Knowledge-driven fine-tuning of perovskite-based electrode materials for reversible Chemicals-to-Power devices

Informazioni relative al progetto

KNOWSKITE-X

ID dell’accordo di sovvenzione: 101091534

DOI

10.3030/101091534

Data della firma CE 29 Novembre 2022

Data di avvio 1 Gennaio 2023

Data di completamento 31 Dicembre 2026

Finanziato da

• Digital, Industry and Space

Costo totale

€ 5 168 000,00

Contributo UE

€ 5 168 000,00

Coordinato da

CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRS
France

Periodic Reporting for period 2 - KNOWSKITE-X (Knowledge-driven fine-tuning of perovskite-based electrode materials for reversible Chemicals-to-Power devices)

Periodo di rendicontazione: 2024-07-01 al 2025-12-31

Sintesi del contesto e degli obiettivi generali del progetto

The transition towards low-carbon energy systems requires advanced materials capable of operating efficiently and durably under realistic and reversible conditions. Solid oxide fuel cell and electrolysis cell (SOFC/SOEC) technologies are key enablers for hydrogen production, energy conversion and electricity storage, but their deployment remains limited by materials complexity, lack of harmonised characterisation methodologies and reliance on empirical development approaches.

KNOWSKITE-X addresses these challenges by establishing a science-based methodology for the rational design of perovskite-based electrode materials, integrating advanced characterisation, multi-scale modelling and data-driven approaches. The project focuses on developing transferable methodologies to understand and optimise composition–structure–property relationships under use-relevant conditions, rather than optimising a single material.

The objectives of KNOWSKITE-X are to generate new scientific knowledge on structure–property–performance correlations, to develop and harmonise advanced characterisation and modelling methods supported by FAIR data frameworks, and to demonstrate the applicability of these methodologies for materials optimisation at laboratory and pre-industrial scales.

Overall, KNOWSKITE-X contributes to the advancement of low-carbon energy technologies while supporting European priorities on open science, data interoperability, reduced critical raw material dependency and sustainable innovation.

Lavoro eseguito dall’inizio del progetto fino alla fine del periodo coperto dalla relazione e principali risultati finora ottenuti

During the reporting period, KNOWSKITE-X consolidated a coherent methodological framework combining multi-scale modelling, advanced characterisation and data-driven approaches for the rational design of perovskite-based materials relevant to reversible SOFC/SOEC applications.

On the modelling side, activities progressed beyond initial atomic-scale and kinetic models. Density Functional Theory (DFT) was systematically used to generate physically meaningful descriptors related to surface energetics, redox behaviour and defect chemistry of representative perovskite compositions. These descriptors were coupled with kinetic models and integrated into surrogate and machine-learning workflows, with targeted DFT calculations addressing critical data gaps. This resulted in a validated hybrid DFT–ML approach enabling accelerated materials screening within defined applicability domains.

Experimentally, the consortium synthesised and characterised benchmark, model and candidate materials with reduced critical raw material content based on abundant elements such as Fe, Ca or Ti. Scalable synthesis routes and textural control strategies were explored to tailor surface and microstructural properties relevant to electrochemical performance.

Harmonised routine and advanced characterisation techniques, including structural, redox and operando analyses, generated robust and comparable datasets linking preparation routes to functional response. More than one hundred material formulations were investigated, supporting cross-comparison, model validation and methodology assessment.

Overall, KNOWSKITE-X progressed from proof-of-concept towards a transferable methodology for science-based materials design, supported by validated modelling tools, harmonised experimental protocols and FAIR-compliant data workflows.ratory and pre-industrial validation through representative case studies guiding materials optimisation prior to device-level testing.

WP6 – Dissemination, communication and standardisation
WP6 ensured dissemination of results through conferences, workshops and training activities, and engagement with European initiatives to support uptake of harmonised methodologies beyond the project.

Progressi oltre lo stato dell’arte e potenziale impatto previsto (incluso l’impatto socioeconomico e le implicazioni sociali più ampie del progetto fino ad ora)

KNOWSKITE-X has delivered methodological and scientific advances beyond the current state of the art in the rational design of perovskite-based materials for energy applications.

On the modelling side, the project established a hybrid DFT–machine learning framework that fundamentally differs from purely data-driven or purely first-principles approaches. Density Functional Theory calculations explicitly account for realistic compositional disorder and non-symmetrical substitutions in complex perovskite systems, such as LSCF-derived materials, which are often idealised in the literature. This enables the computation of descriptors and activation energies that are more representative of real electrode materials and operating conditions, improving physical relevance and transferability.

KNOWSKITE-X further demonstrated how DFT-derived descriptors can be systematically integrated into ML-assisted workflows to enable accelerated materials screening within controlled validity domains, addressing key limitations of current ML approaches such as data sparsity, limited interpretability and poor extrapolation.

Experimentally, the project moved beyond benchmark-driven optimisation by identifying new candidate materials with reduced critical raw material content and by developing preparation routes enabling improved control of texture and surface properties. These advances support rational optimisation strategies rather than empirical trial-and-error.

In terms of methodology and standardisation, KNOWSKITE-X operationalised CHADA and MODA workflows for characterisation and modelling, enabling traceable, interoperable and reusable datasets. These contributions represent a concrete step beyond fragmented data practices and support the emergence of future standards in materials characterisation and modelling.