Accepted as a speech
in session Academy of Finland session: Novel Applications of AI in Physical Sciences and Engineering research
in Hall 6 on March 6th, 16:30

# STRUCTURE PREDICTION OF HYBRID NANOPARTICLES VIA ARTIFICIAL INTELLIGENCE

### Antti Pihlajamäki1, Joakim Linja2, Joonas Hämäläinen2, Paavo Nieminen2, Sami Malola1, Tommi Kärkkäinen2, Hannu Häkkinen1,3

antti.e.pihlajamaki@student.jyu.fi
1 University of Jyväskylä, Department of Physics, Nanoscience Center
2 University of Jyväskylä, Faculty of Information Technology
3 University of Jyväskylä, Department of Chemistry, Nanoscience Center

In the fields of imaging, catalysis and medicine hybrid nanoparticles or
monolayer protected clusters (MPCs) have shown great results and promises
[1, 2, 3]. The structure of these particles consist of three parts: metallic
core, metal-ligand interface and protecting ligands [4]. In order to truly
understand the properties and behaviour of them, both experiments and
theoretical computations are needed. The usual choice of computational
method is density functional theory (DFT), which is known to be accurate
but also demanding. Our goal is utilize artificial intelligence and machine
learning methods to predict and model the properties of MPCs. Ultimately
these methods should reduce the computational burden.

Currently we are developing methods to predict the positions of sulfur
atoms in the metal-ligand interface of the gold clusters [5]. The positions of
the sulfur atoms are determined by the local chemical environments. This
chemical information is extracted from the known structures and in the cur-
rent algorithm it is used to define the rules of sulfur atom positions. The
algorithm uses these rules to predict where are the most probable positions
of sulfur atoms for the structure of interest. The versions under development
are relaying on machine learning and heuristic algorithms.

References
[1] Y. Zhu, H. Qian, M. Zhu, and R. Jin, Thiolate-protected $\text{Au}_n$ nanoclus-
ters as catalysts for selective oxidation and hydrogenation processes, Adv.