The Effect of Cooling Rate on Crystallinity of Au Aerosol Nanoparticles

YI WANG, Sotiris Pratsinis, ETH Zurich, Switzerland

     Abstract Number: 200
     Working Group: Nanoparticles and Materials Synthesis

Abstract
Gold nanoparticles find widespread applications in many technological areas, such as medicine, microelectronics, and sensing. Their chemo-physical properties strongly depend on their crystalline structures, making investigations into crystallinity crucial. Molecular dynamics simulations employing the embedded atom method are conducted [1] to investigate the formation of Au crystals by gas-phase (flame, plasma, laser, etc.) processes. Non-isothermal homogeneous crystallization simulations are performed at various cooling rates for Au nanoparticles of different sizes, and their crystallization evolution is examined. The crystalline extent of nanoparticles, predominantly face-centered cubic with some hexagonal close-packed structures near their surface, is quantified based on their local crystalline disorder. The appearance, growth and coalescence pathways of the crystals are tracked through cross-sectional trajectories. The steep decline of the average atomic potential energy and amorphous fraction marked the crystalline transition at the end of the metastable phase region, which occurs earlier for larger particles. Subsequently, the crystals continue growing as temperature decreases, resulting in less distorted atoms at the grain boundaries and outer surface with more recognizable crystal domains. The retained atoms fraction is used to quantify the transition from sub-critical to supercritical nuclei and thus the onset of nucleation. X-ray diffraction patterns are generated [2] from which the dynamics of crystal growth are elucidated and compared to direct tracing and mapping of crystal sizes, revealing that lower cooling rates and larger particle sizes produce larger crystals. The impact of cooling rates on larger particles is more significant in terms of the final crystallite sizes. The crystallization is prolonged and the final crystallite size is larger at lower cooling rates than at higher ones as the modified Avrami equation is employed for describing the crystallization kinetics.

[1] Goudeli, E. and Pratsinis, S. E. (2016) AIChE J. 62:589-98.
[2] Goudeli, E. and Pratsinis, S. E. (2017) ACS Nano. 11:11653-60.