Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

  1. Home
  3. Research Domains
  5. Physical Sciences
  7. Atomic, Molecular And Optical Physics
  9. Nonlinear Optics And Spectroscopy
  11. Prediction And Screening Of Lead-free Double Perovskite Photovoltaic Materials Based On Machine Learning

Prediction and Screening of Lead-Free Double Perovskite Photovoltaic Materials Based on Machine Learning

Juan Wang1, Yizhe Wang1, Xiaoqin Liu1

  • 1Xi'an Key Laboratory of Advanced Photo-Electronics Materials and Energy Conversion Device, School of Electronic Information, Xijing University, Xi'an 710123, China.

Molecules (Basel, Switzerland)
|June 13, 2025

Related Experiment Videos

Facile Synthesis of Colloidal Lead Halide Perovskite Nanoplatelets via Ligand-Assisted Reprecipitation
04:14

Facile Synthesis of Colloidal Lead Halide Perovskite Nanoplatelets via Ligand-Assisted Reprecipitation

Published on: October 1, 2019

12.9K
Influence of Hybrid Perovskite Fabrication Methods on Film Formation, Electronic Structure, and Solar Cell Performance
11:38

Influence of Hybrid Perovskite Fabrication Methods on Film Formation, Electronic Structure, and Solar Cell Performance

Published on: February 27, 2017

18.4K
Monovalent Cation Doping of CH3NH3PbI3 for Efficient Perovskite Solar Cells
08:30

Monovalent Cation Doping of CH3NH3PbI3 for Efficient Perovskite Solar Cells

Published on: March 19, 2017

16.6K

View abstract on PubMed

Summary
This summary is machine-generated.

Machine learning identified 99 lead-free double perovskites with ideal bandgaps for efficient solar cells. This research guides the design of stable, non-toxic photovoltaic materials for sustainable energy solutions.

Area of Science:

  • Materials Science
  • Computational Chemistry
  • Renewable Energy

Background:

  • The development of efficient and environmentally friendly energy solutions necessitates stable, lead-free perovskite materials.
  • Traditional photovoltaic materials often contain toxic elements like lead, posing environmental concerns.

Purpose of the Study:

  • To employ machine learning for predicting the bandgap and formation energy of double perovskites.
  • To identify promising lead-free double perovskite candidates for photovoltaic applications.

Main Methods:

  • A dataset of 1053 double perovskites was curated from the Materials Project database.
  • Feature selection using Pearson correlation and mRMR identified key descriptors for prediction models.
  • XGBoost machine learning model was trained and validated for predicting bandgap and formation energy, achieving high accuracy (R² > 0.93).
Keywords:
XGBoost modelbandgap predictiondouble perovskitelead free and non-toxic

Related Experiment Videos

Facile Synthesis of Colloidal Lead Halide Perovskite Nanoplatelets via Ligand-Assisted Reprecipitation
04:14

Facile Synthesis of Colloidal Lead Halide Perovskite Nanoplatelets via Ligand-Assisted Reprecipitation

Published on: October 1, 2019

12.9K
Influence of Hybrid Perovskite Fabrication Methods on Film Formation, Electronic Structure, and Solar Cell Performance
11:38

Influence of Hybrid Perovskite Fabrication Methods on Film Formation, Electronic Structure, and Solar Cell Performance

Published on: February 27, 2017

18.4K
Monovalent Cation Doping of CH3NH3PbI3 for Efficient Perovskite Solar Cells
08:30

Monovalent Cation Doping of CH3NH3PbI3 for Efficient Perovskite Solar Cells

Published on: March 19, 2017

16.6K
  • SHAP analysis was used to interpret the model's predictions and identify influential features.

    • Main Results:

    • XGBoost demonstrated superior performance in predicting bandgap (R² = 0.934) and formation energy (R² = 0.959).
    • Key factors influencing bandgap include X-site electron affinity and B″-site ionization energies.
    • Formation energy is primarily determined by X-site ionization energy and B'/B″ site electronegativities.
    • Generated and screened 4573 double perovskites, selecting 2054 structurally stable candidates.
    • Identified 99 lead-free double perovskites with optimal bandgaps (1.3–1.4 eV) for photovoltaics.
    • Four known compounds and 95 novel perovskite compositions were identified as promising candidates.

    Conclusions:

    • The study successfully identified 99 lead-free double perovskites with suitable bandgaps for photovoltaic applications.
    • Specific elemental compositions, particularly involving X-site elements (Se, S, O, C) and B″-site elements (Pd, Ir, Fe, Ta, Pt, Cu), favor narrow bandgaps.
    • The findings provide a valuable roadmap for designing next-generation, non-toxic, high-performance photovoltaic materials.
    material screening