Eigenschaften von gepfropften Polymeren unter veränderlichen Lösungsmittelbedingungen und ihre Wechselwirkung mit Nanoteilchen
Final Report Abstract
The major goals of the project were to develop powerful lattice algorithms to simulated polymer systems with explicit solvent models, and to apply these methods to study structure formation of polymer brushes under poor solvent conditions. The further acceleration of simulation models can be achieve by massive parallelization and translating the codes to graphic processing units (GPU). The bond fluctuation model (BFM), one of the most performant and well tested polymer simulation models, can be implemented on a base centered cubic lattice (bcc) where independent moves are realized for very large subsets of the monomers. This led to the development of a new variant of the BFM within this project which can be efficiently implemented to GPUs. The new algorithm has been tested and compared to the classical serial algorithm, and we have found that it realized the same universality class for polymers both with respect to static and to dynamic quantities. Usually lattice models are not developed to simulate hydrodynamic (HD) properties of polymers but can be used to simulated dynamical properties in the Langevin-limit such as for dense polymers or as to study theoretical models without HD interactions. Within this project we were able to implement the so-called particle interaction model for a fluid on the same (bcc) lattice which resembles HD interactions in average. Coupling this model to the polymer subsystem has been realized. Thus, HD simulations of polymers are possible using a highly performant lattice model which scale only to the third power of the system size. First results for single chains are in good agreement with pre-average HD models (Zimm-model) for polymer chains. In this project explicit solvent models have been used to simulate homopolymer chains under the influence of poor and non-solvents. Single chain properties display a strong collapse transition without freezing the chain conformations as this is the case for short-range attractive pair-interactions as commonly used in implicit solvent models. Using this approach the stretching transition of a single chain in poor solvent have been investigated in detail. Using the bcc-BFM polymer brushes in poor solvents have been simulated for a broad range of parameters. Long simulation runs reveal homogeneous surface properties above the stretching transition of the chains. At lower grafting densities lateral structure formation takes place. In particular so-called octopus micelles have been studied. Octopus micelles result from the competition between droplet formation, as a consequence of surface minimization, and stretching of the grafted chains in order to join the droplets. In contrast to earlier predictions based on mean-field models the micelles are not spherical but are rather flat cylinders on the substrate with an almost constant height. We have developed a scaling model to understand this effect, and which reveals the role of the overlap threshold as the limiting case for octopus micelle formation. Our simulation results are in good agreement with the scaling model and the predicted scaling variable can be used to describe simulation data for different chain lengths and grafting densities under poor solvent conditions. [10] The lateral structure formation is enhanced in charged systems which we have studied using a standard beadspring molecular dynamics model. Using various counterions models including attractive interactions with respect to the monomers we could show that the counterion entropy is the driving force for lateral structure formation. Using a virtual piston to compress the charged brush we were able to calculate the corresponding minimal work and thus the entropy of the system. The results correspond to the predictions from an ideal gas model for the counterions.
Publications
- Polymers under poor solvent conditions simulated using the Bond Fluctuation Model and new routes to accelerate dynamic lattice polymer simulations. ISMC 2013, Rom , 15.09.-19.09.2013
C. Jentzsch, M. Werner and J.-U. Sommer
- Single polymer chains in poor solvent: Using the bond fluctuation method with explicit solvent. Journal of Chemical Physics 138, 094902 (2013)
C. Jentzsch, M. Werner and J.-U. Sommer
- The bond fluctuation model on graphics processing units: applications and new perspectives. 2nd International Symposium “Computer Simulations on GPU”, Freudenstadt, 27.05.-30.05.13
M. Werner, C. Jentzsch, R. Dockhorn and J.-U. Sommer
- Unwinding globules under stretching force: A Monte Carlo Study. DPG Spring Meeting 2013, Regensburg, 10.03.-15.03.2013
C. Jentzsch, M. Werner and J.-U. Sommer
- A Highly Parallelizable Monte Carlo Method For Simulating Polymeric Systems. DPG Frühjahrstagung Dresden 2014, Dresden, 30.03. - 04.04.2014
C. Jentzsch, R. Dockhorn, M. Werner and J.-U. Sommer
- Molecular dynamics simulations of polyelectrolyte brushes under poor solvent conditions: Origins of bundle formation. Journal of Chemical Physics 140, 1044911 (2014)
G.-L. He, H. Merlitz and J.-U. Sommer
(See online at https://doi.org/10.1063/1.4867466) - Polymer brushes in explicit poor solvents studied using a new variant of the bond fluctuation model. Journal of Chemical Physics 141, 104908 (2014)
C. Jentzsch and J.-U. Sommer
(See online at https://doi.org/10.1063/1.4895555)
