Atomic-scale 3D structural dynamics and functional degradation of Pt alloy nanocatalysts during the oxygen reduction reaction

Supplementary Data Codes The data and source codes for the paper [C. Jeong, J. Lee, H. Jo, KH. Lee, SJ. Lee, C. Ophus, P. Ercius, EA. Cho, and Y. Yang, "Atomic-scale 3D structural dynamics and functional degradation of Pt alloy nanocatalysts during the oxygen reduction reaction", Nat. Commun. 16, 8026 (2025).] are posted below. Overview Pt-based electrocatalysts are the primary choice for fuel cells due to their superior oxygen reduction reaction (ORR) activity. To enhance ORR performance and durability, extensive studies have investigated transition metal alloying, doping, and shape control to optimize the three key governing factors for ORR: geometry, local chemistry, and strain of their surface and subsurface. However, systematic optimization remains incomplete, as it requires an atomic-scale understanding of these factors and their dynamics over potential cycling, as well as their relationship to ORR activity. Here, we implement neural network-assisted atomic electron tomography to measure the 3D atomic structural dynamics and their effects on the functional degradation of PtNi alloy catalysts. Our results reveal that PtNi catalysts undergo shape changes, surface alloying, and strain relaxation during cycling, which can be effectively mitigated by Ga doping. By combining geometry, local chemistry, and strain analysis, we calculated the changes in ORR activity over thousands of cycles and observed that Ga doping leads to higher initial activity and greater stability. These findings offer a pathway to understanding 3D atomic structural dynamics and their relation to ORR activity during cycling, paving the way for the systematic design of durable, high-efficiency nanocatalysts. System Requirements 1. All software dependencies and operating systems (including version numbers) Software: MATLAB R2021a – academic use Operating systems: CentOS Linux 7 2. Versions the software has been tested on MATLAB R2021a – academic use 3. Any required non-standard hardware There are no non-standard hardware Repositary Contents 1. Raw Experiment Data Folder: 1_Raw_data This folder contains raw experimental projections before denoising as well as their corresponding angles for 8 PtNi nanoparticles (2 pristine and 6 after 12k cycles) and 9 Ga-PtNi nanoparticles (3 pristine, 2 after 4k cycles, 2 after 8k cycles, and 2 after 12k cycles). For PtNi: pristine-p1, pristine-p2, 12k-p1, 12k-p2, 12k-p3, 12k-p4, 12k-p5, 12k-p6 For GaPtNi: pristine-p1, pristine-p2, pristine-p3, 4k-p1, 4k-p2, 8k-p1, 8k-p2, 12k-p1, 12k-p2 File name : ImageData.mat: raw experimental projections : Angles.mat: corresponding angles 2. Experiment Data Folder: 2_Measured_data This folder contains experimental projections after denoising and alignment as well as their corresponding angles for 8 PtNi nanoparticles (2 pristine and 6 after 12k cycles) and 9 Ga-PtNi nanoparticles (3 pristine, 2 after 4k cycles, 2 after 8k cycles, and 2 after 12k cycles). File name : Projs.mat: experimental projections after denoising and alignment : ZeroProjs.mat: zero-degree projections at the beginning, in the middle, and at the end of the experiment : Angles.mat: corresponding angles 3. Reconstructed 3D Volume before neural network-based volume data augmentation Folder: 3_Raw_reconstruction This folder contains the experimental tomograms, reconstructed using the GENFIRE algorithm, before neural network-based volume data augmentation. The original GENFIRE URL is below. (https://www.physics.ucla.edu/research/imaging/dataSoftware.html) File name : GF_bfDL.mat: Reconstructed 3D volume before neural network-based volume data augmentation : species_classify_input_new.mat: 3D atomic model and chemical species before neural network-based volume data augmentation 4. Neural Network-Based Volume Data Augmentation Folder: 4_DL_augmentation Pre-processing (matlab) (Normalize experimental tomograms used for neural network training) run ./Preprocessing_for_DL_aug/cal_mean_peak_PtNi.m For running neural network-based volume data augmentation 1. Please prepare the input data with a size of 384 × 384 × 384 voxels. 2. Choose 'run_ML_NiPt.ipynb' for jupyter notebook. File name ./Inputdata/Particle #/input_normalized.mat: normalized experimental tomograms used for neural network training ./Outputdata/Particle #/ML_output_fix5_e30_1.3.mat: 3D tomograms after neural network-based volume data augmentation 5. Final Atomic Structure Folder: 5_Final_atomic_structures This folder contains the final 3D atomic model and chemical species (i.e. type 1 for Ni atom, type 2 for Pt atom) of the PtNi or GaPtNi nanoparticles after neural network-based volume data augmentation. File name : Rotated_Vol.mat: Reoriented volume along crystallographic direction : Rotated_matrix.mat: Rotation matrix for rotating along crystallographic direction : Final_atomic_model.mat: Final 3D atomic model and chemical species before reorientation along crystallographic direction 6. Data analysis Folder : 6_Data_analysis This folder contains the analysis of geometry, local chemistry, strain, and the ORR. ./Main1_Geometrical_dynamics: Geometric analysis run 'Step1_Define_surface_atoms.m' to obtain surface atoms run 'Step2_Sphericity.m' to calculate sphericity of each nanoparticle run 'Step3_Assign_facet.m' to assign each surface atom to a crystallographic facet group ({100}, {110}, or {111}). ./Main2_Local_chemistry_dynamics: Local chemistry analysis run 'Step1_Obtain_Nifraction.m' to obtain Ni fraction of PtNi and Ga-PtNi nanoparticles during potential cycles. run 'Step2_Obtain_CSRO.m' to calculate Chemical short-range order parameter (CSRO) as a function of distance from the surface during potential cycles. ./Main3_Strain_dynamics: Strain analysis run 'Step1_Global_FCC_fitting.m' for global fcc fitting. run 'Step2_Local_FCC_fitting.m' for local fcc fitting. run 'Step3_Plot_strain_dynamics.m' to calculate Local volumetric strain of surface Pt atoms during potential cycles, and strain of Pt as a function of distance from the surface during potential cycles. ./Main4_ORR_calculations: ORR analysis run 'Step1_ORR.m' to calculate ORR activity during potential cycles based on modified alloy-sensitive generalized coordination number (MASGCN). Additional Notes If you use any of the above data or source codes in your publications and/or presentations, our paper should be properly cited: C. Jeong, J. Lee, H. Jo, KH. Lee, SJ. Lee, C. Ophus, P. Ercius, EA. Cho, and Y. Yang, "Atomic-scale 3D structural dynamics and functional degradation of Pt alloy nanocatalysts during the oxygen reduction reaction", Nat. Commun. 16, 8026 (2025). If you have any questions regarding the above data or source codes, please contact Yongsoo Yang, Department of Physics, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 34141, Korea. Email: [e-mail] The code is distributed under the terms of the GNU General Public License v3.0 (GPL-3.0).