# Enhancing oxygen reduction reaction mass activity in polymer electrolyte fuel cells via molecular machine learning

https://mdr.nims.go.jp/datasets/55a97262-d049-4c9d-ba25-dd35fe09230c

## File

- [Gracheva_2026_Mach._Learn.__Sci._Technol._7_045006.pdf](https://mdr.nims.go.jp/filesets/345a8416-8bbb-4fbc-a76f-75daff30f77b/download) ([Detail](https://mdr.nims.go.jp/filesets/345a8416-8bbb-4fbc-a76f-75daff30f77b.md))

## Id

55a97262-d049-4c9d-ba25-dd35fe09230c

## Local identifier



## Visibility

open_to_public

## State

published

## Created at

2026-08-26T06:14:54.983402Z

## Updated at

2026-08-26T07:18:22.852839Z

## Published at

2026-08-26T09:30:08.402500Z

## Doi



## First published url

https://doi.org/10.1088/2632-2153/ae7f7c

## Date published

2026-08-01

## Recorded date published

2026-8-1

## Resource type

journal_article

## Manuscript type

vor

## Collection



## Title

- title: Enhancing oxygen reduction reaction mass activity in polymer electrolyte
    fuel cells via molecular machine learning
  title_type: original
  lang: en

## Description

- description: The search for efficient fuel cell additives is a critical challenge
    for improving their oxygen reduction reaction activity. Here, we present the first
    successful application of active learning (AL) to guide the experimental discovery
    of organic molecules that enhance the mass activity of Pt cathode catalysts in
    polymer electrolyte fuel cells (PEFCs), a class of fuel cells often referred to
    in the literature as proton-exchange membrane fuel cells (PEMFCs). Using our in-house
    SMILES-X molecular characterization tool, we trained neural networks, autonomously
    tailored for our data, on an initial set of 96 organic compounds, of which 84
    passed RDKit validity checks for modelling. Model performance improved over three
    AL cycles, with the mean absolute error decreasing from 0.202 to 0.165 and the
    root mean square error decreasing from 0.271 to 0.218. The cycles were guided
    by predictive ranking, experimental feasibility constraints, and LSTM-based molecule
    generation. Among the additives identified, melam, a melamine dimer, achieved
    a 67% improvement in mass activity over the no-additive baseline. These results
    demonstrate the feasibility of AL in low-data regimes for PEFC additive discovery
    and highlight the importance of data consistency, model retraining, and close
    ML–experiment coordination.
  description_type: abstract
  lang: und

## Creator

- name: Ekaterina Gracheva
  role: author
  orcid: https://orcid.org/0000-0002-9704-5939
- name: Shin-ichi Yamazaki
  role: author
- name: Guillaume Lambard
  role: author
  orcid: https://orcid.org/0000-0003-0275-4079
- name: Keitaro Sodeyama
  role: author
  orcid: https://orcid.org/0000-0002-9228-0729
- name: Tsutomu Ioroi
  role: author
- name: Masafumi Asahi
  role: author
  orcid: https://orcid.org/0000-0002-2122-3073

## Contact agent



## Publisher

organization: IOP Publishing

## Managing organization



## Keyword

- subject: materials science
  schema: not_defined
- subject: machine learning
  schema: not_defined
- subject: chemoinformatics
  schema: not_defined
- subject: polymer electrolyte fuel cell (PEFC)
  schema: not_defined
- subject: oxygen reduction reaction
  schema: not_defined
- subject: active learning
  schema: not_defined

## Rights

- identifier: https://creativecommons.org/licenses/by/4.0/

## Other identifier(s)



## Data origin

- data_origin_type: other

## Embargo



## Journal

- title: 'Machine Learning: Science and Technology'
  issn: '26322153'
  volume: '7'
  issue: '4'
  article_number: '045006'

## Conference



## Related item



## Funding

- funder_name: New Energy and Industrial Technology Development Organization

## Instrument



## Instrument operator



## Instrument managing organization



## Measurement method



## Specimen



## Chemical composition



## Structure for specimen



## Structural feature for specimen



## Specific property for specimen



## Process for specimen treatment



## Computational method



## Energy level/transition state



## Software



## Custom property



## Fileset

- id: 345a8416-8bbb-4fbc-a76f-75daff30f77b
  filename: Gracheva_2026_Mach._Learn.__Sci._Technol._7_045006.pdf
  content_type: application/pdf
  size: 1440145
  md5: 5931017c262487229433fe95479c89d4

## Thumbnail

fileset_id: 345a8416-8bbb-4fbc-a76f-75daff30f77b
filename: Gracheva_2026_Mach._Learn.__Sci._Technol._7_045006.pdf