Phase behavior of colloidal dimers and hydrodynamic instabilities in binary mixtures

K. Milinkovic

Research output: ThesisDoctoral thesis 1 (Research UU / Graduation UU)

Abstract

We use computer simulations to study colloidal suspensions comprised of either bidisperse spherical particles or monodisperse dimer particles. The two main simulation techniques employed are a hybrid between molecular dynamics and stochastic rotation dynamics (MD-SRD), and a Monte Carlo (MC) algorithm. MD-SRD allows us to take Brownian motion and hydrodynamic interactions into account, while we use MC simulations to study equilibrium phase behavior. The first part of this thesis is dedicated to studying the Rayleigh-Taylor-like hydrodynamic instabilities which form in binary colloidal mixtures. Configurations with initially inhomogeneous distributions of colloidal species let to sediment in confinement will undergo the instability, and here we have studied the formation, evolution and the structural organization of the colloids within the instability as a function of the properties of the binary mixture. We found that the distribution of the colloids within the instability does not depend significantly on the composition of the mixtures, but does depend greatly on the relative magnitudes of the particle Peclet numbers. To follow the time evolution of the instability formation we calculated the spatial colloid velocity correlation functions, observing alternating regions in which the particle sedimentation velocities are correlated and anticorrelated. These observations are consistent with the network-like structures which are characteristic for Rayleigh-Taylor instabilities. We also calculated the growth rates of the unstable modes both from our simulation data and theoretically, finding good agreement between the obtained results. The second part of this thesis focuses on the phase behavior of monodisperse dimer systems. We first studied the phase behavior of hard snowman-shaped particles which consist of tangential hard spheres with different diameters. We used Monte Carlo simulations and free energy calculations to obtain the phase diagram as a function of the sphere diameter ratio, predicting stable isotropic fluid, plastic crystal and aperiodic crystalline phases. The crystalline phases found to be stable for a given diameter ratio at high densities correspond to the close packed structures of equimolar binary hard-sphere mixtures with the same diameter ratio. However, we also predict several crystal-crystal phase transitions, such that the best packed structures are stable at higher densities, while those with a higher degree of degeneracy are stable at lower densities. To explore the effects of degeneracy entropy on the phase behavior of dimer particles, we calculated the phase diagram of hard asymmetric dumbbells. These particles consist of two spheres with fixed diameters and varying center-to-center separation. We predicted stable isotropic fluid, plastic crystal, and periodic NaCl-based and both periodic and aperiodic CrB-based crystalline phases, and found that reducing the sphere separation results in the aperiodic crystalline phases of snowman-shaped particles becoming destabilized. Finally, we have also studied the phase behavior of dumbbell particles interacting with hard-core repulsive Yukawa potentials. We found that dumbbells with sufficiently long-ranged interactions crystallize spontaneously into plastic crystals in which the particle centers of mass are located on average on a BCC crystal lattice. The auto- and spatial orientational correlation functions reveal no significant hindrance of the particle rotations even for the shortest ranged interactions studied.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Utrecht University
Supervisors/Advisors
  • Dijkstra, Marjolein, Primary supervisor
Award date13 May 2013
Publisher
Print ISBNs978-90-393-5944-0
Publication statusPublished - 13 May 2013

Fingerprint

Dive into the research topics of 'Phase behavior of colloidal dimers and hydrodynamic instabilities in binary mixtures'. Together they form a unique fingerprint.

Cite this