Investigation of the Effect of Sulfur Heteroatom on Asphaltene

Subscriber access provided by University of Sussex Library

Article

Investigation of the Effect of Sulfur Heteroatom on Asphaltene Aggregation Ana Carolina Rennó Sodero, Hugo Santos-Silva, Patricia Guevara Level, Brice Bouyssiere, Jean-Pierre Korb, Herve Carrier, Ahmad Alfarra, Didier Bégué, and Isabelle Baraille Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.6b00757 • Publication Date (Web): 09 May 2016 Downloaded from http://pubs.acs.org on May 14, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Investigation of the Effect of Sulfur Heteroatom on Asphaltene Aggregation Ana C. R. Sodero†‡*, Hugo Santos Silva‡, Patricia Guevara Level‡, Brice Bouyssiere‡, Jean-Pierre Korb§, Herve Carrier┴, Ahmad Alfarra#, Didier Bégué‡,and Isabelle Baraille‡ † Laboratório de Modelagem Molecular e QSAR (MODMOLQSAR), Departamento de Fármacos e Medicamentos, Universidade Federal do Rio de Janeiro (UFRJ), Av. Carlos Chagas Filho, 373, 21941-599, Brazil. ‡ Institut Pluridisciplinaire de Recherche sur l’Environnement et les Matériaux (IPREM), UMR 5254, Université de Pau et des Pays de l’Adour, Hélioparc, 2 Avenue du Président Angot, 64053 Pau Cedex 9, France. ┴ Laboratoire des Fluides Complexes, UMR 5150, Université de Pau et des Pays de l’Adour, BP 1155 64013 Pau Cedex, France § Physique de la Matière Condensée, Ecole Polytechnique−Centre National de la Recherche Scientifique (CNRS), 91128 Palaiseau, France. #Total Research&TechnologyGonfreville, BP 27, 76700 Harfleur, France. ABSTRACT: We present molecular dynamics simulations (MDS) for interpreting the molecular aggregation of four different asphaltene molecular models. These simulations are based on recent X-ray (SAXS) and neutron (SANS) small-angle scattering experiments from Eyssautier et al (2011) which proposed a discoidal asphaltene nanoaggregate structure of 0.67 nm height and its chemical composition. Basically, we have investigated the sulfur atom position in the asphaltene structure. The π-π is the main type of contact and it seems to be responsible to the parallel orientation of molecules. This geometry is also corroborated by the cosine of angles between the poly-aromatic cores. However, the polar side chain is also important to the aggregation, since it occurs to the systems during all the simulations. Three systems, namely systems 1T, 2T, and 4T are in good agreement with experimental work, indicating an average height lower than 0.8 nm. Concerning the place occupied by the sulfur atom in the molecule, we have found that when it is grafted directly to the conjugated core no effect can be deducted from the stability of dimers and trimers and all the other studied parameters. This is not the case when sulfur is located in the lateral chain, where it has a particular affinity with oxygen and hydrogen atoms of the acid chain ends of neighbor molecules. Consequently, we suggest that these configurations of the sulfur atom are more likely to be problematic in oil industry.

1 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

INTRODUCTION Asphatenes molecules are known as a fraction of native crude oil defined by their solubility class. They are insoluble in light n-alkanes like n-pentane or nheptane and soluble in aromatic solvents like benzene or toluene.1 The presence of asphaltenes is critical to many aspects of petroleum utilization, from oil recovery and transportation to refining. Complications related to asphaltene stability within the supporting oil matrix affect the entire production chain.2 The tendency of asphaltene aggregation causes several effects on oil viscosity, adsorption at solid surfaces, precipitation, fluid rheology, and emulsion stability.3 The structure of asphaltenes is complex and is normally assumed to be constituted by polycyclic aromatic hydrocarbons substituted with alkyl side chains and the presence of heteroatoms, including nitrogen, sulfur, oxygen, and metals.1, 4, 5 In spite of recent advances in asphaltenes characterization,3-5 their molecular structures remain poorly understood.6 Some analytical chemistry techniques enable the identification of many distinct chemical species in crude oil and asphaltenes. Recent X-ray (SAXS) and neutron (SANS) small-angle scattering studies in asphaltene solutions4 indicated asphaltenes as a discoidal nanoaggregates with a height of 0.67 nm. The authors also gave insights into the core and the side chain dimensions, as well as the chemical composition of asphaltenes. Carbon and hydrogen are the major atoms, but the molecules also contain heteroatoms, mainly sulfur. The exact formulation is unknown and highly dependent on oil sources. Thereby, we have chosen four asphaltene molecular models, as dimers and trimers in toluene, to investigate the sulfur atom position in the structure and the consequent effect on the stability of dimers and trimers in

Page 2 of 23

solution. These models are based on the C5Pe model module, a prototypical asphaltene molecule that has been reported and used in several studies.7, 8 Experimental work in progress using this molecule also justifies this choice that allows us to compare experience and theory. This work aims to investigate the influence of sulfur atoms on C5Pe’s dimers and trimers stabilization during a 150 ns time-delay in toluene at 298K and 1 atm conditions. In this way, all the other possible parameters influencing the nanoaggregation of asphaltenes (size of conjugated core, lateral chains, etc…) are excluded and one can really screen the effect of sulfur atoms in this physicalchemical process. COMPUTATIONAL METHODS Simulation Procedure The MD simulations (MDS) were carried out using the GROMACS 4.6.39-11 software package. The GROMOS96 force field12, 13 with the 53a6 parameter set was used in all calculations. The GROMOS96 force field was already tested to polyaromatic molecules and shown to be useful in the dynamics of poly-aromatic nanoaggregate formation.8, 14-16 The N-(1-Hexylheptyl)-N’-(5-carboxylicpentyl)-perylene-3,4,9,10tetracarboxylicbisimide molecule (C5Pe, molecule 1) and three C5Pe analogues (molecules 2, 3, and 4), with the presence of sulfur atom at different positions (core and side chain), were studied (Figure 1). The initial molecular geometries (PDB coordinates) of the asphaltene molecules were built using Avogadro software Version 1.1.1.17 The PDB coordinates were used as an input file to the PRODRG 2.5 49 server18 to generate the molecular topology and GROMACS structure files. Two other molecules were studied: the situation where all the C=O groups are replaced by C=S ones (the 2

ACS Paragon Plus Environment

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

extreme of molecule 2) and the situation where no heteroatom is found in this position (C-H bond, quinoid structure imposed), as one can see in Figure S1. The complete analysis of these molecules is found in the Electronic Supporting Information (ESI) material. Toluene was chosen as the solvent for this study, in a strategy to mimic the infinite dissolution of the asphaltene molecules. The charges and topology of the toluene molecule were generated from the phenylalanine amino acid fraction of GROMOS96 force field. A box of 3 nm3 was filled with 110 toluene molecules and the system energy was minimized using the steepest descent method, followed by the conjugated gradient one. To equilibrate the system, the simulation was carried out for 100 ps under the NVT ensemble at 298 K, using the Berendsen thermostat,19 followed by 100 ps under NPT ensemble at 298 K, using Berendsen thermostat and barostat. Finally, the system was simulated under NPT ensemble at 298 K and 1 atm, using Nosé-Hoover thermostat20, 21 and Parrinello-Rahman pressure coupling algorithm,22 for 20 ns. The average density was further calculated during the last 20ns from the production step,23 yielding a value of 0.88 g/cm3, in good agreement with the 0.86 g/cm3 value extracted from NIST WebBook (National Institute of Standards and Technology).24 For each solvated asphaltene system, the initial configuration was constructed in a box with the 2 or 3 asphaltene molecules models. Two starting points were envisaged: 1) their poly-aromatic cores are anti-parallel to each other and the aggregate is already formed (this is called the “organized” configuration); and 2), the molecules were randomly inserted in the simulation box. For the former case, the spacing between the mass centers of two neighboring asphaltene

molecules is 0.35 nm in the π-stacking direction. Finally, for both cases, the box was randomly filled with toluene molecules. As this paper’s aim is to study the stability of the aggregate, only the former case is thoroughly reported in the main body text. Table 1 shows the systems features for this particular situation. The results for the latter case can be found in the ESI. All partial charges for the compounds were used in the topologies based on the GROMOS96 force field parameters.12 For C=S group, the charges were applied as described by Huang and co-workers.25 The system energy was minimized using the steepest descent method, followed by conjugate gradient. A cutoff of 1.2 nm was used for both Coulomb and van der Waals interactions during energy minimization. Simulations were carried out under the NPT ensemble at 298 K and 1 bar pressure during 150 ns.16 For the first 3 ns, the Berendsen thermostat and barostat19 were used, followed by NoséHoover thermostat20, 21 and ParrinelloRahman pressure coupling algorithm.22 Periodic boundary conditions (PBC) were applied in the x, y, and z directions, and a leapfrog Verlet algorithm26 with a time step of 2 fs was used for integration of the trajectories. The electrostatic interaction was computed using the Particle-Mesh Ewald summation (PME)27 method with a fast Fourier transform (FFT) grid spacing of 0.16 nm to account for long-range electrostatic interactions of the system. A cutoff of 1.4 nm was used for the van der Waals interactions. All bond lengths in our system were constrained using LINCS algorithm.28 A neighbor list with a cutoff of 1.2 nm was updated every 5 steps. The initial atomic velocities of the system were set using the MaxwellBoltzmann distribution employed in GROMACS at the specified temperature of 298 K. 3


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Data Analysis After 150 ns of MDS, the system’s properties were analyzed using the GROMACS built-in analytical tools. To quantify the intensity of the interactions, we investigated the number of contact frequency by the minimum distances between poly-aromatic cores (π−π interaction), between poly-aromatic core and the side chains (π−θ interaction), and between side chains (θ−θ interaction) were investigated, as described by Jian and co-workers.29 A contact was considered when the minimum distance between any atoms of the two groups was ≤ 0.5 nm. The geometry of the poly-aromatic cores was also quantified by the cosine of angle ω between the poly-aromatic planes. The distances and the hydrogen bonds between the monomers were analyzed as well. RESULTS AND DISCUSSION Molecule 1 was already studied by MDS by Teklebrhan and co-workers8, 16 and Gao and co-workers,7 with results similar to the ones found in this study, i.e., the dimers of these molecules are stable and keep interacting throughout the simulation. Figures 2 and 3 show the initial (0 ns) and the final configurations (150 ns) of all systems for which the ordered starting point was used. Visual inspection of the MDS trajectories indicates that all systems keep their aggregated stack in place throughout the simulation, although some of them, such as 3T, has a final configuration where only two of the asphaltene molecules are in an aggregated state whereas the third is not in interaction with the stack. This is not found for the case where the starting configuration is random, as it can be seen in Figure S14. Further details why this happens are given in the next sections.

Page 4 of 23

Thus, the behavior of both dimers and trimers are explained in more details in the next sections. Dimers (Systems 1D, 2D, 3D, and 4D): The analysis of monomer distances shows that systems 1D through 4D are stable as dimers, regardless the initial configuration within the box (Figure 4 and S7). For 1D, the average distance between the two molecules is lower that 0.4 nm during ~88% of the simulation, when the aggregated starting point is used and during 90% for the random starting point situation. The other systems (2D, 3D and 4D) are also stable and are found in the aggregated state during 85 (96), 94 (98) and 96 (94)% of the simulation for the “organized” (random) starting point. These details, for the former case, are presented in Table 2. The formation of hydrogen bonds both between two acid chain ends (called herein θ−θ) or between an acid chain end and the carbonyl group (called π−θ) of the conjugated core are probable to take place for all the 1D-4D systems. In Figure 51, one can find the counting of these bonds formed during the simulation. In this run, the molecules of systems 2D and 4D rotate around their normal axis and have a final configuration where the θ−θ hydrogen bond configurations are more likely than the π−θ. Moreover, one should note that, as it was expected, system 2D has a lower probability of forming π−θ hydrogen bonds interactions because a carbonyl group was replaced by a thionyl one. 1

Hydrogen bonds are defined based on a geometric criterion, specified by the maximum hydrogen-donor-acceptor angle (α≤30º) and donor-acceptor distance (r≤0.35nm).The percentage is related to the number of frames in which the HB is presented between defined groups, following the above description.


Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

In order to investigate the mean type of interactions between monomers, the intermolecular contacts were calculated. Figure 6 shows the numbers of π−π, π−θ, and θ−θ contacts as a function of simulation time for systems 1D, 2D, 3D, and 4D. There is no meaningful change in the number of contacts for all systems during all the simulations. Figure 6 and Table 3 show the great difference at the average number of π−π contacts when compared to θ−π and θ−θ contacts and it suggests that the main type of interactions is produced by the poly-aromatic cores, what is expected for this type of conjugated systems. The cosine of the angle between the aromatic cores’ planes is a useful complementary information towards the determination of the stability of the stack.29 Values of ω = 0° (cos ω = 1) or ω = 180° (cos ω = -1) corresponds to parallel monomers (polyaromatic cores as π−π stacking), and ω = 90° (cos ω = 0) corresponds to perpendicular molecules (T-shape). For the “organized” starting point, the cosine indicates planar co-facial interacting structures, except for system 1D, which has some fluctuations during the simulation, as it can be seen in Figure 7. For the random starting point (Figure S8), these fluctuations are more common for all the systems in the beginning of the simulation and system 2D has it in the middle of the simulation time. This indicates that, although the parallel cofacial interaction is preferential, as it is also indicated by the minimum distances values, the molecules have some freedom to rotate around their normal axis during the MDS under these thermodynamic conditions. Up to this point, we cannot discern and indicate if this is an effect of the place of the sulfur atom in the structure or if is is just the fact that the

hydrogen bonds are dynamic and allow the molecules to rotate. In any case, these results are in agreement with the ones found by Teklebrhan and co-workers.16 What can be said though is that the 1D4D systems do not perform T-shape interactions, but this is the case for the system 5D and 6D, as it is reported in the ESI (Figure S10). Trimers (Systems 1T, 2T, 3T, and 4T): Previous molecular dynamics simulations have shown that molecule 1 (C5Pe) has a high tendency to dimerize in toluene, but it does not form larger aggregates, since it has a long and linear alkyl side chains.16 However, the distance between the monomers of system 1T is constant from the beginning of the simulation (Figure 8), indicating an average height lower than 0.8 nm (Table 4). This is in agreement with experimental work,4 which suggest a height of 0.67 nm. Systems 2T and 4T (after 33 ns) exhibit similar results (Figure 8, Table 4). As it can be seen from these figures and the ones found in the ESI (S12-S15), the fluctuations in these distances are a function of the starting configuration. In “organized” starting configuration, system 3T shows larger distances after 45 ns and the molecules are close to each other (< 0.4 nm) at only 4% of the simulation. For system 3T, the results show variable molecular dissociation of one monomer, but the remaining dimer is stable. For the random starting configurations, systems 1T and 3T are stable as face-to-face aggregates during almost all the simulation time, whereas 2T takes 75 ns to stabilize in this configuration. The 4T now has the same effect in this situation that 3T had before: the trimer is not stable in the face-to-face configuration although the three molecules are kept together by the 5


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

hydrogen bonds and the two of them assume the face-to-face configuration, as it is the case of 4D system. When the predominant type of interactions was investigated for the case with the “organized” starting point, it was noticed they are constant during the whole simulation for system 1T, 2T, and 4T (Figure 9). Those molecules seem to be stable as a trimer. The average number of π−π contacts (including the three monomers) is 769, 765, and 700, and the π−θ contacts is 137, 114, and 112, respectively (Table 5). Even more, for systems 1T, 2T, and 4T, the number of contacts π−π is doubled compared to system 3T. The interaction between the first two aromatic cores is not influenced when a third monomer is added, which in turn, has an identical behavior with the second molecule. System 3T also shows a lower average number of θ−θ contacts (Figure 9, Table 5), since it corresponds only to the remaining dimer. This is a result of the fact that the third monomer is not found in a face-to-face configuration concerning the rest of the aggregate. The number of π−π contacts in general is significantly higher than π−θ and θ−θ contacts to all systems. It suggests that the interaction driving the aggregation is made from the poly-aromatic cores. Although, the side chain should contribute to aggregation since it interacts during all the simulation for the four systems. The monomers geometries can be also analyzed by the cosine values of the angle between poly-aromatic cores (Figure 10). For all trimer systems, the planar configuration is more likely than others, namely the T-shape. System 3T, in the “organized” starting point (Figure 10) and system 4T in the random starting point (Figure S15) show now variations in these angles, either in the beginning or during the whole simulation. This, added

Page 6 of 23

to the fact that the intermolecular distance is similar for both 1-3 and 2-3 molecules, is an indication that the dimer is stable and interacts with the third molecule via the hydrogen bonds. The stable trimers also interact through hydrogen bonds during the simulation. On systems 2T, 3T and 4T, the hydrogen bonds interactions occur basically between carboxylic groups of side chains (Figure 11). For systems 4T, the frequency of these hydrogen bonds is higher than 96% of the total simulation. For systems 3T, the numbers decrease to 63%, since only side chains from monomers 1 and 2 interact. Furthermore, only for system 1T, the numbers of hydrogen bonds between the side chains and the carbonyl group of poly-aromatic core are significant (80%). Moreover, the presence of carboxylic acid groups is important to aggregation, but they were not enough present to maintain the trimer in a parallel orientation, as observed in systems 1T, 3T and 4T (from random starting point). General effect of the Sulfur atom: Up to now, the properties of the aggregates of molecules 1-4 has been discussed for dimers and trimers. Although some differences in their dynamics in solution could be noticed during MDS, no final conclusion on the role of the sulfur hetero-element can be extracted so far. In order to circumvent this shortcut, we have introduced, for the random starting configurations, the study of the radial distribution functions (RDF) for the S-X pairwise interactions, where X can be: 1 – the oxygen atoms of the aromatic core (aromatic carbonyl groups), 2 – the oxygen atoms of the carboxylic acid group (COOH), and 3 – the hydrogen atom of the same group. This was motivated since it is known that sulfur atoms can form non-covalent (Coulomb 6


Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

and/or van der Waals) interactions with oxygen and hydrogen,32 sometimes strong enough to lock conformations in place. These graphics can be found in Figure S11 for the dimer and S18 for the trimer systems. In dimer and trimer systems, all the peaks found in the S-aromatic carbonyl oxygen RDFs are due to intra-molecular configurations. This is known by measuring the S-O distances of an isolated molecule of each type. Minor peaks, as it is the case of the ones found around 3, 8 and 14 Angstroms for system 2D, are due to the interaction with oxygen atoms of the other molecule interacting with the one bearing the reference sulfur atom. The other systems show similar behavior with wider peaks now due to the fact that sulfur atoms are in flexible parts of the molecule. This allows us to state that there is no particular affinity between sulfur atoms and aromatic carbonyl atoms in the studied structures and this cannot be a likely interaction mechanism. The S-O (COOH) and S-H have now new features that could not be observed in the previous case. Considering both dimer and trimer systems, these two RDFs have intra-molecular peaks centered at around 18 Å for molecule 2 (systems 2D and 2T), 22 Å for molecule 3 (systems 3D and 3T), and 26 Å for molecule 4 (systems 4D and 4T – out of the measurement range, but one can see the peak in the end of the graph). The other peaks are due to inter-molecular interactions between adjacent molecules. Molecule 2, regardless if it is a dimer or a trimer, does not have these additional peaks, indicating that the presence of the sulfur atom on the conjugated backbone does not have any particular affinity with neither oxygen or polar hydrogen atoms of the other molecules. On the other hand, molecule 3 (system 3D) has a “bump” for both S-O (COOH) and S-H (COOH)

RDFs centered at 8 Å that does not come from an intra-molecular interaction and is a signature of inter-molecular affinity between the sulfur atom and these oxygen and hydrogen atoms. This behavior is more pronounced in trimer systems, as it expected, since this interaction can take place more easily. This behavior is still more pronounced for molecule 4 (system 4D and 4T): the S-O and S-H inter-molecular interactions are now very likely to take place and, considering the peak centered in 2.5 Å of the S-H RDF that is not present in the SO curve (see Figure S11), one could say that there is also an affinity of this terminal –SH group with the acid chain ends of the neighbor molecule. This same characteristic is found for trimer systems (Figure S18). Based on these observations, one can state that sulfur atoms grafted directly to the conjugated backbone has a little influence on the dynamic behavior of dimer and trimer asphaltene systems and one might not be able to differentiate this with what is observed for the case where there is no sulfur atom (molecule 1). Oppositely, the presence of sulfur atoms on the lateral chains, either in the middle of it or as a chain end, has a more pronounced effect on the properties of these molecules (molecules 3 and 4). This is due to the fact that this atom can now interact with acid chain ends of neighbor molecules by means of non-covalent interactions like Coulomb and/or van der Waals. Although the differences herein highlighted can, in our scale, not be of uppermost importance, it is the case in oil industry where huge quantities of molecules are being treated at the same time. These results can help towards the comprehension why crude oils with an elevated sulfur content are more difficult to crack, as a matter of fact. 7


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

CONCLUDING REMARKS Several asphaltene structural models have been proposed in the literature.14, 16, 29-31 Based on experimental work of Eyssautier and co-workers (2011),4 we have chosen four different model molecules for investigating the role of the sulfur atom position within the asphaltene structure. The molecular dynamics simulations have allowed understanding the stability of asphaltene dimers and trimers in toluene as well as showing that the position of the sulfur atom has an effect on the intermolecular non-covalent interactions that can take place in these systems. Although the polar side chain is important for determining the aggregation, the main interactions between the molecules are produced by the poly-aromatic cores, which are responsible to the parallel orientation of molecules. This geometry is confirmed by the cosine of angles between the polyaromatic cores. We suggest that the presence of one sulfur atom in the poly-aromatic core

Page 8 of 23

(molecule 2) do not prevent the aggregation, since systems 2D and 2T are stable and the radial distribution functions of the pairwise interaction of this atom and the oxygen atoms of the neighborhood do not show any particular affinity. However, the presence of four sulfur atoms in poly-aromatic core avoids the molecules aggregation in a major part of the simulation and this may be due to both steric hindrance, Coulomb repulsion between sulfur and nitrogen atom of the adjacent molecule. Moreover, this molecular structure can not form hydrogen bonds between the acid chain end and the conjugated core of the molecule. On the other hand, sulfur atoms in the middle of the aliphatic chain, as well as a chain end, have a measurable effect on the stability of both dimers and trimers and this is due to the fact that they have a low, still measurable, interaction with the acid chain end oxygen and hydrogen atoms. Finally, these results can help towards the comprehension why heavy crude oils containing high values of sulfur are experimentally more difficult to treat and to crack. 33-35


Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1

2

3

4 Figure 1. Molecular structures of asphaltene molecules used in this study.


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 23

Figure 2. Snapshots of the initial configuration and the final frame of the dimers simulations. Carbon atoms are in grey, oxygen in red, nitrogen in blue and sulfur in yellow. Hydrogen atoms are omitted for clarity.

Figure 3. Snapshots of the initial configuration and the final frame of the trimers simulations. Carbon atoms are in grey, oxygen in red, nitrogen in blue and sulfur in yellow. Hydrogen atoms are omitted for clarity.


Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 4. Distances between monomers. System 1D is in black, system 2D is in orange, system 3D is in purple, and system D is in blue.


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 23

Figure 5. Number of hydrogen bonds between the carboxylic acids groups from the side chains (θ-θ, in orange) and between the carboxylic acids groups from the side chains and the carbonyl group from the polyaromatic core (π-θ, in black).


Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 6. Number of contacts of systems 1D, 2D, 3D, and 4D. π−π contacts are in black, θ−π contacts are in orange, and θ−θ contacts are in purple.


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

Figure 7. Cosine of angle between aromatic cores in function of distance between dimers monomers.Carbon atoms are in grey, oxygen in red, nitrogen in blue and sulfur in yellow. Hydrogen atoms are omitted for clarity.


Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 8. Distances between monomers of systems 1T, 2T, 3T, and 4T. Distances between monomers 1 and 2 are in black, distances between monomers 2 and 3 are in orange and distances between monomers 1 and 3 are in purple.

Figure 9. Number of contacts of systems 1T, 2T, 3T, and 4T. π−π contacts are in black, θ−π contacts are in orange, and θ−θ contacts are in purple. The number of contacts takes into account the contributions of interactions between monomers 1 and 2, molecules 2 and 3 and molecules 1 and 3.


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 23

Figure 10. Cosine of angle between aromatic cores in function of time and distance. Cosine between monomers 1 and 2 are in black, 2 and 3 in orange, and 1 and 3 in purple.Carbon atoms are in grey, oxygen in red, nitrogen in blue and sulfur in yellow. Hydrogen atoms are omitted for clarity.


Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 11. Number of hydrogen bonds between the carboxylic acids groups from the side chains (θ-θ, in orange) and between the carboxylic acids groups from the side chains and the carbonyl group from the polyaromatic core (π-θ, in black).


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 23

Table 1. Systems simulated in toluene. System Compound

Number of Asphaltene Molecules

Number of Solvent Molecules

Size of Simulation Box (nm3)

Simulation Time (ns)

1D

1

2

585

5x5x5

150

2D

2

2

584

5x5x5

150

3D

3

2

581

5x5x5

150

4D

4

2

581

5x5x5

150

1T

1

3

902

5 x 7.5 x 5

150

2T

2

3

888

5 x 7.5 x 5

150

3T

3

3

888

5 x 7.5 x 5

150

4T

4

3

902

5 x 7.5 x 5

150

Table 2. Average distances between monomers during the molecular dynamic simulation. System Average Distance 1-2 (nm) ± SD

%*

1D

0.45 ± 0.16

87.9

2D

0.41 ± 0.05

84.4

3D

0.40 ± 0.04

93.9

4D

0.39 ± 0.05

96.4

*Percentage of frames which distance is under 0.4 nm


Page 19 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Table 3. Average number of contacts during the molecular dynamic simulation of dimers. AverageNumber of Contacts Contacts System 1D System 2D System 3D System 4D π-π

373

386

392

392

π- θ

66

40

47

55

θ-θ

22

27

18

36

Table 4. Distances between monomers of trimers during the molecular dynamic simulation. System

Average Distance 1-2 (nm)

%*


%*


%**

1T

0.44 ± 0.12

58.9

0.46 ± 0.07

46.6

0.74 ± 0.09

90.9

2T

0.41 ± 0.05

88.0

0.40 ± 0.04

93.5

0.79 ± 0.06

88.9

3T

5.00 ± 2.24

0.0

0.55 ± 0.06

4.22

5.03 ± 2.22

0.0

4T

0.96 ± 0.51

0.0

0.44 ± 0.08

62.7

0.70 ± 0.69

77.7***

*Percentage of frames which distance is under 0.4 nm **Percentage of frames which distance is under 0.8 nm ***Monomer 3 of system 4T has 73.9% of frames with distance under 0.4 nm from monomer 1


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 23

Table 5. Average number of contacts during the molecular dynamic simulation of trimers. AverageNumber of Contacts Contacts System 1T System 2T System 3T System 4T π-π

769

765

395

700

π- θ

137

114

37

112

θ-θ

45

81

28

101

ASSOCIATED CONTENT Supporting Information. Two other molecules (molecules 5 and 6) were also studied and are explained at Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author *[email protected] PresentAddresses † Laboratório de Modelagem Molecular e QSAR (MODMOLQSAR), Departamento de Fármacos e Medicamentos, Universidade Federal do Rio de Janeiro (UFRJ), Av. Carlos Chagas Filho, 373, 21941-599, Brazil.

ACKNOWLEDGMENT We are thankful to Universidade Federal do Rio de Janeiro (UFRJ), Brazil as well as the Université de Pau et des Pays de l’Adour for computational power. We are also warmly thankful to the Mésocentre de Calcul Intensif Aquitain and Centre Informatique National de l’Enseignement Supérieur, who also provided computational power needed to accomplish this work. ABBREVIATIONS SAXS, X-ray small-angle scattering studies; SANS, neutron small-angle scattering studies; MDS, molecular dynamic simulations.


Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

REFERENCES 1. Mullins, O. C., The asphaltenes. Annu Rev Anal Chem (Palo Alto Calif) 2011, 4, 393-418. 2. Adams, J. J., Asphaltene Adsorption, a Literature Review. Energy & Fuels 2014, 28, (5), 2831-2856. 3. Benamsili, L.; Korb, J.-P.; Hamon, G.; Louis-Joseph, A.; Bouyssiere, B.; Zhou, H.; Bryant, R. G., Multi-dimensional Nuclear Magnetic Resonance Characterizations of Dynamics and Saturations of Brine/Crude Oil/Mud Filtrate Mixtures Confined in Rocks: The Role of Asphaltene. Energy & Fuels 2013, 28, (3), 1629-1640. 4. Eyssautier, J.; Levitz, P.; Espinat, D.; Jestin, J.; Gummel, J.; Grillo, I.; Barre, L., Insight into asphaltene nanoaggregate structure inferred by small angle neutron and X-ray scattering. J Phys Chem B 2011, 115, (21), 6827-37. 5. Korb, J. P.; Louis-Joseph, A.; Benamsili, L., Probing structure and dynamics of bulk and confined crude oils by multiscale NMR spectroscopy, diffusometry, and relaxometry. J Phys Chem B 2013, 117, (23), 7002-14. 6. Sabbah, H.; Morrow, A. L.; Pomerantz, A. E.; Zare, R. N., Evidence for Island Structures as the Dominant Architecture of Asphaltenes. Energy & Fuels 2011, 25, (4), 1597-1604. 7. Gao, F.; Xu, Z.; Liu, G.; Yuan, S., Molecular Dynamics Simulation: The Behavior of Asphaltene in Crude Oil and at the Oil/Water Interface. Energy & Fuels 2014, 28, (12), 7368-7376. 8. Teklebrhan, R. B.; Ge, L.; Bhattacharjee, S.; Xu, Z.; Sjoblom, J., Initial partition and aggregation of uncharged polyaromatic molecules at the oil-water interface: a molecular dynamics simulation study. J Phys Chem B 2014, 118, (4), 1040-51. 9. Berendsen, H. J. C.; van der Spoel, D.; van Drunen, R., GROMACS: A messagepassing parallel molecular dynamics implementation. Computer Physics Communications 1995, 91, (1–3), 43-56. 10. Lindahl, E.; Hess, B.; van der Spoel, D., GROMACS 3.0: a package for molecular simulation and trajectory analysis. Molecular modeling annual 2001, 7, (8), 306-317. 11. Van Der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A. E.; Berendsen, H. J., GROMACS: fast, flexible, and free. J Comput Chem 2005, 26, (16), 1701-18. 12. Oostenbrink, C.; Soares, T. A.; van der Vegt, N. F.; van Gunsteren, W. F., Validation of the 53A6 GROMOS force field. Eur Biophys J 2005, 34, (4), 273-84. 13. Oostenbrink, C.; Villa, A.; Mark, A. E.; van Gunsteren, W. F., A biomolecular force field based on the free enthalpy of hydration and solvation: the GROMOS force-field parameter sets 53A5 and 53A6. J Comput Chem 2004, 25, (13), 1656-76. 14. Kuznicki, T.; Masliyah, J. H.; Bhattacharjee, S., Molecular Dynamics Study of Model Molecules Resembling Asphaltene-Like Structures in Aqueous Organic Solvent Systems. Energy & Fuels 2008, 22, (4), 2379-2389. 15. Kuznicki, T.; Masliyah, J. H.; Bhattacharjee, S., Aggregation and Partitioning of Model Asphaltenes at Toluene−Water Interfaces: Molecular Dynamics Simulations. Energy & Fuels 2009, 23, (10), 5027-5035. 16. Teklebrhan, R. B.; Ge, L.; Bhattacharjee, S.; Xu, Z.; Sjoblom, J., Probing structurenanoaggregation relations of polyaromatic surfactants: a molecular dynamics simulation and dynamic light scattering study. J Phys Chem B 2012, 116, (20), 5907-18.


Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 23

17. Hanwell, M. D.; Curtis, D. E.; Lonie, D. C.; Vandermeersch, T.; Zurek, E.; Hutchison, G. R., Avogadro: an advanced semantic chemical editor, visualization, and analysis platform. J Cheminform 2012, 4, (1), 17. 18. Schuttelkopf, A. W.; van Aalten, D. M., PRODRG: a tool for high-throughput crystallography of protein-ligand complexes. Acta Crystallogr D Biol Crystallogr 2004, 60, (Pt 8), 1355-63. 19. Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.; DiNola, A.; Haak, J. R., Molecular dynamics with coupling to an external bath. J Chem Phys 1984, 81, (8), 3684-3690. 20. Hoover, W. G., Canonical dynamics: Equilibrium phase-space distributions. Physical Review A 1985, 31, (3), 1695-1697. 21. Nosé, S., A molecular dynamics method for simulations in the canonical ensemble. Molecular Physics 1984, 52, (2), 255-268. 22. Parrinello, M.; Rahman, A., Polymorphic transitions in single crystals: A new molecular dynamics method. Journal of Applied Physics 1981, 52, (12), 7182-7190. 23. Lide, D. R., CRC Handbook of Chemistry and Physics. Internet Version 2005 ed.; CRC Press Boca Raton, FL, 145, 2005. 24. Shen, V. K.; Siderius, D. W.; Krekelberg, W. P., NIST Standard Reference Simulation Website, NIST Standard Reference Database Number 173. National Institute of Standards and Technology: Gaithersburg MD, 20899, 146, http://www.nist.gov/mml/csd/informatics_research/srsw.cfm, retrieved 22, december 2015. 25. Huang, W.; Eichenberger, A. P.; van Gunsteren, W. F., Molecular dynamics simulation of thionated hen egg white lysozyme. Protein Sci 2012, 21, (8), 1153-61. 26. Verlet, L., Computer "Experiments" on Classical Fluids. I. Thermodynamical Properties of Lennard-Jones Molecules. Physical Review 1967, 159, (1), 98-103. 27. Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.; Pedersen, L. G., A smooth particle mesh Ewald method. J Chem Phys 1995, 103, (19), 8577-8593. 28. Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M., LINCS: A linear constraint solver for molecular simulations. J Comput Chem 1997, 18, (12), 1463-1472. 29. Jian, C.; Tang, T.; Bhattacharjee, S., Molecular Dynamics Investigation on the Aggregation of Violanthrone78-Based Model Asphaltenes in Toluene. Energy & Fuels 2014, 28, (6), 3604-3613. 30. Acevedo, S.; Castro, A.; Negrin, J. G.; Fernández, A.; Escobar, G.; Piscitelli, V.; Delolme, F.; Dessalces, G., Relations between Asphaltene Structures and Their Physical and Chemical Properties: The Rosary-Type Structure. Energy & Fuels 2007, 21, (4), 21652175. 31. Andrews, A. B.; McClelland, A.; Korkeila, O.; Demidov, A.; Krummel, A.; Mullins, O. C.; Chen, Z., Molecular orientation of asphaltenes and PAH model compounds in Langmuir-Blodgett films using sum frequency generation spectroscopy. Langmuir 2011, 27, (10),6049-58. 32. Jackson, N. E.; Savoie, B. M.; Kohlstedt, K. L.; de la Cruz, M. O.; Schatz, G. C.; Chen, L. X.; Ratner, M. A., Controlling conformations of conjugated polymers and small molecules: the role of nonbonding interactions. Journal of the American Chemical Society 2013, 135, 10475-83.


Page 23 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

33. Nishioka, M., Whiting, D. G., Campbell, R. M., & Lee, M. L. (1986). Supercritical fluid fractionation and detailed characterization of the sulfur heterocycles in a catalytically cracked petroleum vacuum residue. Analytical Chemistry, 58(11), 2251-2255. 34. Lewan, M. D. (1998). Sulphur-radical control on petroleum formation rates. Nature, 391(6663), 164-166. 35. Drushel, H. V., & Sommers, A. L. (1967). Isolation and characterization of sulfur compounds in high-boiling petroleum fractions. Analytical Chemistry, 39(14), 1819-1829.


Investigation of the Effect of Sulfur Heteroatom on Asphaltene

Recommend Documents