agent for Organic Semiconductor Device Research

Large language models (LLMs) have attracted widespread attention recently, however, their application in specialized scientific fields still requires deep adaptation. Here, an artificial intelligence (AI) agent for organic field-effect transistors (OFETs) is designed by integrating the generative pre-trained transformer 4 (GPT-4) model with well-trained machine learning (ML) algorithms. It can efficiently extract the experimental parameters of OFETs from scientific literature and reshape them into a structured database, achieving precision and recall rates both exceeding 92%. Combined with well-trained ML models, this AI agent can further provide targeted guidance and suggestions for device design. With prompt engineering and human-in-loop strategies, the agent extracts sufficient information of 709 OFETs from 277 research articles across different publishers and gathers them into a standardized database containing more than 10 000 device parameters. Using this database, a ML model based on Extreme Gradient Boosting is trained for device performance judgment. Combined with the interpretation of the high-precision model, the agent has provided a feasible optimization scheme that has tripled the charge transport properties of 2,6-diphenyldithieno[3,2-b:2',3'-d]thiophene OFETs. This work is an effective practice of LLMs in the field of organic optoelectronic devices and expands the research paradigm of organic optoelectronic materials and devices.

Large Language Models for Inorganic Synthesis Predictions

We evaluate the effectiveness of pretrained and fine-tuned large language models (LLMs) for predicting the synthesizability of inorganic compounds and the selection of precursors needed to perform inorganic synthesis. The predictions of fine-tuned LLMs are comparable to─and sometimes better than─recent bespoke machine learning models for these tasks but require only minimal user expertise, cost, and time to develop. Therefore, this strategy can serve both as an effective and strong baseline for future machine learning studies of various chemical applications and as a practical tool for experimental chemists.

Solving Advanced Task-Specific Problems in Measurement Sciences with Generative AI

The Generative Pre-Trained Transformer known as ChatGPT-4 has undergone extensive pretraining on a diverse data set, enabling it to generate coherent and contextually relevant text based on the input it receives. This capability allows it to perform tasks from answering questions and has attracted significant interest in material sciences, synthetic chemistry, and drug discovery. In this work, we posed four advanced task-specific problems to ChatGPT, which were recently published in leading journals for topics in analytical chemistry, spectroscopy, bioimage super-resolution, and electrochemistry. ChatGPT-4 successfully implemented the four ideas after assigning the "persona" to the AI and posing targeted questions. We show two cases where "unguided" ChatGPT could complete the assignments with minimal human direction. The construction of a microwave spectrum from a free induction curve and super-resolution of bioimages was accomplished using this approach. Two other specific tasks, correcting a complex baseline with morphological operations of set theory and estimating the diffusion current, required additional input, e.g., equations and specific directions from the user. In each case, the MATLAB code was eventually generated by ChatGPT-4 even when the original authors did not provide any code themselves. We show that a validation test must be implemented to detect and correct mistakes made by ChatGPT-4, followed by feedback correction. These approaches will pave the way for open and transparent science and eliminate the black boxes in measurement science when it comes to advanced data processing.

Evaluation of Open-Source Large Language Models for Metal-Organic Frameworks Research

Along with the development of machine learning, deep learning, and large language models (LLMs) such as GPT-4 (GPT: Generative Pre-Trained Transformer), artificial intelligence (AI) tools have been playing an increasingly important role in chemical and material research to facilitate the material screening and design. Despite the exciting progress of GPT-4 based AI research assistance, open-source LLMs have not gained much attention from the scientific community. This work primarily focused on metal-organic frameworks (MOFs) as a subdomain of chemistry and evaluated six top-rated open-source LLMs with a comprehensive set of tasks including MOFs knowledge, basic chemistry knowledge, in-depth chemistry knowledge, knowledge extraction, database reading, predicting material property, experiment design, computational scripts generation, guiding experiment, data analysis, and paper polishing, which covers the basic units of MOFs research. In general, these LLMs were capable of most of the tasks. Especially, Llama2-7B and ChatGLM2-6B were found to perform particularly well with moderate computational resources. Additionally, the performance of different parameter versions of the same model was compared, which revealed the superior performance of higher parameter versions.

Multiscale Modeling of Physical Properties of Nanoporous Frameworks: Predicting Mechanical, Thermal, and Adsorption Behavior

ConspectusNanoporous frameworks are a large and diverse family of supramolecular materials, whose chemical building units (organic, inorganic, or both) are assembled into a 3D architecture with well-defined connectivity and topology, featuring intrinsic porosity. These materials play a key role in various industrial processes and applications, such as energy production and conversion, fluid separation, gas storage, water harvesting, and many more. The performance and suitability of nanoporous materials for each specific application are directly related to both their physical and chemical properties, and their determination is crucial for process engineering and optimization of performances. In this Account, we focus on some recent developments in the multiscale modeling of physical properties of nanoporous frameworks, highlighting the latest advances in three specific areas: mechanical properties, thermal properties, and adsorption.In the study of the mechanical behavior of nanoporous materials, the past few years have seen a rapid acceleration of research. For example, computational resources have been pooled to create a public large-scale database of elastic constants as part of the Materials Project initiative to accelerate innovation in materials research: those can serve as a basis for data-based discovery of materials with targeted properties, as well as the training of machine learning predictor models.The large-scale prediction of thermal behavior, in comparison, is not yet routinely performed at such a large scale. Tentative databases have been assembled at the DFT level on specific families of materials, such as zeolites, but prediction at larger scale currently requires the use of transferable classical force fields, whose accuracy can be limited.Finally, adsorption is naturally one of the most studied physical properties of nanoporous frameworks, as fluid separation or storage is often the primary target for these materials. We highlight the recent achievements and open challenges for adsorption prediction at a large scale, focusing in particular on the accuracy of computational models and the reliability of comparisons with experimental data available. We detail some recent methodological improvements in the prediction of adsorption-related properties: in particular, we describe the recent research efforts to go beyond the study of thermodynamic quantities (uptake, adsorption enthalpy, and thermodynamic selectivity) and predict transport properties using data-based methods and high-throughput computational schemes. Finally, we stress the importance of data-based methods of addressing all sources of uncertainty.The Account concludes with some perspectives about the latest developments and open questions in data-based approaches and the integration of computational and experimental data together in the materials discovery loop.

Structural Design for EMI Shielding: From Underlying Mechanisms to Common Pitfalls

Modern human civilization deeply relies on the rapid advancement of cutting-edge electronic systems that have revolutionized communication, education, aviation, and entertainment. However, the electromagnetic interference (EMI) generated by digital systems poses a significant threat to the society, potentially leading to a future crisis. While numerous efforts are made to develop nanotechnological shielding systems to mitigate the detrimental effects of EMI, there is limited focus on creating absorption-dominant shielding solutions. Achieving absorption-dominant EMI shields requires careful structural design engineering, starting from the smallest components and considering the most effective electromagnetic wave attenuating factors. This review offers a comprehensive overview of shielding structures, emphasizing the critical elements of absorption-dominant shielding design, shielding mechanisms, limitations of both traditional and nanotechnological EMI shields, and common misconceptions about the foundational principles of EMI shielding science. This systematic review serves as a scientific guide for designing shielding structures that prioritize absorption, highlighting an often-overlooked aspect of shielding science.

Synthetic Strategies to Extended Aromatic Covalent Organic Frameworks

π-Conjugated materials are highly attractive owing to their unique optical and electronic properties. Covalent organic frameworks (COFs) offer a great opportunity for precise arrangement of building units in a π-conjugated crystalline matrix and tuning of the properties through choice of functionalities or post-synthetic modification. With this review, we aim at summarizing both the most representative as well as emerging strategies for the synthesis of π-conjugated COFs. We give examples of direct synthesis using large, π-extended building blocks. COFs featuring fully conjugated linkages such as vinylene, pyrazine, and azole are discussed. Then, post-synthetic modification methods that result in the extension of the COF π-system are reviewed. Throughout, mechanistic insights are presented when available. In the context of their utilization as film devices, we conduct a concise survey of the prominent COF layer deposition techniques reported and their aptness for the deposition of fused aromatic systems.

Analysis and prediction in SCR experiments using GPT-4 with an effective chain-of-thought prompting strategy

This study explores the use of large language models (LLMs) in interpreting and predicting experimental outcomes based on given experimental variables, leveraging the human-like reasoning and inference capabilities of LLMs, using selective catalytic reduction of NOx with NH3 as a case study. We implement the chain of thought (CoT) concept to formulate logical steps for uncovering connections within the data, introducing an "Ordered-and-Structured" CoT (OSCoT) prompting strategy. We compare the OSCoT strategy with the more conventional "One-Pot" CoT (OPCoT) approach and with human experts. We demonstrate that GPT-4, equipped with this new OSCoT prompting strategy, outperforms the other two settings and accurately predicts experimental outcomes and provides intuitive reasoning for its predictions.

Machine Learning-Assisted Discovery of Propane-Selective Metal-Organic Frameworks

Machine learning is gaining momentum in the prediction and discovery of materials for specific applications. Given the abundance of metal-organic frameworks (MOFs), computational screening of the existing MOFs for propane/propylene (C3H8/C3H6) separation could be equally important for developing new MOFs. Herein, we report a machine learning-assisted strategy for screening C3H8-selective MOFs from the CoRE MOF database. Among the four algorithms applied in machine learning, the random forest (RF) algorithm displays the highest degree of accuracy. We experimentally verified the identified top-performing MOF (JNU-90) with its benchmark selectivity and separation performance of directly producing C3H6. Considering its excellent hydrolytic stability, JNU-90 shows great promise in the energy-efficient separation of C3H8/C3H6. This work may accelerate the development of MOFs for challenging separations.

Shaping the Water-Harvesting Behavior of Metal-Organic Frameworks Aided by Fine-Tuned GPT Models

We construct a data set of metal-organic framework (MOF) linkers and employ a fine-tuned GPT assistant to propose MOF linker designs by mutating and modifying the existing linker structures. This strategy allows the GPT model to learn the intricate language of chemistry in molecular representations, thereby achieving an enhanced accuracy in generating linker structures compared with its base models. Aiming to highlight the significance of linker design strategies in advancing the discovery of water-harvesting MOFs, we conducted a systematic MOF variant expansion upon state-of-the-art MOF-303 utilizing a multidimensional approach that integrates linker extension with multivariate tuning strategies. We synthesized a series of isoreticular aluminum MOFs, termed Long-Arm MOFs (LAMOF-1 to LAMOF-10), featuring linkers that bear various combinations of heteroatoms in their five-membered ring moiety, replacing pyrazole with either thiophene, furan, or thiazole rings or a combination of two. Beyond their consistent and robust architecture, as demonstrated by permanent porosity and thermal stability, the LAMOF series offers a generalizable synthesis strategy. Importantly, these 10 LAMOFs establish new benchmarks for water uptake (up to 0.64 g g-1) and operational humidity ranges (between 13 and 53%), thereby expanding the diversity of water-harvesting MOFs.