Machine learning approach to atomic simulations of protected gold nanoclusters
Julkaistu sarjassa
JYU dissertationsTekijät
Päivämäärä
2022Tekijänoikeudet
© The Author & University of Jyväskylä
In the nanometer lengthscale, the boundaries between physics, chemistry and biology disappear and all phenomena are reduced to the level of atomic interactions. Technological advancement has provided means to measure what happens at the atomic level but there are limitations and experiments cannot tell everything. Here computational studies provide further insight. The most accurate computational methods are based on the quantum mechanics, which explains atomic interactions at the level of electrons. However, these methods are computationally demanding, which limits their usage. One can also compromise the accuracy and use efficient force field methods. During the last two decades a third type of method, machine learning (ML), has become increasingly popular. ML methods utilize data from other computational methods or measurements to "learn" underlying trends. This way they reproduce the behavior of the high-level methods with significantly reduced computational cost. Their usage is not restricted to imitate other methods but they can also be used for data analysis. In this thesis, four studies demonstrate three different applications of ML methods in studies of gold nanoclusters protected by organic ligands. Wavelet-based image comparison method was used to analyze experimental and theoretical transmission electron microscope (TEM) images of the crystal lattice made of nanoclusters. The analysis ruled out the possible structural isomer and shed light to the cluster orientation in TEM images. So-called distance-based ML methods were utilized for dynamic simulations of the similar clusters. Based on the given configuration the ML method predicted potential energies, which were used to run Monte Carlo simulations emulating the dynamics of the clusters. After this, a new distance-based ML method was designed to estimate forces affecting to the individual atoms of the nanoclusters. Estimated force vectors enabled ML assisted structure optimization of the goldthiolate systems. The results showed the great potential of the distance-based methods on simulations of the complex nanostructures.
...
Julkaisija
Jyväskylän yliopistoISBN
978-951-39-9309-2ISSN Hae Julkaisufoorumista
2489-9003Julkaisuun sisältyy osajulkaisuja
- Artikkeli I: Yao, Q., Liu, L., Malola, S., Xu, H., Wu, Z., Chen, T., Cao, Y., Matus, M. F., Pihlajamäki, A., Zang, S., Han, Y., Häkkinen, H. and Xie, J. (2022). Engineering Colloidal Crystals of Atomically Precise Gold Nanoparticles Promotedby Particle Surface Dynamics. Accepted in Nature Chemistry.
- Artikkeli II: Pihlajamäki, A., Hämäläinen, J., Linja, J., Nieminen, P., Malola, S., Kärkkäinen, T., & Häkkinen, H. (2020). Monte Carlo Simulations of Au38(SCH3)24 Nanocluster Using Distance-Based Machine Learning Methods. Journal of Physical Chemistry A, 124(23), 4827-4836. DOI: 10.1021/acs.jpca.0c01512. JYX: jyx.jyu.fi/handle/123456789/69062
- Artikkeli III: Pihlajamäki, A., Linja, J., Hämäläinen, J., Nieminen, P., Malola, S., Kärkkäinen, T., & Häkkinen, H. (2021). Orientation Adaptive Minimal Learning Machine for Directions of Atomic Forces. In ESANN 2021 : Proceedings of the 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning Online event (Bruges, Belgium), October 06 - 08 (pp. 529-534). ESANN. DOI: 10.14428/esann/2021.es2021-34
- Artikkeli IV: Pihlajamäki, A., Malola, S., Kärkkäinen,T. and Häkkinen. H. (2022). Orientation Adaptive Minimal Learning Machine: Application to Thiolate- Protected Gold Nanoclusters and Gold-Thiolate Rings. Arxiv
Metadata
Näytä kaikki kuvailutiedotKokoelmat
- JYU Dissertations [784]
- Väitöskirjat [3457]
Lisenssi
Samankaltainen aineisto
Näytetään aineistoja, joilla on samankaltainen nimeke tai asiasanat.
-
Photodynamics studies of ligand-protected gold nanoclusters by using ultrafast transient infrared spectroscopy
Mustalahti, Satu (University of Jyväskylä, 2015)Highly monodisperse samples of three ligand-protected gold nanoclusters Au102(pMBA)44, Au144(SC2H4Ph)60, and a cluster tentatively identified as Au130(pMBA)50, were characterized by UV/vis and infrared spectroscopy, ... -
Dynamic Stabilization of the Ligand-Metal Interface in Atomically Precise Gold Nanoclusters Au68 and Au144 Protected by meta-Mercaptobenzoic Acid
Tero, Tiia-Riikka; Malola, Sami; Koncz, Benedek; Pohjolainen, Emmi; Lautala, Saara; Mustalahti, Satu; Permi, Perttu; Groenhof, Gerrit; Pettersson, Mika; Häkkinen, Hannu (American Chemical Society, 2017)Ligand-stabilized, atomically precise gold nanoclusters with a metal core of a uniform size of just 1-3 nm constitute an interesting class of nanomaterials with versatile possibilities for applications due to their ... -
Computational Criteria for Hydrogen Evolution Activity on Ligand-Protected Au25-Based Nanoclusters
López-Estrada, Omar; Mammen, Nisha; Laverdure, Laura; Melander, Marko M.; Häkkinen, Hannu; Honkala, Karoliina (American Chemical Society (ACS), 2023)The hydrogen evolution reaction (HER) is a critical reaction in addressing climate change; however, it requires catalysts to be generated on an industrial scale. Nanomaterials offer several advantages over conventional HER ... -
What Contributes to the Measured Chiral Optical Response of the Glutathione-Protected Au25 Nanocluster?
Monti, Marta; Matus, María Francisca; Malola, Sami; Fortunelli, Alessandro; Aschi, Massimiliano; Stener, Mauro; Häkkinen, Hannu (American Chemical Society (ACS), 2023)The water-soluble glutathione-protected [Au25(GSH)18]−1 nanocluster was investigated by integrating several methodologies such as molecular dynamics simulations, essential dynamics analysis, and state-of-the-art time-dependent ... -
Dynamics of weak interactions in the ligand layer of meta-mercaptobenzoic acid protected gold nanoclusters Au68(m-MBA)32 and Au144(m-MBA)40
Mammen, Nisha; Malola, Sami; Honkala, Karoliina; Häkkinen, Hannu (Royal Society of Chemistry (RSC), 2020)Atomically precise metal nanoclusters, stabilized and functionalized by organic ligands, are emerging nanomaterials with potential applications in plasmonics, nano-electronics, bio-imaging, nanocatalysis, and as therapeutic ...
Ellei toisin mainittu, julkisesti saatavilla olevia JYX-metatietoja (poislukien tiivistelmät) saa vapaasti uudelleenkäyttää CC0-lisenssillä.