Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine Learning

Organic electrode materials (OEMs), composed of abundant elements such as carbon, nitrogen, and oxygen, offer sustainable alternatives to conventional electrode materials that depend on finite metal resources. The vast structural diversity of organic compounds provides a virtually unlimited design s...

Full description

Saved in:
Bibliographic Details
Main Author: Jaehyun Park (2402962) (author)
Other Authors: Farshud Sorourifar (14553482) (author), Madhav R. Muthyala (20026184) (author), Abigail M. Houser (20026187) (author), Madison Tuttle (20026190) (author), Joel A. Paulson (1631845) (author), Shiyu Zhang (372275) (author)
Published: 2024
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1852025488005922816
author Jaehyun Park (2402962)
author2 Farshud Sorourifar (14553482)
Madhav R. Muthyala (20026184)
Abigail M. Houser (20026187)
Madison Tuttle (20026190)
Joel A. Paulson (1631845)
Shiyu Zhang (372275)
author2_role author
author
author
author
author
author
author_facet Jaehyun Park (2402962)
Farshud Sorourifar (14553482)
Madhav R. Muthyala (20026184)
Abigail M. Houser (20026187)
Madison Tuttle (20026190)
Joel A. Paulson (1631845)
Shiyu Zhang (372275)
author_role author
dc.creator.none.fl_str_mv Jaehyun Park (2402962)
Farshud Sorourifar (14553482)
Madhav R. Muthyala (20026184)
Abigail M. Houser (20026187)
Madison Tuttle (20026190)
Joel A. Paulson (1631845)
Shiyu Zhang (372275)
dc.date.none.fl_str_mv 2024-11-01T11:03:52Z
dc.identifier.none.fl_str_mv 10.1021/jacs.4c11663.s002
dc.relation.none.fl_str_mv https://figshare.com/articles/dataset/Zero-Shot_Discovery_of_High-Performance_Low-Cost_Organic_Battery_Materials_Using_Machine_Learning/27447206
dc.rights.none.fl_str_mv CC BY-NC 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Biochemistry
Medicine
Pharmacology
Science Policy
Infectious Diseases
Biological Sciences not elsewhere classified
Chemical Sciences not elsewhere classified
Information Systems not elsewhere classified
vast structural diversity
simultaneously balance reward
organic compounds provides
offer sustainable alternatives
human intuition alone
finite metal resources
conventional electrode materials
combines computational chemistry
000 organic compounds
performing oems among
oems ), composed
solubility ),
oems selected
discovered oems
among sparkle
synthesizability ).
six years
shot predictions
shot discovery
new framework
molecular generation
machine learning
fold improvement
extrapolation tasks
experimental testing
error approaches
edisonian trial
cycling stability
cell batteries
abundant elements
dc.title.none.fl_str_mv Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine Learning
dc.type.none.fl_str_mv Dataset
info:eu-repo/semantics/publishedVersion
dataset
description Organic electrode materials (OEMs), composed of abundant elements such as carbon, nitrogen, and oxygen, offer sustainable alternatives to conventional electrode materials that depend on finite metal resources. The vast structural diversity of organic compounds provides a virtually unlimited design space; however, exploring this space through Edisonian trial-and-error approaches is costly and time-consuming. In this work, we develop a new framework, SPARKLE, that combines computational chemistry, molecular generation, and machine learning to achieve zero-shot predictions of OEMs that simultaneously balance reward (specific energy), risk (solubility), and cost (synthesizability). We demonstrate that SPARKLE significantly outperforms alternative black-box machine learning algorithms on interpolation and extrapolation tasks. By deploying SPARKLE over a design space of more than 670,000 organic compounds, we identified ≈5000 novel OEM candidates. Twenty-seven of them were synthesized and fabricated into coin-cell batteries for experimental testing. Among SPARKLE-discovered OEMs, 62.9% exceeded benchmark performance metrics, representing a 3-fold improvement over OEMs selected by human intuition alone (20.8% based on six years of prior lab experience). The top-performing OEMs among the 27 candidates exhibit specific energy and cycling stability that surpass the state-of-the-art while being synthesizable at a fraction of the cost.
eu_rights_str_mv openAccess
id Manara_7b2dd5c5a30dbebcdd0fb1dd42e49ade
identifier_str_mv 10.1021/jacs.4c11663.s002
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/27447206
publishDate 2024
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY-NC 4.0
spelling Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine LearningJaehyun Park (2402962)Farshud Sorourifar (14553482)Madhav R. Muthyala (20026184)Abigail M. Houser (20026187)Madison Tuttle (20026190)Joel A. Paulson (1631845)Shiyu Zhang (372275)BiochemistryMedicinePharmacologyScience PolicyInfectious DiseasesBiological Sciences not elsewhere classifiedChemical Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedvast structural diversitysimultaneously balance rewardorganic compounds providesoffer sustainable alternativeshuman intuition alonefinite metal resourcesconventional electrode materialscombines computational chemistry000 organic compoundsperforming oems amongoems ), composedsolubility ),oems selecteddiscovered oemsamong sparklesynthesizability ).six yearsshot predictionsshot discoverynew frameworkmolecular generationmachine learningfold improvementextrapolation tasksexperimental testingerror approachesedisonian trialcycling stabilitycell batteriesabundant elementsOrganic electrode materials (OEMs), composed of abundant elements such as carbon, nitrogen, and oxygen, offer sustainable alternatives to conventional electrode materials that depend on finite metal resources. The vast structural diversity of organic compounds provides a virtually unlimited design space; however, exploring this space through Edisonian trial-and-error approaches is costly and time-consuming. In this work, we develop a new framework, SPARKLE, that combines computational chemistry, molecular generation, and machine learning to achieve zero-shot predictions of OEMs that simultaneously balance reward (specific energy), risk (solubility), and cost (synthesizability). We demonstrate that SPARKLE significantly outperforms alternative black-box machine learning algorithms on interpolation and extrapolation tasks. By deploying SPARKLE over a design space of more than 670,000 organic compounds, we identified ≈5000 novel OEM candidates. Twenty-seven of them were synthesized and fabricated into coin-cell batteries for experimental testing. Among SPARKLE-discovered OEMs, 62.9% exceeded benchmark performance metrics, representing a 3-fold improvement over OEMs selected by human intuition alone (20.8% based on six years of prior lab experience). The top-performing OEMs among the 27 candidates exhibit specific energy and cycling stability that surpass the state-of-the-art while being synthesizable at a fraction of the cost.2024-11-01T11:03:52ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.1021/jacs.4c11663.s002https://figshare.com/articles/dataset/Zero-Shot_Discovery_of_High-Performance_Low-Cost_Organic_Battery_Materials_Using_Machine_Learning/27447206CC BY-NC 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/274472062024-11-01T11:03:52Z
spellingShingle Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine Learning
Jaehyun Park (2402962)
Biochemistry
Medicine
Pharmacology
Science Policy
Infectious Diseases
Biological Sciences not elsewhere classified
Chemical Sciences not elsewhere classified
Information Systems not elsewhere classified
vast structural diversity
simultaneously balance reward
organic compounds provides
offer sustainable alternatives
human intuition alone
finite metal resources
conventional electrode materials
combines computational chemistry
000 organic compounds
performing oems among
oems ), composed
solubility ),
oems selected
discovered oems
among sparkle
synthesizability ).
six years
shot predictions
shot discovery
new framework
molecular generation
machine learning
fold improvement
extrapolation tasks
experimental testing
error approaches
edisonian trial
cycling stability
cell batteries
abundant elements
status_str publishedVersion
title Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine Learning
title_full Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine Learning
title_fullStr Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine Learning
title_full_unstemmed Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine Learning
title_short Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine Learning
title_sort Zero-Shot Discovery of High-Performance, Low-Cost Organic Battery Materials Using Machine Learning
topic Biochemistry
Medicine
Pharmacology
Science Policy
Infectious Diseases
Biological Sciences not elsewhere classified
Chemical Sciences not elsewhere classified
Information Systems not elsewhere classified
vast structural diversity
simultaneously balance reward
organic compounds provides
offer sustainable alternatives
human intuition alone
finite metal resources
conventional electrode materials
combines computational chemistry
000 organic compounds
performing oems among
oems ), composed
solubility ),
oems selected
discovered oems
among sparkle
synthesizability ).
six years
shot predictions
shot discovery
new framework
molecular generation
machine learning
fold improvement
extrapolation tasks
experimental testing
error approaches
edisonian trial
cycling stability
cell batteries
abundant elements