About

Tonio Buonassisi is a Professor of Mechanical Engineering at the Massachusetts Institute of Technology (MIT). His research focuses on the application of artificial intelligence to develop new materials for societally beneficial applications, with particular emphasis on solar photovoltaics and technoeconomic analysis. His work has contributed to technology developments in numerous companies and has earned him several prestigious awards, including a US Presidential Early Career Award for Scientists and Engineers (PECASE), a National Science Foundation CAREER Award, and a Google Faculty Award. He directs the ADDEPT Center, a DOE-funded national center dedicated to making semi-transparent perovskite solar cells durable for terrestrial tandem applications. Additionally, he is the PI of the Accelerated Materials Lab for Sustainability (AMLS) at MIT and has served as the founding director of the Accelerated Materials Development for Manufacturing Programme in Singapore. Buonassisi is recognized for his dedication to education, evidenced by the MIT Everett Moore Baker Memorial Award for Excellence in Undergraduate Teaching and the widespread viewership of his OpenCourseware/YouTube lectures on photovoltaic technology.

Research topics

• Computer Science
• Artificial Intelligence
• Machine Learning
• Materials science
• Chemistry
• Nanotechnology
• Inorganic chemistry
• Physics
• Management science
• Telecommunications
• Political Science
• Data science
• Engineering
• Biological system
• Organic chemistry
• Philosophy
• Mechanics
• Engineering management
• Optoelectronics
• Knowledge management
• Mathematics
• Epistemology
• Engineering ethics
• Chemical engineering

Selected publications

• Multi-variable batch Bayesian optimization in materials research: Synthetic data analysis of noise sensitivity and problem landscape effects

  Journal of materials research/Pratt's guide to venture capital sources · 2026-02-26 · 1 citations

  articleOpen access

  Abstract Bayesian optimization (BO) holds promise for accelerating materials science research; however, it faces challenges with high-dimensional inputs and experimental noise in real-world problems. This study addresses these issues by benchmarking batch BO on two synthetic six-variable optimization tasks at varying noise levels: a needle-in-a-haystack task (Ackley function) representing rare materials properties, and a smooth landscape (Hartmann function) simulating process optimization. We evaluate key BO strategies, including acquisition functions, batch-picking methods, and exploration hyperparameter tuning, while presenting a framework for tracking high-dimensional optimization progress. Results show optimization outcomes are highly sensitive to noise levels and landscape shapes. This information enables the design of robust materials optimization campaigns with pre-planned experimental budgets that account for real-world uncertainties. Our methodology facilitates greater BO utilization in experimental materials research, particularly for multi-variable optimization problems, by providing practical guidance for configuring BO campaigns in challenging scientific applications. Graphical abstract

  PublisherOA PDFDOI

• Science acceleration and accessibility with self-driving labs

  Nature Communications · 2025-04-24 · 55 citations

  reviewOpen access

  In the evolving landscape of scientific research, the complexity of global challenges demands innovative approaches to experimental planning and execution. Self-Driving Laboratories (SDLs) automate experimental tasks in chemical and materials sciences and the design and selection of experiments to optimize research processes and reduce material usage. This perspective explores improving access to SDLs via centralized facilities and distributed networks. We discuss the technical and collaborative challenges in realizing SDLs’ potential to enhance human–machine and human–human collaboration, ultimately fostering a more inclusive research community and facilitating previously untenable research projects. Collaborative self-diving research is crucial to research acceleration amidst ever more complex problems. Here, authors identify the key challenges to the dual cultivation of centralised self-driving user facilities and networks of self-driving labs.

  PublisherOA PDFDOI

• Archerfish: a retrofitted 3D printer for high-throughput combinatorial experimentation <i>via</i> continuous printing

  Digital Discovery · 2025-01-01 · 7 citations

  articleOpen accessSenior authorCorresponding

  Archerfish is a low-cost, high-throughput tool for combinatorial materials research. Retrofitted with in situ mixing, Archerfish prints 250 unique compositions per min—a 100× acceleration factor—for aqueous, nanoparticle, and crystalline materials.

  PublisherOA PDFDOI

• High-throughput micro-scale bandgap mapping for perovskite-inspired materials with complex composition space

  Nature Communications · 2025-08-12 · 5 citations

  articleOpen accessSenior author

  Abstract To realize the full promise of high-throughput experimental workflows, the rate of sample synthesis must be matched by that of characterization. Of growing interest are contactless optical techniques that can rapidly measure material homogeneity and properties. Here, we present a hyperspectral imaging method to measure local optical bandgap distributions within samples, utilizing spatially-resolved reflectance spectra coupled with automated data analysis. We collect approximately one million optical bandgap data across the compositional space of Cs 3 (Bi x Sb 1- x ) 2 (Br y I 1- y ) 9 perovskite-inspired materials. Our results show non-monotonic bandgap variations (i.e., bandgap bowing) along six composition gradient sequences, in addition to identifying samples with multiple bandgaps in statistics. High-throughput transient absorption spectroscopy reveals that within these compositions, the depletion of the ground state carriers to excited states occurred at discrete energy levels with independent carrier dynamics, consistent with the bandgap observation and indicative of phase separation. This work demonstrates the potential for rapid optical measurements to assess material quality and homogeneity in a high-throughput experimental setting, supporting screening and recipe optimization of optoelectronic material candidates with desired carrier dynamics and optical properties.

  PublisherOA PDFDOI

• A tomographic interpretation of structure-property relations for materials discovery

  ArXiv.org · 2025-01-30 · 2 citations

  preprintOpen accessSenior author

  Recent advancements in machine learning (ML) for materials have demonstrated that "simple" materials representations (e.g., the chemical formula alone without structural information) can sometimes achieve competitive property prediction performance in common-tasks. Our physics-based intuition would suggest that such representations are "incomplete", which indicates a gap in our understanding. This work proposes a tomographic interpretation of structure-property relations of materials to bridge that gap by defining what is a material representation, material properties, the material and the relationships between these three concepts using ideas from information theory. We verify this framework performing an exhaustive comparison of property-augmented representations on a range of material's property prediction objectives, providing insight into how different properties can encode complementary information.

  PublisherOA PDFDOI

• A closed-loop AI framework for hypothesis-driven and interpretable materials design

  ArXiv.org · 2025-09-23 · 1 citations

  preprintOpen accessSenior author

  Scientific hypothesis generation is central to materials discovery, yet current approaches often emphasize either conceptual (idea-to-data) reasoning or data-driven (data-to-idea) analysis, rarely achieving an effective integration of both. Here, we present a generalizable active learning workflow that integrates top-down, theory-driven hypothesis generation, guided by a large language model. This is complemented by bottom-up, data-driven hypothesis testing through a root-cause association study. We demonstrate this approach through the design of equimolar quinary-cation two-dimensional perovskite, a chemically complex system with over 850,000 possible cation combinations. In the top-down component, the large language model drives closed-loop optimization by proposing candidates that are likely to achieve phase purity, leveraging domain knowledge and chain-of-thought reasoning. With each iteration, the model identifies an increasing number of near phase-pure compositions, sampling less than 0.004% of the design space. In parallel, the bottom-up association study identifies molecular features with statistically significant influences on phase purity. The integration of these approaches enables the convergence of conceptual and statistical hypotheses, leading to generalizable and rational design rules for phase-pure quinary-cation two-dimensional perovskites. As a proof of concept, we applied the optimized phase-pure quinary-cation two-dimensional perovskite film as a surface capping layer in perovskite solar cells, achieving good performance and stability. Our framework enables the development of interpretable and generalizable design rules that are applicable to a wide range of optimization processes within complex design spaces, providing a foundational strategy for rational, scalable, and efficient materials discovery.

  PublisherOA PDFDOI

• Long-term research and design strategies for fusion energy materials

  Matter · 2024-12-01 · 6 citations

  articleSenior author
  PublisherDOI

• Flexible batch electrodialysis for low-cost solar-powered brackish water desalination

  Nature Water · 2024-03-26 · 44 citations

  articleOpen access

  Abstract Globally, 1.6 billion people in rural regions face water scarcity. Expanding freshwater access via brackish groundwater desalination can provide additional resources to address this challenge. In this study, we have developed a time-variant electrodialysis reversal (EDR) technology that flexibly uses available solar energy for desalination. Our proposed photovoltaic-powered desalination system can vary pumping and EDR power to match the availability of intermittent solar power, maximizing the desalination rate. Our results show improved system performance with the direct use of 77% of available solar energy—91% more than in conventional systems—and a 92% reduction in battery reliance. In a village-scale desalination case study in India, these system improvements lead to a 22% reduction in water cost, making the technology competitive with the currently used on-grid, village-scale reverse osmosis systems that are mainly powered by fossil fuels. Future advances could further reduce costs, providing an improved, sustainable solution to water scarcity in remote areas.

  PublisherOA PDFDOI

• Machine Learning Accelerates Innovation in Perovskite Manufacturing Scale-up (Final Technical Report (FTR))

  2024-10-19

  reportOpen access1st authorCorresponding

  We propose to address the challenge of the vast parameter space associated with perovskite manufacturing optimization, by developing a machine learning (ML)-assisted optimization framework for a scalable perovskite PV manufacturing tool. This framework will be interpretable, sequential, and rapidly adaptable to upgraded systems (e.g., via transfer learning). The tool is an open-air rapid spray plasma process (RSPP) of perovskite films, which has already been established at Stanford and is a unique platform to test and deploy the proposed ML-guided framework because the RSPP technique is able to conduct optimization experiments with a high throughput, and easily adjust a wide range of process variables.

  PublisherDOI

• Transfer Learning for Material Parameter Extraction from Current-Voltage Characteristics of Solar Cells

  SSRN Electronic Journal · 2024-01-01

  preprintOpen access
  PublisherDOI

Recent grants

• EPAS: Hierarchical Characterization of Optoelectronic Hyperdoped Silicon Devices for Terawatt-Scale Photovoltaics

  NSF · $370k · 2011–2015

• CAREER:Toward robust, scalable, and non-intermittent solar power: Silicon-based multijunction devices with integrated photocatalysis

  NSF · $400k · 2012–2017

• Collaborative Research: SusChEM: Air-stable, high-lifetime bismuth compounds as solar absorbers with perovskite-like band structures

  NSF · $300k · 2016–2020

Frequent coauthors

• Ian Marius Peters

  150 shared

• Barry Lai

  126 shared

• Riley E. Brandt

  Massachusetts Institute of Technology

  125 shared

• Zekun Ren

  124 shared

• Shijing Sun

  University of Washington

  115 shared

• Roy G. Gordon

  Harvard University

  108 shared

• Noor Titan Putri Hartono

  Helmholtz-Zentrum Berlin für Materialien und Energie

  105 shared

• Sin Cheng Siah

  Massachusetts Institute of Technology

  91 shared

Labs

• MIT Mechanical EngineeringPI

Awards & honors

• Presidential Early Career Award for Scientists and Engineers…
• NSF CAREER Award (2012)
• BOSCH Energy Research Network Award (2012)
• European Materials Research Society (E-MRS) Young Scientist…
• German Academic Exchange Service (DAAD) Graduate Research Fe…