60th Annual Report on Research 2015

I. Introduction

During the first grant year, we focused on the development of the basic methodology that enables accurate mesoscale simulations of asphaltene aggregation. The developments include (a) dissipative particle dynamics (DPD) simulations with beads of different sizes that are necessary for studies of the interactions between the primary asphaltene aggregates (b) coarse-grained methodology for large-scale simulations, where each primary asphaltene aggregate is modelled as a single particle.

II. Extending DPD methodology Quantitative models for dissipative particle dynamics (DPD) simulations of asphaltenes need to reproduce (a) surface density of carbon atoms in polyaromatic flat sheets (b) the distance between the sheets in a “stack” of flat polyaromatic molecules that form the cores of the primary asphaltene aggregates. This requires dissection of asphaltene molecules onto beads of different sizes (otherwise simulations become extremely expensive computationally). DPD simulations with differently sized beads are very rare in the literature and do not target specific complex systems; parameterization strategies for such systems are essentially non-existent. We have extended the technique of mesoscale models parameterization from activity coefficients of infinitely diluted reference solutions (developed previously by the PI [1]) to systems with beads of different diameters. The methodology was applied to asphaltenes and implemented into DL_MESO [2], one of the standard packages for mesoscale simulations. With the models obtained, we have performed DPD simulations of asphaltene aggregation in hexane and estimated the distribution of aggregate sizes. Currently, we are performing simulations of interactions of the primary aggregates for asphaltenes of different molecular structures and chemical compositions.

 III. Large-scale simulations of aggregation of primary particles

 The precipitation of asphaltenes from solution towards the bulk bitumen is driven by the attraction between the polyaromatic “cores” and resisted by the entropic repulsion between the aliphatic tails which are attached either to the asphaltene polyaromatic cores or to resin molecules. As primary aggregates associate (due to van der Walls forces or hydrogen bonds between hydrophilic groups attached to them) the entropic repulsion between them and non-associated primary aggregates strengthens due to the increases surface density of the aliphatic tails. This determines the dynamics of precipitation, which occurs via formation of fractal-type structures of varying dimensions.

To enable the mesoscale simulations of agglomeration of primary asphaltene aggregates, we have developed a simulation scheme, where each primary aggregate is modeled by one particle. The method is based on Brownian Dynamics (BD). The particles interact via soft harmonic repulsive potentials similar to those used in DPD and steep r^-12 potentials on very short distances. This form follows ref [3] where the forces between primary aggregates were experimentally measured. If two primary aggregates (due to thermal fluctuations) approach each other to distances closer than the quarter of the particle size, they can associate by forming reversible Morse bonds. The bonds are form and dissolve via canonical Monte Carlo random steps. The strength of the repulsive potentials between the primary particles changes with the number of bonds they have formed. To prove our concept, we implemented the scheme in DL_MESO and modelled the viscosities of systems of primary aggregates.

IV. Influence on other fields

The methodology for simulations of precipitations turned out to be applicable to very different class of systems: it allowed implementation of coordination into DPD. In this scheme, we present a transition metal ion as a semi-rigid cluster of coordinating centers whose arrangement follows the geometry of the actual complex (we implemented tetrahederal and octahederal geometries). The centers form dissociable bonds with the ligands, which are modelled as standard DPD beads. We tested this idea by calculating the viscosities of model poly(vinyl pyrrolidone) concentrated solutions, to which model cobalt chloride was added. Cobalt cations form reversible bonds with the polymer. No thorough parameterization of the models was performed for this system, but we qualitatively reproduced the influence of the coordinating cation on viscosity.

V. Impact on students involved in the project.

The project involves two Rutgers students pursuing Master degree in chemical engineering: Tianying Ma and Ravish Kumar. The PI teaches the mater students directly; they receive one-on-one instruction. Working on the project not only give them experience in modeling techniques. Most importantly, it improves their understanding of nanostructures systems formed by dissimilar fragments (typical examples of which are surfactants, block copolymers etc.) and different aspects of particulate systems. The skills and knowledge acquired by the students are of interest for various industries including those not related directly to crude oil. For example in personal care products, understanding of the structure and rheology of micellar and other segregating systems is indispensable. The PI has a positive experience of mentoring master students and expects that this tendency continue with T. Ma and R. Kumar who are expected to graduate in June 2016.

VI. Other developments

Starting this project allowed the PI to establish collaboration with the Rutgers Department of Civil and Environmental Engineering (Prof. Hao Wang). This research focuses on the structure and properties of asphalt, a field entirely different from the original topic of asphaltene aggregation. Of particular interest are the adsorption of asphaltenes on quartz that influences the properties of asphalt. Currently, the PI informally co-advices Guangji Xu PhD student of Prof. Wang. Guangji Xu also attends PI’s course “Fundamentals of Nanoscale Thermo- and Hydrodynamics”.

VII. References

1. Vishnyakov, A., M.-T. Lee, and A.V. Neimark, Prediction of the Critical Micelle Concentration of Nonionic Surfactants by Dissipative Particle Dynamics Simulations. Journal of Physical Chemistry Letters, 2013. 4(5): p. 797-802.

2. Seaton, M.A., et al., DL_MESO: highly scalable mesoscale simulations. Molecular Simulation, 2013. 39(10): p. 796-821.

3. Wang, S., et al., Colloidal Interactions between Asphaltene Surfaces in Toluene. Energy & Fuels, 2009. 23(1): p. 862-869.

(( word_count = 970 ))