# Sustainable Materials Discovery in the Era of Artificial Intelligence

arXiv: https://arxiv.org/abs/2601.21527 · HTML: https://mlsanity.com/paper/2601.21527/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 110 references that are in the map.
- First posted on arXiv 29 Jan 2026.

## Builds on (most cited first, 20 of 20)

- [Extracting Accurate Materials Data from Research Papers with Conversational Language Models and Prompt Engineering](https://mlsanity.com/paper/2303.05352/) (2023)
- [14 Examples of How LLMs Can Transform Materials Science and Chemistry: A Reflection on a Large Language Model Hackathon](https://mlsanity.com/paper/2306.06283/) (2023)
- [LLMs4OL: Large Language Models for Ontology Learning](https://mlsanity.com/paper/2307.16648/) (2023)
- [All-atom Diffusion Transformers: Unified generative modelling of molecules and materials](https://mlsanity.com/paper/2503.03965/) (2025)
- [ChemVLM: Exploring the Power of Multimodal Large Language Models in Chemistry Area](https://mlsanity.com/paper/2408.07246/) (2024)
- [Are LLMs Ready for Real-World Materials Discovery?](https://mlsanity.com/paper/2402.05200/) (2024)
- [MatSci-NLP: Evaluating Scientific Language Models on Materials Science Language Tasks Using Text-to-Schema Modeling](https://mlsanity.com/paper/2305.08264/) (2023)
- [Polymer Informatics with Multi-Task Learning](https://mlsanity.com/paper/2010.15166/) (2020)
- [Materials science in the era of large language models: a perspective](https://mlsanity.com/paper/2403.06949/) (2024)
- [Inorganic Materials Synthesis Planning with Literature-Trained Neural Networks](https://mlsanity.com/paper/1901.00032/) (2018)
- [Multimodal Foundation Models for Material Property Prediction and Discovery](https://mlsanity.com/paper/2312.00111/) (2023)
- [Foundational Large Language Models for Materials Research](https://mlsanity.com/paper/2412.09560/) (2024)
- [Looking Through Glass: Knowledge Discovery from Materials Science Literature using Natural Language Processing](https://mlsanity.com/paper/2101.01508/) (2021)
- [Polymers for Extreme Conditions Designed Using Syntax-Directed Variational Autoencoders](https://mlsanity.com/paper/2011.02551/) (2020)
- [MatSciRE: Leveraging Pointer Networks to Automate Entity and Relation Extraction for Material Science Knowledge-base Construction](https://mlsanity.com/paper/2401.09839/) (2024)
- [Bioplastic Design using Multitask Deep Neural Networks](https://mlsanity.com/paper/2203.12033/) (2022)
- [Space Group Conditional Flow Matching](https://mlsanity.com/paper/2509.23822/) (2025)
- [Surrogate Modeling and Explainable Artificial Intelligence for Complex Systems: A Workflow for Automated Simulation Exploration](https://mlsanity.com/paper/2510.16742/) (2025)
- [Energy & Force Regression on DFT Trajectories is Not Enough for Universal Machine Learning Interatomic Potentials](https://mlsanity.com/paper/2502.03660/) (2025)
- [Ontology-aligned structuring and reuse of multimodal materials data and workflows towards automatic reproduction](https://mlsanity.com/paper/2601.12582/) (2026)

## Abstract (arXiv)

Artificial intelligence (AI) has transformed materials discovery, enabling rapid exploration of chemical space through generative models and surrogate screening. Yet current generative AI models for materials discovery, which now drive exploration of vast chemical and structural spaces, optimize candidates exclusively for structural stability and functional properties, with no integration of environmental assessment at any stage of the design loop. Prospective and ex-ante life cycle assessment methods exist and have been applied to emerging technologies, but they operate as standalone downstream analyses, not as active constraints within generative or active-learning pipelines. The result is that environmental feedback, even when produced, arrives after design decisions have been made rather than informing them. The disconnect between atomic-scale design and lifecycle assessment (LCA) reflects fundamental challenges: (i) data scarcity across heterogeneous sources, (ii) scale gaps from atoms to industrial systems, (iii) uncertainty in synthesis pathways, and (iv) the absence of frameworks that co-optimize performance with environmental impact. In this Perspective, we propose integrating upstream ML-assisted materials discovery with downstream LCA into the ML-LCA framework, comprising five components: information extraction for building materials-environment knowledge bases, harmonized databases linking properties to sustainability metrics, multi-scale models bridging atomic properties to lifecycle impacts, ensemble prediction of manufacturing pathways with uncertainty quantification, and uncertainty-aware optimization enabling simultaneous performance-sustainability navigation. Case studies spanning polymers, glass, photoresists, and cement demonstrate both necessity and feasibility while identifying material-specific integration challenges.