2025 Call for Seed Grant Proposals

Scientific Area: Earth Systems: Oceans to near Space

Abstract: Climate is changing. As Earth warms, the oceans are projected to lose dissolved oxygen, with countless impacts on the distribution and productivity of global fisheries, the rates of respiration that governs natural carbon sequestration, and the flux of the potent greenhouse gas nitrous oxide from the oceans. Yet, the driving mechanisms escape science, preventing accurate climate predictions and precluding a scientific basis for maintaining robust fishery and conservation policy. Here we present a new mechanism for not only revealing the natural cycles and anthropogenic perturbation of oxygen in the ocean but also for monitoring into the future. We intend to deploy a prototype electronic tag that can measure dissolved oxygen, developed by our Portuguese collaborators, to monitor the ocean environment using sharks as ocean observing platforms. These novel measurements are critical for constructing a dynamic view of ocean change and making predictions for both fisheries management and climate change mitigation.

MIT PI:
Andrew Babbin, Associate Professor, Department of Earth, Atmospheric & Planetary Sciences

PT PI:
Dr. Nuno Queiroz, Principal Researcher Research Centre in Biodiversity and Genetic Resources and the Universidade do Porto

Scientific Area: Earth Systems: Oceans to near Space

Abstract: Near the sea surface, interactions between small-scale ocean processes, turbulence, and Earth’s climate are not fully resolved by either models or observations. However, we can make decisive progress by fusing models with data through innovative AI techniques. We propose to advance understanding by integrating MIT’s ocean models with satellite data from the Azores ESA Lab. Our focus is on two critical regions for climate change, where well developed AI techniques can help address key questions about small-scale processes. In the Azores region, we will forecast thermal fronts, mesoscale eddies, internal waves, and other mixing indices. Additionally, we will train AI to detect and predict internal waves at the equator during La Ninã seasons, where their interactions with thermal fronts have important implications for climate. In both cases, simulation data from MIT models will be used to train generative AI and fine-tune foundation models, which will then be applied to satellite data.

MIT PI:
Gael Forget, Research Scientist, Department of Earth, Atmospheric & Planetary Sciences

PT PI:
Dr. Adriana Ferreira (AIR Centre)
Dr. Jorge M. Magalhães (CIIMAR)
M. João Pinelo (AIR Centre)
Pr. José da Silva (FCUP).

Scientific Area: Earth Systems: Oceans to near Space

Abstract: In this project we proposed to develop algorithms and software for coordinated path planning of multi-robot marine robotic platforms. The goal of this work is to enable effective deployments of large numbers of vehicles, both at the surface and underwater, with very little need for operator control. The new algorithms will leverage MIT’s unique software libraries for multi-objective optimization and decentralized decision-making available in MIT’s online moos-ivp.org open-source project. The initial phase of this project will focus on collaborative surface vehicle autonomy, and the second phase will focus on nested swarm clusters of both surface and underwater vehicles. The end-of-project target experiment is deployment of surface and underwater vehicles in Pico Straight in Horta Portugal. MIT will validate Year 1 algorithms in its 100-node sim cluster and using 10 unmanned surface vehicles at the MIT Marine Autonomy Lab on the Charles River. Prof. Haitong Xu from Instituto Superior Técnico, University of Lisbon, will develop complementary algorithms and visit MIT lab for the Spring Semester 2026, co-testing autonomy capabilities.

MIT PI:
Michael Benjamin, Principal Research Scientist, Department of Mechanical Engineering

PT PI:
Prof. Haitong Xu, Instituto Superior Técnico, University of Lisbon
Prof. Carlos Guedes Soares, Instituto Superior Técnico, University of Lisbon

Scientific Area: Earth Systems: Oceans to near Space

Abstract: To answer urgent questions about climate change, food security, and sustainability, it is necessary to mechanistically understand microscale soil processes. From interviews with experts in soil microbial biogeochemistry, it is clear that current soil characterization techniques provide inadequate spatial resolution, analyte variety, and levels of perturbation to study dynamic processes in the challenging soil environment. We propose a novel sensing platform to detect diverse soil analytes in two dimensions on the microscale, composed of a planar matrix housing whole-cell microbial biosensors that contacts the soil through an engineered membrane interface. This project is divided into three stages: (1) designing and modeling the sensor platform; (2) building and validating three sensors to map three diverse soil analytes (a plant metabolite, a microbial electron acceptor, and a contaminant); and (3) testing the sensors by using them to investigate scientific questions of interest for the three model analytes.

MIT PI:
Rohit N. Karnik, Professor, Department of Mechanical Engineering

PT PI:
Paula Morais
Associate Professor with Habilitation, Department of Life Sciences, Faculty of Sciences and Technology, Laboratory ARISE
University of Coimbra

Scientific Area: Climate Science & Climate Change

Abstract: Tropical forests regulate Earth’s climate, yet deforestation, mega-droughts, windthrows, seed-dispersal collapse and runaway fires threaten to push them beyond tipping points. We will fuse high-resolution remote-sensing data, causal machine-learning and resampling of legacy plots to pinpoint where and why these thresholds are crossed. A “twin-forest paradox” design—Amazonia’s declining sink versus Africa’s stable counterpart (Fig. 1)—isolates the disturbance combinations that flip forests from sink to source. Soil re-cores of 40 Amazon plots, first sampled two decades ago, will be isotopically analysed at INIAV laboratories in Portugal to test whether stressed soils buffer or amplify canopy loss. The project will result in driver-resolved early-warning indicators and a public dashboard ready for climate modeling and REDD+ accounting. Portuguese partners within the TERRA network contribute Europe’s burn-scar atlas, fire-physics labs and Copernicus cloud, ensuring Amazon-derived alerts translate directly into wildfire-resilient management of Mediterranean pine and eucalyptus landscapes.

MIT PI:
Cesar Terrer, Associate Professor, Department of Civil and Environmental Engineering

PT PI:
José Miguel C. Pereira, Professor, Forest Research Centre, School of Agriculture, University of Lisbon.
Ruben Heleno, Associate Professor, Centre for Functional Ecology, University of Coimbra.

Scientific Area: Energy

Abstract: We propose to develop diodes and prototype solar cells using chalcogenide perovskite thin films. We will deposit BaZrS3 thin films, and BaZr(S,Se)3 alloys with tunable band gap, using previouslyestablished methods of molecular beam epitaxy. We will select contact materials, and will develop methods to deposit and evaluate the contact materials as thin films. We will form p-n heterojunction diodes and test their electrical performance. Finally, we will evaluate the photovoltage and solar-cell performance of the top-performing diodes. Our project will build on a new collaboration between MIT and the International Iberian Nanotechnology Laboratory (INL), leveraging world-leading expertise in chalcogenide perovskite deposition (MIT) and in scanning probe microscopy (INL). We will be the first to characterize and optimize chalcogenide perovskite thin-film diodes and photovoltaic performance, which will be a major advance in this growing research field.

MIT PI:
Rafael Jaramillo, Associate Professor, Department of Materials Science and Engineering

PT PI:
Sascha Sadewasser
Principal Investigator, Laboratory for Nanostructured Solar Cells International Iberian Nanotechnology Laboratory (INL)

Scientific Area: Energy

Abstract: The motivation for this project is the recent Iberian blackout. We study it by modeling and simulating multi-country interconnected electric power grid affected by this event using modified publicly available data of European Union grid known as PECASE. We will then use extended AC Optimal Power Flow software to assess the most vulnerable parts of the grid, and, to, consequently, reconstruct the events which led to massive loss of electricity service. Of particular interest will be to understand the role of coordinating inter-countries power exchanges, notably between France and Spain for preventing system voltages from collapsing. In parallel, we will introduce adaptive power electronically switched control of intermittent resources for stabilizing voltage and frequency during such extreme events. The study will set a basis for general framework needed in other parts of the world for the same purposes, including US. Results will be demonstrated using Power Digital Twin at MIT.

MIT PI:
Marija Ilic, Senior Research Scientist, Department of Electrical Engineering and Computer Science

PT PI:
Pedro Carvalho Full Professor, IST Técnico Lisboa, Institute for Systems and Computer Engineering, Research and Development

Scientific Area: Energy

Abstract: Offshore wind energy leverages the high intensity and consistency of oceanic winds, playing a key role in the transition to renewable energy. As energy demands grow, larger turbines are needed to optimize power generation and reduce costs. However, upscaling introduces structural design and manufacturing challenges. Designing better wind turbines is therefore essential. A key challenge is the time-consuming nature of multiphysics simulations, involving interactions between wind, waves, and ocean currents, limiting exploration of design alternatives. While AI provides a promising way to help alleviate this challenge, most current AI-accelerated models lack the capability to capture multi-physics, leading to untrustworthy or structurally invalid designs. We propose a physics-guided generative design framework that combines Graph Neural Nets (GNNs) and diffusion models, trained on high-fidelity multiphysics simulation data and validated against experimental data from full-scale offshore structures, to design and evaluate offshore structures. This approach enables faster development, improved accuracy, and scalable digital twins for Portugal’s partners in the offshore wind sector.

MIT PI:
Faez Ahmed, Associate Professor, Department of Mechanical Engineering

PT PI:
Sérgio Tavares (University of Aveiro),
Filipe Magalhães (University of Porto)

Scientific Area: Energy

Abstract: Boiling of refrigerants enables high heat dissipation and is critical in heat pumps, refrigeration, solar power systems, and emerging microsystems. Understanding refrigerant boiling can drive the development of clean technologies for process heat, cooling, and power while reducing greenhouse gas emissions by replacing harmful refrigerants with eco-friendly alternatives. This project aims to enhance boiling heat transfer using next-generation low-global-warming-potential refrigerants. We will conduct pioneering experiments using advanced optical and infrared diagnostics, paired with engineered surfaces, to identify designs that maximize performance. These surfaces will be tested in prototypical conditions, focusing on applications such as refrigeration systems, concentrated solar power and passive, thermally powered cooling. Insights gained will also benefit broader water-based thermal systems, including steam generation and solar cooking.

MIT PI:
Matteo Bucci, Associate Professor, Department of Nuclear Science & Engineering

PT PI:
Prof. Ana Moita Universidade de Lisboa Instituto Superior Técnico

Scientific Area: Energy

Abstract: Poly(vinyl chloride) (PVC) waste presents a major challenge to circular materials management due to its high chlorine content and incompatibility with existing recycling systems. This project, a collaboration between Prof. Luís Branco (NOVA FCT) and Prof. Yuriy Román (MIT), will develop a catalytic process to convert PVC into energy-dense liquid fuels using mild thermochemical treatment in tailored solvent systems. The approach integrates solvents with tunable acid–base properties (NOVA FCT) and selective hydrogenolysis catalysts (MIT) to enable dechlorination and C–C bond cleavage under mild conditions. Key innovations include nucleophilic, low-volatility solvents that stabilize reactive intermediates and catalysts that promote high selectivity without over-cracking. Unlike existing chemical recycling methods, which suffer from low carbon yields, proof-of-concept results demonstrate dechlorinated PVC carbon recoveries approaching 80%. Portuguese PVC-containing waste streams will be used to validate the process and guide scale-up. This collaboration offers a high-yield, lowimpact route to valorize halogenated plastic waste while supporting national energy and sustainability goals.

MIT PI:
Yuriy Roman, Professor, Department of Chemical Engineering

PT PI:
Prof. Luís Branco, Associate Professor, Department of Chemistry, Principal Investigator, Fotoquímica Research Group NOVA School of Science and Technology (FCT NOVA)

Scientific Area: Chips/Nanotechnology

Abstract: We aim to develop neuromorphic computing systems by exploiting the stochastic nature of nanoscale spintronic devices. Currently neural networks have achieved remarkable success in various application areas. However, existing silicon-based hardware exhibits inefficiency at processing compute-intensive models compared to the human brain. Bio-inspired computing models based on stochastic activations and learning rules provide new opportunities for developing efficient neuromorphic systems with complex cognitive capabilities. In this proposal, we will leverage the intrinsic stochasticity from the compact and energy-efficient spintronic devices for realizing stochastic neuro-mimetic components, and realize algorithms and architecture to achieve balanced performance among accuracy, robustness, synaptic memory requirements, and energy efficiency. Particularly, we will develop stochastic spin-orbit torque magnetic tunnel junctions for both neurons and synapses. Compared to state of the art, the proposed devices and circuits will minimize the influence of noises and device variations, leading to a scalable solution for robust and efficient implementation of stochastic neural networks.

MIT PI:
Luqiao Liu, Associate Professor, Department of Electrical Engineering and Computer Science

PT PI:
Susana Cardoso de Freitas, INESC-MN and Instituto Superior Tecnico, Principal Investigator and Full Professor

Scientific Area: Chips/Nanotechnology

Abstract: Sepsis is a leading cause of death from infection, driven by a dysregulated immune response. There is a need for technology to monitor sepsis progression and guide treatment. Neutrophils are abundant and easily accessible immune cells, and their functional state holds promise as a real-time biomarker for sepsis and other diseases of the immune system. Here we propose to integrate a microfluidic electronic cellular analysis chip developed at MIT called isodielectric separation with a nanoelectronic graphene field-effect transistor chip sensor developed at INL that is capable of detecting reactive oxygen species, a key neutrophil effector function. This combined platform will provide high-resolution, multidimensional profiling of neutrophil function from a drop of blood. We will develop an integrated chip-based platform and validate the system using chemically activated human neutrophils. This work will enable more precise immune monitoring in sepsis and other immune-related conditions, advancing clinical applications and providing substantial commercial potential.

MIT PI:
Joel Voldman, Professor, Department of Electrical Engineering and Computer Science

PT PI:
Pedro Alpuim, Research Group Leader, International Iberian Nanotechnology Laboratory
Jérôme Borme, Research Scientist PI, International Iberian Nanotechnology Laboratory

Scientific Area: Chips/Nanotechnology

Abstract: The advent of 2D crystalline materials has enabled the fabrication of novel devices with unique capabilities different from conventional 3D materials. Of particular interest are devices where the ability to control the stacking order and rotational alignment between 2D materials enables the engineering of novel tunable phases of matter. This proposal aims to investigate a new generation of nanoelectronic sensors based on novel 2D synthetic ferroelectric devices and twisted bilayer graphene devices. These sensors are expected to provide advantages such as ultra-fast switching speed, long endurance, biocompatibility, and compatibility with existing complementary metal-oxide-semiconductor technology. This work will involve a close collaboration between MIT and Portuguese researchers at the International Iberian Nanotechnology Institute (INL) paving the way to advanced technology applications of interest to the health sciences and pharmaceutical industry in Portugal.

MIT PI:
Pablo Jarillo-Herrero, Professor, Department of Physics

PT PI:
Dr. Pedro Alpuim, International Iberian Nanotechnology Institute, Research Group Leader.
Dr. Joaquin Fernandez-Rossier, International Iberian Nanotechnology Institute, Research Group Leader

Scientific Area: Space

Abstract: This project advances a novel plasma-based reactor for carbon dioxide conversion (CO2) and oxygen production under Mars-relevant conditions, supporting in-situ generation of valuable resources on Mars. Building on a prior MIT Portugal seed grant, we propose to design, construct, and test a plasma reactor integrated with oxygenpermeable membranes to selectively extract oxygen from the reaction products. The plasma is produced using Nanosecond Repetitively Pulsed (NRP) discharges, an energy-efficient technique that can reach up to 100% CO2 conversion. The integration of separation-technology addresses one of the main bottlenecks of the technology: the inability of plasmas to produce a pure stream of products. The system will be characterized across key parameters, including pressure, plasma power, and feedstock composition with a focus on measuring reactorperformance metrics: conversion fraction and energy efficiency. The work aims to elevate the technology from NASA Technology Readiness Level (TRL) 3 to TRL 4 by demonstrating a functional laboratory prototype. In addition to enabling sustainable oxygen and fuel production on Mars, the insights gained will inform scalable CO2 conversion systems for Earth-based decarbonization.

MIT PI:
Carmen Guerra-Garcia, Associate Professor, Department of Aeronautics and Astronautics

PT PI:
Vasco Guerra, Professor of Physics, Instituto Superior Técnico (IST), Universidade de Lisboa
Tiago Silva, PhD, Assistant Researcher, Instituto Superior Técnico (IST), Universidade de Lisboa

Scientific Area: Space

Abstract: Earth’s oceans are our planet’s largest defining feature and an invaluable resource as they contain millions of algae and animals responsible for sustaining large ecosystems that supply oxygen and food. Monitoring key features of the ocean provides critical insight to how its behavior is changing and how that affects maritime industries and ecosystems. This seed proposal builds upon prior work with MIT Portugal’s AEROS to take it to the next generation of development. We propose to use MIT’s available NASA CSLI launch opportunity (a $300k value) and existing CubeSat hardware (see Figure to engage with our MIT Portugal collaborators in taking the next-generation step for remote sensing with CubeSats: testing an AI-optimized processor in space and running algorithms relevant to maritime industry, such as looking for ocean fronts, coastal morphology and color change, and harmful algal blooms.

MIT PI:
Kerri Cahoy, Professor, Department of Aeronautics and Astronautics

PT PI:
Catarina Cecilio, Researcher +ATLANTIC CoLAB
Dr. Jorge Fontes, Researcher at the University of the Azores, Okeanos

Scientific Area: Sustainable Cities

Abstract: Sustainable cities face persistent social-environmental externalities that cannot be resolved by urban strategies alone. The forestry sector offers strong potential to mitigate these impacts and generate benefits for urban sustainability. This project focuses on using AI to enhance the efficiency and resilience of forestry supply chains. While AI offers environmental and economic advantages, sociotechnical factors like adoption, trust, and systemic resilience are often overlooked. Building on our sociotechnical AI maturity model, we aim to refine it using evidence-based factors to support responsible AI integration. The rapid uptake of AI across supply chains creates opportunities for innovation and sustainability, yet adoption remains uneven. Existing AI maturity models rarely consider sociotechnical systems thinking or responsible AI. By applying our model to Portugal’s forestry sector, a critical player in achieving Sustainable Smart Cities (SSCs) goals, we will assess current AI maturity and propose sociotechnical interventions, including digital twin concepts for SSCs ecosystems.

MIT PI:
Donna Rhodes, Principal Research Scientist, Sociotechnical Systems Research Center (SSRC)

PT PI:
António Lucas Soares, Researcher at INESC TEC and Associate Professor at Department of Informatics Engineering of the Faculty of Engineering of the University of Porto
André M. Carvalho, Assistant Professor, Department of Mechanical and Industrial Engineering, NOVA SCHOOL OF SCIENCE AND TECHNOLOGY | NOVA FCT, Universidade NOVA de Lisboa

Scientific Area: Digital Transformation in Manufacturing Sustainable Cities

Abstract: Rapid urbanization and extreme climate conditions are driving innovation in design and construction. At the global scale, the square footage of livable floor area must be doubled by 2060, a daunting goal made more challenging by the significant carbon impact of new construction and building operation [1]. Research on Low Carbon Large Scale Additive Manufacturing (LC-LSAM) offers a potential pathway to simultaneously accelerate and decarbonize construction. Portugal’s vernacular earth construction techniques offer time-tested climate adaptation strategies for thermal comfort in buildings, yet their integration with modern performance requirements requires new methods and experimental validation. This proposal combines MIT’s Digital Structures research (led by postdoctoral associate Dr. Alexander Curth) on LC-LSAM, including material-aware computational design, multi-objective toolpath optimization, and zero-waste earth printing, with FEUP’s expertise in Portuguese earthen construction, and low carbon printing admixtures to develop next-generation climate-resilient construction systems [2]. This work addresses a key barrier to scalable construction automation: making local materials a functional feedstock for the additive manufacturing of buildings. Current state of the art printing systems rely on carbon and cost intensive mortars with limited thermal performance. This collaboration will generate and test novel, architectural scale, climate and material adaptive computational design methods for the specific context of Portugal’s urban development needs, leveraging historic passive cooling strategies and locally sourced soils in a reproducible framework for low-carbon additive construction. This research will culminate in full-scale prototypes in Porto to test cooling loads compared to conventional construction and the relative carbon Life Cycle impacts of the 3D printed system. This work establishes new paradigms for performance-based vernacular architecture through computational design, specifically focused on the contemporary needs of a rapidly changing Portuguese urban development.

MIT PI:
Caitlin Mueller, Associate Professor, Department of Architecture

PT PI:
Bárbara Rangel, Assistant Professor, University of Porto, Faculty of Engineering (FUEP), Department of Civil Engineering, DIGI@feup3DC research group

Scientific Area: AI

Abstract: Bispecific antibody drugs provide transformative cures, but they are often refractory to modern bioproduction methods. Here, we will apply large-scale data and AI to integrate manufacturing with early bispecific antibody discovery. We will collect high-throughput manufacturability data and train an AI-based drug design algorithm to enhance drug activity, potency, and product quality. First, we will clone large libraries of bispecific antibody variants into manufacture-ready cell lines and growth conditions. Next, each bispecific antibody-producing cell will be captured emulsion microreactor droplets to analyze functional performance in manufacturing-like conditions. High-throughput sequencing will be used to analyze test results en masse, and AI models will be trained to identify critical features of manufacturable molecules. Finally, we will apply trained AI algorithms to generate new manufacturing-ready bispecific designs and evaluate their improvement related to controls. If successful, we will establish a seamless transition between early discovery and large-scale manufacturing for bispecific antibody drug products.

MIT PI:
Brandon DeKosky, Associate Professor, Department of Chemical Engineering

PT PI:
Paula Alves, CEO of iBET, Professor at NOVA University of Lisbon, Portugal
Antonio Roldao, Head of Cell-based Vaccines Development Lab, Coordinator of Late-stage R&D and Bioproduction Unit, iBET, Portugal
Jose Escandell, Principal Scientist, Animal Cell Technology Unit iBET, Portugal
Patrícia Alves, Coordinator of Analytical Services Unit, iBET, Portugal

Scientific Area: AI

Abstract: This project advances the frontiers of eXplainable Artificial Intelligence (XAI). Modern high-performance prediction models (e.g., random forests, gradient boosting, neural networks, LLMs) are complex black-box objects that are difficult to interpret and use in critical applications. One promising XAI approach is rule-based distillation where a large ensemble of trees is replaced by a small number of rules with model utility almost comparable to the original complex model. Despite their promise, the current scope of these approaches is limited. Our goal is to advance the frontiers of rule-based distillation. Our proposed framework (i) significantly generalizes the notion of rules beyond their traditional usage; (ii) allows the distilled rule ensemble to incorporate various user-defined notions of interpretability, trust, transparency, etc while maximally retaining model utility. We propose integrating these interpretable compact models into performant LLMs for improved in-context learning.

MIT PI:
Rahul Mazumder, Associate Professor, Sloan School of Management

PT PI:
Paulo Cortez, Full Professor, Dep. of Information Systems (DSI), U. Minho.