Search for: All records

We have developed an artificial intelligence tool, XES Neo, for fitting x-ray emission spectroscopy (XES) data using a genetic algorithm. The Neo package has been applied to extended x-ray absorption fine structure [Terry et al., Appl. Surf. Sci. 547, 149059 (2021)] as well as Nanoindentation data [Burleigh et al., Appl. Surf. Sci. 612, 155734 (2023)] and is in development for x-ray photoelectron spectroscopy data. This package has been expanded to the fitting of XES data by incorporating basic background removal methods (baseline and linear) optimized simultaneously with peak-fitting using the active background approach, as well as the peak shapes Voigt, and an asymmetrical Voigt, known as the Double Lorentzian. The fit parameters are optimized using a robust metaheuristic method, which starts with a population of temporary solutions known as the chromosomes. This population is then evaluated and assigned a fitness score, from which the best solution is then found. Future generations are created through crossover of the best sets of parameters along with some random parameters. Mutation is then done on the new generation using random perturbations to the chromosomal parameters. The population is then evaluated again, and the process continues. The analyzed data presented here are available in the corresponding XESOasis discussion forum (https://xesoasis.org/ai_posted). 

Material characterization techniques are widely used to characterize the physical and chemical properties of materials at the nanoscale and, thus, play central roles in material scientific discoveries. However, the large and complex datasets generated by these techniques often require significant human effort to interpret and extract meaningful physicochemical insights. Artificial intelligence (AI) techniques such as machine learning (ML) have the potential to improve the efficiency and accuracy of surface analysis by automating data analysis and interpretation. In this perspective paper, we review the current role of AI in surface analysis and discuss its future potential to accelerate discoveries in surface science, materials science, and interface science. We highlight several applications where AI has already been used to analyze surface analysis data, including the identification of crystal structures from XRD data, analysis of XPS spectra for surface composition, and the interpretation of TEM and SEM images for particle morphology and size. We also discuss the challenges and opportunities associated with the integration of AI into surface analysis workflows. These include the need for large and diverse datasets for training ML models, the importance of feature selection and representation, and the potential for ML to enable new insights and discoveries by identifying patterns and relationships in complex datasets. Most importantly, AI analyzed data must not just find the best mathematical description of the data, but it must find the most physical and chemically meaningful results. In addition, the need for reproducibility in scientific research has become increasingly important in recent years. The advancement of AI, including both conventional and the increasing popular deep learning, is showing promise in addressing those challenges by enabling the execution and verification of scientific progress. By training models on large experimental datasets and providing automated analysis and data interpretation, AI can help to ensure that scientific results are reproducible and reliable. Although integration of knowledge and AI models must be considered for the transparency and interpretability of models, the incorporation of AI into the data collection and processing workflow will significantly enhance the efficiency and accuracy of various surface analysis techniques and deepen our understanding at an accelerated pace. 

Understanding of structural and morphological evolution in nanomaterials is critical in tailoring their functionality for applications such as energy conversion and storage. Here, we examine irradiation effects on the morphology and structure of amorphous TiO2 nanotubes in comparison with their crystalline counterpart, anatase TiO2 nanotubes, using high-resolution transmission electron microscopy (TEM), in situ ion irradiation TEM, and molecular dynamics (MD) simulations. Anatase TiO2 nanotubes exhibit morphological and structural stability under irradiation due to their high concentration of grain boundaries and surfaces as defect sinks. On the other hand, amorphous TiO2 nanotubes undergo irradiation-induced crystallization, with some tubes remaining only partially crystallized. The partially crystalline tubes bend due to internal stresses associated with densification during crystallization as suggested by MD calculations. These results present a novel irradiation-based pathway for potentially tuning structure and morphology of energy storage materials. 

Nanoscale oxide-based negative electrodes are of great interest for lithium ion batteries due to their high energy density, power density and enhanced safety. In this work, we conducted a case study on mesoporous TiO 2 nanoparticle negative electrodes with uniform size and varying crystallinity in order to investigate the trend in the electrochemical properties of oxide-based nanoscale negative electrodes with varying crystallinity. Mesoporous solid spherical TiO 2 nanoparticles with a uniform particle size and varying crystallinity, i.e. , amorphous TiO 2 (A-TiO 2 ), partially crystalline TiO 2 (PC-TiO 2 ) and fully crystalline TiO 2 (FC-TiO 2 ) nanoparticles were studied. At low current rate (quasi steady-state), the specific capacity of the samples drops with the decrease of crystallinity. Ex situ synchrotron pair distribution function analysis reveals that the 1D zigzag Li ion diffusion pathway becomes expanded with the increase of crystallinity, which promotes ion mobility and charge storage. At high current rates (away from equilibrium states), however, the A-TiO 2 sample demonstrates slightly larger capacity than the FC-TiO 2 sample, both of which show larger capacities than that of the PC-TiO 2 sample. Both A-TiO 2 and FC-TiO 2 samples exhibit higher capacitive contribution to the charge storage and larger Li + diffusivity than those of the PC-TiO 2 sample, which explains their better rate capability. Moreover, the larger Li + diffusivity of the A-TiO 2 sample leads to the slightly larger specific capacity than the FC-TiO 2 sample at the highest current rate.