Search for: All records

Discovering novel molecules with targeted properties remains a formidable challenge in materials science, often likened to finding a needle in a haystack. Traditional experimental approaches are slow, costly, and inefficient. In this study, we present an inverse design framework based on a molecular graph conditional variational autoencoder (CVAE) that enables the generation of new molecules with user-specified optical properties, particularly molar extinction coefficient ($$\varepsilon$$). Our model encodes molecular graphs, derived from SMILES strings, into a structured latent space, and then decodes them into valid molecular structures conditioned on a target $$\varepsilon$$ value. Trained on a curated dataset of known molecules with corresponding extinction coefficients, the CVAE learns to generate chemically valid structures, as verified by RDKit. Subsequent Density Functional Theory (DFT) simulations confirm that many of the generated molecules exhibit the electronic structures similar to those molecules with desired $$\varepsilon$$ values. We have also verified the $$\varepsilon$$ values of the generated molecules using a graph neural network (GNN) and the synthesizability of those molecules using an open-source module named ASKCOS. This approach demonstrates the potential of CVAEs to accelerate molecular discovery by enabling user-guided, property-driven molecule generation -- offering a scalable, data-driven alternative to traditional trial-and-error synthesis. 

Abstract The stress field perturbation caused by magmatic intrusions within volcanic systems induces strain in the surrounding region. This effect results in the opening and closing of microcracks in the vicinity of the intrusion, which can affect regional seismic velocities. In late November 2023, we deployed a distributed acoustic sensing interrogator to convert an existing 100‐km telecommunication fiber‐optic cable along the coast of Iceland's Reykjanes peninsula into a dense seismic array, which has run continuously. Measuring changes in surface wave moveout with ambient noise cross‐correlation, we observe up to 2% changes in Rayleigh wave phase velocity following eruptions in the peninsula's 2023–2024 sequence that are likely associated with magmatic intrusions into the eruption‐feeding dike. We apply a Bayesian inversion to compute the posterior distribution of potential dike opening models for each eruption by considering measurements for varying channel pairs and frequency bands, and assuming this velocity change is tied to volumetric strain associated with dike‐opening. Our results are in agreement with those based on geodetic measurement and provide independent constraints on the depth of the dike, demonstrating the viability of this novel inversion and new volcano monitoring directions through fiber sensing. 

Predicting the behavior of nanomaterials under various conditions presents a significant challenge due to their complex microstructures. While high-fidelity modeling techniques, such as molecular dynamics (MD) simulations, are effective, they are also computationally demanding. Machine learning (ML) models have opened new avenues for the rapid exploration of design spaces. In this work, we developed a deep learning framework based on a conditional generative adversarial network (cGAN) to predict the evolution of grain boundary (GB) networks in nanocrystalline materials under mechanical loads, incorporating both morphological and atomic details. We conducted MD simulations on nanocrystalline tungsten and used the resulting ground-truth data to train our cGAN model. We assessed the performance of our cGAN model by comparing it to a Convolutional Autoencoder (ConvAE) model and examining the impact of changes in geometric morphology and loading conditions on the model's performance. Our cGAN model demonstrated high accuracy in predicting GB network evolution under a variety of conditions. This developed framework shows potential for predicting various materials' behaviors across a wide range of nanomaterials.