Asphaltene Aggregation and Impact of Alkylphenols | Langmuir

Article pubs.acs.org/Langmuir

Asphaltene Aggregation and Impact of Alkylphenols Lamia Goual,* Mohammad Sedghi, Xiaoxiao Wang, and Ziming Zhu Department of Chemical and Petroleum Engineering, University of Wyoming, 1000 East University Avenue, Laramie, Wyoming 82071, United States ABSTRACT: The main objective of this study was to provide novel insights into the mechanism of asphaltene aggregation in toluene/heptane (Heptol) solutions and the effect of alkylphenols on asphaltene dispersion through the integration of advanced experimental and modeling methods. Highresolution transmission electron microscope (HRTEM) images revealed that the onset of asphaltene flocculation occurs near a toluene/heptane volume ratio of 70:30 and that flocculates are well below 1 μm in size. To assess the impact of alkylphenols on asphaltene aggregation, octylphenol (OP) and dodecylphenol (DP) were evaluated by impedance analysis based on their ability to delay the precipitation onset and to reduce the size of nonflocculated asphaltene aggregates in 80:20 toluene/heptane solutions. Although a longer dispersant chain length did not affect the precipitation onset, it reduced the size of the aggregates. Molecular dynamics simulations were then performed to understand the mechanism of interaction between a model asphaltene and OP in heptane. OP molecules saturated the H-bonding sites of asphaltenes and prevented them from interacting laterally between themselves. This explained why OP favored the formation of flocculates with filamentary rather than globular structures, which were clearly observed by HRTEM. Although OP proved to be an effective dispersant, its effectiveness was hindered by its self-association and the fact that it interacted at the periphery of asphaltenes, leaving their aromatic cores uncovered.

1. INTRODUCTION Asphaltenes impact virtually all aspects of crude oil utilization. They constitute the heaviest and more polarizable fractions of oil and are a solubility class typically defined as the heptaneinsoluble, toluene-soluble fraction of oil. The structure of asphaltenes consists of a polynuclear aromatic (PNA) core with 4−10 aromatic rings and peripheral aliphatic chains with lengths ranging from 3 to 7 carbons.1 Both the asphaltene core and side chains can contain heteroatoms. Some asphaltenes may also contain porphyrin structures with metals.2 Because of their aromatic and polar character, asphaltene molecules have a high propensity to associate into dimers, trimers, and so forth. However, they stop growing after a few molecules because of the steric stabilization produced by their peripheral side chains.3,4 There is a confluence of evidence from high-Q ultrasonics,5 static light scattering,6 nuclear magnetic resonance (NMR),7 dc conductivity,8,9 and centrifugation10,11 measurements indicating that the concentration at which nanoaggregates stop growing (also called the critical nanoaggregate concentration or CNAC) is about 100 ± 50 mg/L for different asphaltenes. Asphaltene nanoaggregates contain 6−9 molecules stacked on top of each other12−15 and have an average diameter of 2 nm.3 When the concentration of asphaltenes in toluene increases, nanoaggregates can further associate into clusters with an average diameter of 5 nm.16,17 Dc conductivity studies revealed that the critical cluster concentration (CCC) of asphaltenes © XXXX American Chemical Society

occurs at about 1.5−2 g/L in toluene, which is more that 10 times greater than the CNAC.11 Clusters can also form upon destabilization of crude oils by precipitants such as alkanes, eventually leading to flocculation and precipitation. Flocculation occurs when clusters grow to a point where they become colloidally unstable in solution. Precipitation ensues when flocculates are too heavy to remain in solution and undergo phase separation. Recent studies by Hoepfner et al. revealed that clusters become unstable and flocculate at toluene/heptane ratios as low as 70:30 v/v.18,19 This may explain why cluster sizes determined by dynamic light scattering (DLS) remained almost constant with the addition of heptane to toluene up to 20−30 vol %.20,21 Clusters show a very different kinetic behavior than nanoaggregates in toluene/heptane mixtures. Nanoaggregates exhibit a fast diffusion-limited aggregation, whereas clusters exhibit a relatively slow reaction-limited aggregation20,22 that can lead to fractal structures,23−25 with fractal dimension typically ranging from 1.3 to 2.0.3,26,27 In order to decrease the size of aggregates below the flocculation threshold or delay the onset of precipitation, amphiphilic dispersants have been used in the past with variable success. An amphiphilic dispersant generally contains an “anchoring” polar group, which attaches itself to the asphaltene Received: February 19, 2014 Revised: April 29, 2014

A

dx.doi.org/10.1021/la500615k | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

surface, and a “blocking” alkyl group, which blocks other asphaltene molecules and allows the dispersant to be soluble in aliphatic solvents. The polar group contains heteroatoms, whereas the aliphatic tail usually has less than 16 carbons to avoid crystallization,28 unless branched double tails are used.29 In 1994, Chang and Fogler carried out a comprehensive and systematic experimental study to understand the mechanisms of asphaltene dispersion in aliphatic solvents using low-molecularweight alkylbenzene-derived amphiphiles.30,31 The structure of the dispersants included an aliphatic tail attached to a benzene ring with one or two functional or head groups (−OH, −SO3H, and −O−). Among nonionic dispersants, p-alkylphenols displayed a high performance in dispersing asphaltenes for two reasons: (i) hydrogen bonding between amphiphiles and asphaltenes and (ii) π−π interactions between their aromatic rings. The adsorption propensity of alkylphenols on asphaltene surfaces has often been used as a means to probe their efficiency. Leon et al. measured a two-step adsorption isotherm of alkylphenols on asphaltenes: (i) adsorption through interactions with asphaltene surface and (ii) adsorption due to amphiphile−amphiphile interactions, leading to the formation of amphiphile aggregates at the surface.32 The same authors reported that 6 to 7 molecules of nonylphenol are involved in the peptizing of each asphaltene molecule, which was later confirmed by microcalorimetry.33 Rogel and Leon performed molecular dynamics (MD) simulations to study the adsorption behavior of amphiphiles on asphaltene surface and showed that the adsorption energy is proportional to the amphiphile’s dipole moment at low coverage and to its polarizability at high coverage.34 In other Monte Carlo simulations, Barcenas et al. demonstrated that nonylphenols manifest high adsorption ability on asphaltenes at low concentration in toluene, whereas at high concentration, they prefer to self-associate in the bulk rather than adsorb on asphaltenes.35 Thus, the self-assembly of amphiphilic molecules in the solvent may be a reason for their declining efficiency with increasing concentration. Studies relating the structure of alkylphenols to their function revealed that the presence of oxyethylenic groups in the molecule might decrease their dispersal ability.36 However, the performance of ethoxylated nonylphenols increased with the degree of ethoxylation.37 Hernandez-Trujillo et al. introduced the solvation free energies between p-alkylphenols and heptane computed by quantum mechanics as a way to evaluate their performance.38 The alkyl chain length and its relationship with the electronic structure of molecules was examined as a function of solvation energies. Pentylphenol had the shortest alkyl chain necessary to dissolve asphaltenes in heptane efficiently. From these results, amphiphile−solvent interactions explained the ability of alkylphenols to disperse asphaltenes in heptane, which means that we should not only consider amphiphile−asphaltene interactions when testing for a potential dispersant but also its interactions with the solvent. Laux et al. proposed that dispersants can be effective if their Hbonding with asphaltenes is near 10 [MJ m−3]0.5.39 Thus, the cohesion energy parameters according to Hansen can be used as a good criterion for the selection of asphaltenes dispersants.40 On the basis of this theory, alkylphenols may have a good stabilizing effect on asphaltenes in heptane if they have at least 9 carbon atoms in their alkyl chain. Although p-alkylphenols were found to be effective dispersants for asphaltenes to some extent, the mechanism by

which they stabilize asphaltenes is not very clear. The majority of previous studies were performed in heptane, where asphaltenes are already precipitated, and very few have targeted asphaltene aggregates before or near flocculation. The main objective of this study is to provide novel insights into both the mechanism of asphaltene aggregation in toluene/heptane solutions and the effect of octylphenol (OP) on the dispersion of asphaltene aggregates. To this end, three different methods were used: (i) impedance analysis (IA), (ii) high-resolution transmission electron microscopy (HRTEM), and (iii) MD. The goal of IA was to measure the onset of asphaltene precipitation in toluene/heptane solutions containing OP dispersants and to calculate the size of nonflocculated aggregates from their electrical conductivity in Heptol. With this approach, the dispersant could be evaluated on the basis of its ability to delay the precipitation onset and to reduce the size of asphaltene aggregates. Dodecylphenol (DP) was also used to investigate the impact of dispersant chain length on aggregation. The efficiency of OP was further verified by direct HRTEM imaging of asphaltenes in toluene/heptane solutions. MD simulations using high-performance computing were then performed with a model asphaltene and OP dispersant in heptane. The structure of the model asphaltene was chosen so that different types of interactions (including hydrogen bonding) were possible with dispersant molecules. Umbrella sampling and long-term MD simulations could explain the unique asphaltene flocculate structures observed by HRTEM.

2. EXPERIMENTAL METHODS 2.1. Materials. Materials include a Canadian crude oil from Alberta (see the crude oil properties in Table 1), anhydrous heptane (>99%), anhydrous toluene (>99.8%), 4-octylphenol (>99%), and 4dodecylphenol (>99%), all from Sigma-Aldrich.

Table 1. Properties of Canadian Crude Oil Property

Value

ρ20°C (g/cm3) C (%) H (%) N (%) O (%) S (%) H/C ASPH (wt %) TAN (mg of KOH/g) TBN (mg of KOH/g) TBN/TAN

0.943 81.29 8.18 1.10 1.41 8.28 1.21 14.04 1.140 2.797 2.5

2.2. Separation of Asphaltenes. The separation of asphaltenes was carried out according to ASTM D2007 by mixing crude oil with nheptane at a volume ratio of 1:40.41 The mixture was allowed to equilibrate after stirring and left overnight at room temperature. It was then filtered under vacuum using 0.2 μm pore size Whatman filter paper. The filter cake was repeatedly washed with n-heptane to remove any resins until the effluent from the filter became colorless. The asphaltenes were recovered from the filter cake by dissolution in toluene and then dried after toluene evaporation. 2.3. Impedance Analysis. Low-frequency impedance measurements were conducted using a 4294A precision impedance analyzer connected to a 16452A liquid test fixture through a 16048G 1 m port extension cable with four terminals, all from Agilent Technologies (Santa Clara, CA). The cell has a volume capacity of 4.8 mL and consists of two nickel-coated cobalt electrodes. The equipment was calibrated by performing phase compensation with a 4294A-1D5 highB


Langmuir

Article

stability frequency reference (Agilent, Santa Clara, CA). A short compensation was also performed using a shorting plate with a 1.3 × 10−3 m spacer in the liquid test fixture to account for lead capacitance. The equivalent circuit and schematic of the experimental setup can be found elsewhere.8,11 The parallel resistivity, Rp, of asphaltene solutions is determined from a fit with Z-view software and is used to calculate the dc conductivity according to σdc =

t AR p

Table 2. Structural Parameters of Asphaltene Model Molecule

(1) −3

where t is the electrode spacing (t = 10 m) and A is the electrode area (A = 0.001134 m2). The dc conductivity is related to the diffusion coefficient (D) via the Nernst−Einstein equation

D=

kBTMσdc CNAxe 2

(2)

where kB is the Boltzmann constant, T is the absolute temperature, e is the coulomb charge, NA is Avogadro’s number, M is the molecular weight of asphaltenes (∼1 kg/mol), C is the concentration of asphaltenes in kilograms per liter, and x is the ionic fraction of asphaltenes (∼10−4 in toluene).42 Because OP is a nonionic dispersant, x remains unchanged upon the addition of small amounts of OP. Assuming particles with spherical shape, their diameters, d, is calculated from the Stokes−Einstein equation, where ηs is the viscosity of the solvent. d=

kBT 3πηsD

The bias potentials can have any functional form. Often, harmonic potentials shown in eq 4 are used for their simplicity, where ξci is the position at which the system is restrained with a force constant Ki. K wi = i (ξ − ξic)2 (4) 2 The optimum umbrella force constant of asphaltene-like molecules is 1000 kJ/(mol nm2).48 The potential of mean force (PMF) can be calculated from unbiased probability distributions of the system through eq 5, where ξ0 is an arbitrary point and the PMF is zero.

(3)

2.4. High-Resolution Transmission Electron Microscopy. The Tecnai TF20 S-Twin HRTEM from FEI was used in this study. The microscope features a TIETZ F415MP 4k × 4k multiport CCD camera with a 4-port readout and 15 μm pixel size. Several asphaltene stock solutions were dissolved in Heptol with different toluene/ heptane ratios to image the size of aggregates and flocculates in the absence and presence of OP dispersant. The mixtures were placed inside an incubating shaker from VWR for 2 h and then allowed to rest under ambient conditions for 1 day before observation. A 15 nm silicon nitride (Si3N4) grid with 9 windows (0.1 × 0.1 mm) was dipped almost horizontally inside each sample for at least 30 seconds, dried for 5 min, and then imaged by the microscope at 200 kV accelerating voltage under bright-field illumination mode. ImageJ software was used to process the images. The images obtained by this procedure are different from those in previous HRTEM studies in which asphaltenes were preadsorbed on the grids prior to imaging.43,44

⎛ P(ξ) ⎞ w(ξ) = −KBT ln⎜ ⎟ ⎝ P(ξ0) ⎠

(5)

The unbiased probability distributions were obtained from the weighted histogram analysis method (WHAM) implemented in GROMACS. For statistical error calculations, we used Bayesian bootstrapping of complete histograms provided by the g_wham program in GROMACS.50 The PMF profile of two molecules (asphaltene−asphaltene and asphaltene−dispersant) in heptane versus the distance between their centers of mass (COM) was obtained by umbrella sampling simulations conducted in a cubic box (6.25 × 6.25 × 6.25 nm3) with periodic boundary conditions. All simulations were performed at a constant temperature (T) of 300 K and pressure (P) of 1 bar. The initial configurations for each umbrella window simulation can be generated from pulling the COM of one of the molecules along the z axis with the pull code implemented in GROMACS package at a rate of 2.6 nm/ns. During each window simulation, the two molecules could present any energetically favorable configurations because only one atom of the fixed molecule was constrained. The distance between two consecutive window simulations was about 0.5 Å. For each window, NPT simulation (i.e., at constant T, P, and number of molecules) was run for 300 ps to bring the system into the equilibrium temperature and pressure followed by an umbrella simulation that lasted for 10 ns (with a time step of 2 fs) to allow enough time for the two molecules to explore all possible configurations. A Nosé−Hoover thermostat was used for temperature control, whereas the Parrinello−

3. MOLECULAR DYNAMICS SIMULATIONS 3.1. Asphaltene Molecular Structure. MD simulations were performed using the GROMACS 4.5.5 simulation package and the OPLS-AA force field.45,46 The molecular structure of the model asphaltene used in this study is similar to structures provided by others47,48 and is shown in Table 2. The model asphaltene contains a pyridine group in one aromatic ring, a hydroxyl group in an aliphatic chain, and a sulfide group in another chain. The purpose of adding these three functional groups was to enable us to understand the effect of heteroatoms on the mechanism of asphaltene−dispersant interaction. For example, the hydroxyl group in the chain length can form Hbonding with some dispersants, whereas the nitrogen in the aromatic core can provide acid−base as well as H-bonding interactions. The structural parameters of this asphaltene model molecule are also listed in Table 2. Aromaticity factor (γ) is defined as the ratio of the number of aromatic carbons to the total number of carbons and can be measured by 13C NMR. 3.2. Umbrella Sampling. A series of windows were considered in which MD simulations were performed along the reaction coordinate ξ, biased by umbrella potentials wi(ξ).49 C


Langmuir

Article

Rahman algorithm was used for pressure coupling. The particlemesh Ewald (PME) algorithm was used to account for longrange electrostatic interactions, and the van der Waals (VDW) cutoff radius was set to 1.2 nm. 3.3. Asphaltene Aggregation. In long-time-scale MD simulations, 36 molecules (or 6.6 wt %) of model asphaltenes were placed in a heptane-filled cubic box (10 × 10 × 10 nm3) with periodic boundary conditions. In order to investigate the aggregation behavior of asphaltenes in the presence of dispersant, 6.6 wt % of OP dispersant was also considered. All simulations were performed at a constant temperature of 300 K and pressure of 1 bar. The original distances between each molecule were more than 1.5 nm. To equilibrate our systems, a quick energy minimization was first run followed by 200 ps of MD simulations to bring asphaltene and solvent molecules to equilibrium. The average T and P during NPT simulations was checked to ensure that the system was in equilibrium. The production simulations were run up to 150 ns. The distance between each pair of asphaltene molecules measured every 10 ps during the MD simulations provides a measure of aggregation. On the basis of our criteria, aggregation occurs if COM < 0.85 nm and does not occur if COM > 1.25 nm. If 0.85 < COM 1.25 nm, then aggregation would be accepted or rejected on the basis of the track of their distances in the past or following time frames. The z-average aggregation number, gz, was calculated from eq 6, where ni is the number of aggregates i containing gi monomers.21,48 gz =

Figure 1. HRTEM micrographs of 5000 ppm asphaltenes in (a) 80:20 and (b) 70:30 Heptol solutions imaged on Si3N4 grids.

5 nm, where stable asphaltene clusters have been detected. This study bridges this size gap and shows continuous connection. On the basis of these observations, mixtures of 5000 ppm asphaltenes in 80:20 Heptol were selected to investigate their size by impedance analysis below the onset of flocculation. Because the rodlike aggregates observed in Figure 1a have an average volume of 400 nm3, the solid sphere equivalent size of these aggregates is approximately 10 nm. The ionic fraction of asphaltenes in eq 3 was adjusted to 3 × 10−4 in 80:20 Heptol (instead of 10−4 in toluene) to obtain aggregate sizes of 10 nm at 5000 ppm concentration. Figure 2 displays the variations of asphaltene size with concentration in 80:20 Heptol. The data suggest that asphaltenes grow from individual clusters to clusters of clusters as their concentration increases in Heptol. 4.1.2. Effect of Dispersant on the Size of Asphaltene Aggregates. Before testing the performance of OP dispersant, we measured its conductivity in 80:20 Heptol up to 5000 ppm and found that it is more than 3 orders of magnitude lower than that of asphaltenes and remained constant with concentration. Moreover, when different amounts of OP were added to 5000 ppm asphaltenes in 80:20 Heptol, the conductivity of asphaltenes remained almost unchanged. Therefore, any changes in dc conductivity observed in the next experiments are mainly attributed to changes in asphaltene aggregation and not to OP. To investigate the impact of dispersant on asphaltene size, 5000 ppm of OP was first added to Heptol solutions, which were then mixed with asphaltenes. Although the addition of OP increased the conductivity of asphaltenes, it did not affect their CCC (∼1500 ppm). The increase in conductivity could be attributed to a nearly 15% reduction in aggregate size, as seen in Figure 2. This effect was even more pronounced at very low asphaltene concentration. For instance, at 20 ppm, asphaltenes were no longer in cluster but nanoaggregate state with a diameter of ∼2 nm. When the dispersant chain length was increased from 8 to 12, a more pronounced reduction in size

∑i nigi3 ∑i nigi2

(6)

4. RESULTS 4.1. Experimental Results. 4.1.1. Effect of Solvent on Asphaltene Aggregation. Toluene/heptane (Heptol) solutions were used to investigate the effect of solvent on the aggregation behavior of asphaltenes. Asphaltenes (5000 ppm) were added to 90:10, 80:20, and 70:30 Heptol solutions, stirred, and left overnight at room temperature. They were imaged the next day with HRTEM using the procedure described in Section 2.4. Asphaltenes in 90:10 Heptol could not be seen because the contrast between the aggregates and the Si3N4 grid was not high enough. However, we could clearly observe asphaltene aggregates in 80:20 Heptol (Figure 1a). The micrographs show individual aggregates and others interacting with each other. Most aggregates have a rodlike shape with a 5 nm diameter and 20 nm length, consistent with apparent sizes previously measured in Heptol.20,21 A closer look at these aggregates indicates that they consist of 5 nm spherical units, possibly clusters, almost aligned on top of each other. When 5000 ppm asphaltenes were mixed with 70:30 Heptol, the aggregates agglomerated into small colloidally unstable flocculates with almost globular shapes, as seen in Figure 1b. Some flocculates interacted with each other and grew into larger structures. This indicates that the onset of asphaltene flocculation is near 70:30 Heptol, in excellent agreement with recent studies.18,19 The flocculates have a diameter of 50−100 nm and are 1 order of magnitude smaller than values reported in previous studies.20,21,23,51 Indeed, most of previous determinations of asphaltene flocculation onsets were performed through optical detection, which is limited to 1 μm by the diffraction limit. There are relatively few studies covering the size range between 1 μm, where optical methods work, and D


Langmuir

Article

Figure 2. Variation of asphaltene average diameter with concentration with and without 5000 ppm of OP and DP dispersants in 80:20 Heptol solution.

Figure 3. HRTEM micrographs of (a) 5000 ppm asphaltenes in 30:70 Heptol with and without 5000 ppm of OP after 1 day and (b) 500 ppm asphaltenes with 500 ppm of OP in 70:30 Heptol after 1 month.

dispersant. Without OP, large asphaltene precipitates were clearly seen on the grid windows. However, almost no precipitate was observed in the presence of OP. It was possible to image asphaltenes stabilized by OP at higher magnification. Figure 3a shows that OP was able to reduce asphaltenes from

was observed (∼20%) because of the higher steric stabilization provided by DP. The performance of OP was further examined by HRTEM. Figure 3a presents the micrographs of 5000 ppm asphaltenes in 30:70 Heptol solution with and without 5000 ppm of OP E


Langmuir

Article

Figure 4. Flocculation and precipitation onsets of 5000 ppm asphaltenes: (a) without any dispersant and (b) with 5000 ppm of OP and DP.

Figure 5. PMF profiles of asphaltene−OP association in heptane: (a) H-bonding between the hydroxyl groups of asphaltene and OP and (b) aromatic interactions and H-bonding between the hydroxyl group of OP and the nitrogen atom of asphaltene.

precipitate to flocculate size. Some flocculates tend to align on top of each other to form long and thin structures. They are very different from flocculates in 70:30 Heptol without dispersants, which have a more globular shape (Figure 1b). In an attempt to image smaller aggregates, 500 ppm asphaltenes were mixed with 500 ppm OP in 70:30 Heptol. After 1 day, the contrast between aggregates and microscope grid was not high enough to observe clear images. However,

after aging the mixture for 1 month, very long chains of flocculates linked together (like filaments) were clearly observed and are shown for the first time in Figure 3b. These findings highlight the importance of kinetics when studying asphaltene aggregation, as emphasized in another work.52 The reason why such flocculates are obtained with OP will be explained in Section 4.2 through MD simulations. F


Langmuir

Article

Figure 6. Asphaltene and OP aggregation number in heptane versus time. (a) OP could break asphaltene clusters only when added in a sufficiently large amount (i.e., 33 wt %). (b) OP molecules tend to form trimers and quadrumers.

4.1.3. Effect of Dispersant on the Onset of Asphaltene Flocculation and Precipitation. The variations of asphaltene conductivity with the volume fraction of heptane in Heptol are plotted in Figure 4a for 5000 ppm asphaltenes. Two breaks in the slope of asphaltene conductivity versus heptane vol % can be seen. The first one occurs at the onset of asphaltene flocculation (∼25 vol % heptane), in agreement with the micrographs of Figure 1, and is caused by the size increase of colloidally unstable aggregates. The second break represents the onset of asphaltene precipitation (∼52 vol % heptane) and is due to the lower amount of asphaltenes in solution after phase separation. Figure 4b shows the effect of 5000 ppm OP on both onsets. Although the addition of OP did not change the onset of flocculation, it increased the onset of precipitation from 52 to 62 vol %. Furthermore, the performance of DP was almost the same as that of OP, suggesting that the amphiphile’s chain length does not affect the onset of precipitation. 4.2. Molecular Dynamics Simulation Results. 4.2.1. Umbrella Sampling. The Gibbs energy of association (ΔGa) between asphaltene model molecule (Table 2) and OP molecule was calculated from the PMF profiles obtained by

umbrella sampling simulation (Figure 5) to reveal the dispersant−asphaltene interaction strength and mechanism in heptane. The statistical error (∼1 kJ/mol) was calculated from the g_wham program provided in GROMACS. H-bonding between the hydroxyl groups of asphaltene and OP manifests as a minimum at 0.3 nm separation. An additional minimum can be seen at 1 nm and represents aromatic interactions and Hbonding between the hydroxyl group of phenol and the nitrogen atom of asphaltene. 4.2.2. Long-Time-Scale MD Simulations. MD simulations of 36 asphaltene molecules (or 6.6 wt %) in heptane were first run for 150 ns without OP to form nanoaggregates and clusters. The z-average aggregation number (gz) of asphaltenes in heptane is shown every 500 ps in Figure 6a. Nanoaggregates stop growing beyond 12 molecules and start forming clusters after 97 ns. Because of the limited number of molecules in the simulation box, clusters here consist only of two nanoaggregates. The presence of a hydroxyl group in asphaltene molecules increased the aggregation number of nanoaggregates well above 9, previously obtained using a similar molecule without the hydroxyl group (A04).48 When 6.6 wt % OP was G


Langmuir

Article

Figure 7. MD snapshots of OP molecules interacting at the periphery of an asphaltene (a) nanoaggregate and (b) cluster, leaving their aromatic cores uncovered.

added to asphaltenes at 50 ns, they could not break the nanoaggregates (Figure 6a). The same behavior was obtained with 33.0 wt % OP. The reason is that the aromatic interactions between OP’s benzene ring and asphaltene’s aromatic core are significantly smaller than those between asphaltene’s aromatic cores; therefore, OP molecules cannot interrupt the stacking of asphaltene molecules into nanoaggregates. However, OP can curb the clustering of asphaltene nanoaggregates. When OP molecules were added to asphaltene clusters at 110 ns, they could break clusters into nanoaggregates only when added in sufficiently large amounts (i.e., 33.0 wt %). This is in agreement with the low concentration data in Figure 2. Although 6.6 wt % OP was not able to break clusters, it prevented them from growing further. The data reveal that in order for OP to be effective (even at low concentration) it should be added at early stages of asphaltene aggregation. MD snapshots provided in Figure 7 show that OP interacts at the periphery of nanoaggregates and clusters by forming H-bonds with −OH and −N groups of asphaltenes. Thus, asphaltene aggregates are unable to get close to each other because of the steric repulsion between the aliphatic tails of OP. Figure 8 shows that the average number of hydrogen bonds between asphaltenes, calculated over a 10 or 30 ns period, was significantly reduced when OP was added to nanoaggregates and clusters. In addition to providing steric repulsion, OP dispersants saturated the Hbonding sites of asphaltenes and therefore diminished their

ability to form H-bonds between themselves. This may explain why OP favored the formation of the filamentary structures observed in Figure 3. The z-average aggregation number of OP molecules is presented in Figure 6b. OP molecules show a high tendency to aggregate into trimers and quadrumers because of their high dipole moment and H-bonding ability, in agreement with previous work.35 Although OP proved to be an effective dispersant by curbing further aggregation, its effectiveness was hindered by two factors: (i) OP molecules could self-associate through H-bonding and (ii) OP molecules preferred to stay at the periphery of asphaltenes and left their aromatic cores uncovered. Thus, a powerful dispersant blend should combine a dispersant like OP that interacts at the periphery of asphaltenes and another dispersant that interacts with the asphaltene aromatic cores, limiting their growth into long flocculates. Knowing that asphaltenes are a solubility class with a wide range of structures, the conclusions made herein may be valid only for asphaltenes with high H-bonding ability, such as the ones used in this study.

5. CONCLUSIONS This study provided novel insights into the mechanism of asphaltene aggregation in Heptol solutions and the effect of alkylphenols on asphaltene aggregates and flocculates. Rodshaped clusters of clusters in 80:20 Heptol solution could be seen with HRTEM. The onset of asphaltene flocculation occurred near 70:30 Heptol, in line with recent studies.18,19 Asphaltene flocculates have a diameter of 50−100 nm and are 1 order of magnitude smaller than the 1 μm threshold value reported in previous studies.20,21,23,51 On the basis of these results, impedance analysis was used to investigate the effect of alkylphenols (OP and DP) on the aggregation of asphaltene clusters in 80:20 Heptol. Dispersant efficiency was examined on the basis of its ability to delay the precipitation onset and to reduce the size of asphaltene aggregates. Although a longer dispersant chain length did not affect the onset of precipitation, it reduced the size of asphaltene aggregates. MD simulations were later performed to verify and interpret the experimental findings. The simulations revealed that OP molecules tend to form H-bonds with asphaltenes, in agreement with previous work.30,34 However, these bonds occur at the periphery of asphaltenes. As a result, nanoaggregates and clusters are unable to aggregate because of the steric repulsion between the aliphatic tails of OP. In addition, OP saturated the H-bonding sites of asphaltenes, diminishing their ability to form bonds between themselves. This explained why OP favored the

Figure 8. Average number of hydrogen bonds between asphaltenes with and without 6.6 wt % OP dispersant. OP was added to nanoaggregates at 50 ns and to clusters at 110 ns. H


Langmuir

Article

formation of filamentary flocculates, which were clearly observed by HRTEM. The effectiveness of OP was hindered by its self-association and the fact that it interacted at the periphery of asphaltenes, leaving their aromatic core uncovered.

■

petroleum- and coal-derived asphaltenes. Energy Fuels 2014, 28, 475− 482. (15) Majumdar, R. D.; Gerken, M.; Mikula, R.; Hazendonk, P. Validation of the Yen−Mullins model of Athabasca oil-sands asphaltenes using solution-state 1H NMR relaxation and 2D HSQC spectroscopy. Energy Fuels 2013, 27, 6528−6537. (16) 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, 7002−7014. (17) Mullins, O. C.; Seifert, D. J.; Zuo, J. Y.; Zeybek, M. Clusters of asphaltene nanoaggregates observed in oilfield reservoirs. Energy Fuels 2013, 27, 1752−1761. (18) Hoepfner, M. P.; Vilas Boas Favero, C.; Haji-Akbari, N.; Fogler, H. S. The fractal aggregation of asphaltenes. Langmuir 2013, 29, 8799−8808. (19) Hoepfner, M. P.; Limsakoune, V.; Chuenmeechao, V.; Maqbool, T.; Fogler, H. S. A fundamental study of asphaltene deposition. Energy Fuels 2013, 27, 725−735. (20) Burya, Y. G.; Yudin, I. K.; Dechabo, V. A.; Kosov, V. I.; Anisimov, M. A. Light-scattering study of petroleum asphaltene aggregation. Appl. Opt. 2001, 40, 4028−4035. (21) Mansur, C. R. E.; de Melo, A. R.; Lucas, E. F. Determination of asphaltenes particles size: Influence of flocculant, additive and temperature. Energy Fuels 2012, 26, 4988−4994. (22) Yudin, I. K.; Nikolaenko, G. L.; Gorodetskii, E. E.; Kosov, V. I.; Melikyan, V. R.; Markhashov, E. L.; Frot, D.; Briolant, Y. Mechanisms of asphaltene aggregation in toluene-heptane mixture. J. Pet. Sci. Eng. 1998, 20, 297−301. (23) Rahmani, N. H.; Dabros, T.; Masliyah, J. H. Fractal structure of asphaltene aggregates. J. Colloid Interface Sci. 2005, 285, 599−608. (24) Anisimov, M. A.; Yudin, I. K.; Nikitin, V.; Nikolaenko, G. L.; Chernoutsan, A. I.; Toulhoat, H.; Frot, D.; Briolant, Y. Asphaltene aggregation in hydrocarbon solutions studied by photon correlation spectroscopy. J. Phys. Chem. 1995, 99, 9576−9580. (25) Rassamdana, H.; Sahimi, M. Asphalt flocculation and deposition: II. Formation and growth of fractal aggregates. AIChE J. 1996, 42, 3318−3332. (26) Hoepfner, M. P.; Fogler, H. S. Multiscale scattering investigations of asphaltene cluster breakup, nanoaggregate dissociation, and molecular ordering. Langmuir 2013, 29, 15423−15432. (27) Barre, L.; Simon, S.; Palermo, T. Solution properties of asphaltenes. Langmuir 2008, 24, 3709−3717. (28) Chang, C.; Fogler, H. S. Peptization and coagulation of asphaltenes in apolar media using oil-soluble polymers. Fuel Sci. Technol. 1996, 14, 75−100. (29) Wiehe, I. A.; Jermansen, T. G. Design of synthetic dispersants for asphaltenes. Pet. Sci. Technol. 2003, 21, 527−536. (30) Chang, C.; Fogler, H. S. Stabilization of asphaltenes in aliphatic solvents using alkylbenzene-derivated 1. Effect of the chemical structure of amphiphiles on asphaltenes stabilization. Langmuir 1994, 10, 1749−1757. (31) Chang, C.; Fogler, H. S. Stabilization of asphaltenes in aliphatic solvents using alkylbenzene-derived amphiphiles. 2. Study of the asphaltene-amphiphile interactions and structures using Fourier transform infrared spectroscopy and small-angle X-ray scattering techniques. Langmuir 1994, 10, 1758−1766. (32) Leon, O.; Rogel, E.; Urbina, A.; Andujar, A.; Lucas, A. Study of the adsorption of alkyl benzene-derived amphiphiles on asphaltene particles. Langmuir 1999, 15, 7653−7657. (33) Merino-Garcia, D.; Andersen, S. I. Interaction of asphaltenes with nonylphenol by microcalorimetry. Langmuir 2004, 20, 1473− 1480. (34) Rogel, E.; Leon, O. Study of the adsorption of alkyl-benzenederived amphiphiles on an asphaltene surface using molecular dynamics simulations. Energy Fuels 2001, 15, 1077−1086. (35) Barcenas, M.; Orea, P.; Buenrostro-González, E.; ZamudioRivera, L. S.; Duda, Y. Study of medium effect on asphaltene agglomeration inhibitor efficiency. Energy Fuels 2008, 22, 1917−1922.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]; Phone: 307-766-3278. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the Center for Fundamentals of Subsurface Flow and the School of Energy Resources at the University of Wyoming for financial support. We are also grateful to Encana Corporation for providing crude oil samples, Dr. Erwin Sabio for performing HRTEM imaging, and Dr. Oliver Mullins for many helpful discussions.

■

REFERENCES

(1) Groenzin, H.; Mullins, O. C. Asphaltene molecular size and weight by time-resolved fluorescence depolarization. In Asphaltenes, Heavy Oils, and Petroleomics; Mullins, O. C., Sheu, E. Y., Hammami, A., Marshall, A. G., Eds.; Springer: New York, 2007; pp 17−62. (2) Speight, J. G. The Chemistry and Technology of Petroleum; Marcel Dekker, New York, 1980. (3) Mullins, O. C.; Sabbah, H.; Eyssautier, J.; Pomerantz, A. E.; Barré, L.; Andrews, A. B.; Ruiz-Morales, Y.; Mostowfi, F.; McFarlane, R.; Goual, L.; Lepkowicz, R.; Cooper, T.; Orbulescu, J.; Leblanc, R. M.; Edwards, J.; Zare, R. N. Advances in asphaltene science and the Yen− Mullins model. Energy Fuels 2012, 26, 3986−4003. (4) Freed, D. M.; Lisitza, N. V.; Sen, P. N.; Song, Y.-Q. Molecular composition and dynamics of oils from diffusion measurements. In Asphaltenes, Heavy Oils, and Petroleomics; Mullins, O. C., Sheu, E. Y., Hammami, A., Marshall, A. G., Eds.; Springer: New York, 2007; pp 279−300. (5) Andreatta, G.; Bostrom, N.; Mullins, O. C. High-Q ultrasonic determination of the critical nanoaggregate concentration of asphaltenes and the critical micelle concentration of standard surfactants. Langmuir 2005, 21, 2728−2736. (6) Evdokimov, I. N.; Eliseev, N. Y.; Akhmetov, B. R. Asphaltene dispersions in dilute oil solutions. Fuel 2006, 85, 1465−1472. (7) Lisitza, N. V.; Freed, D. E.; Sen, P. N.; Song, Y.-Q. A study of asphaltene nanoaggregation by NMR. Energy Fuels 2009, 23, 1189− 1193. (8) Goual, L. Impedance spectroscopy of petroleum fluids at low frequency. Energy Fuels 2009, 23, 2090−2094. (9) Zeng, H.; Song, Y.-Q.; Johnson, D. L.; Mullins, O. C. Critical nanoaggregate concentration of asphaltenes by low frequency conductivity. Energy Fuels 2009, 23, 1201−1208. (10) Mostowfi, F.; Indo, K.; Mullins, O. C.; McFarlane, R. Asphaltene nanoaggregates and the critical nanoaggregate concentration from centrifugation. Energy Fuels 2009, 23, 1194−1200. (11) Goual, L.; Sedghi, M.; Zeng, H.; Mostowfi, F.; McFarlane, R.; Mullins, O. C. On the formation and properties of asphaltene nanoaggregates and clusters by DC-conductivity and centrifugation. Fuel 2011, 90, 2480−2490. (12) Schneider, M.; Andrews, A. B.; Mitra-Kirtley, S.; Mullins, O. C. Asphaltene molecular size by fluorescence correlation spectroscopy. Energy Fuels 2007, 21, 2875−2882. (13) Tanaka, R.; Sato, E.; Hunt, J. E.; Winans, R. E.; Sato, S.; Takanohashi, T. Characterization of asphaltene aggregates using X-ray diffraction and small-angle X-ray scattering. Energy Fuels 2004, 18, 1118−1125. (14) Wu, Q.; Pomerantz, A. E.; Mullins, O. C.; Zare, R. N. Laserbased mass spectrometric determination of aggregation numbers for I


Langmuir

Article

(36) Gonzalez, G.; Middea, A. Peptization of asphaltene by various oil soluble amphiphiles. Colloids Surf. 1991, 52, 207−217. (37) Pereira, J. C.; Delgado-Linares, J.; Briones, A.; Guevara, M.; Scorzza, C.; Salager, J. L. The effect of solvent nature and dispersant performance on asphaltene precipitation from diluted solutions of instable crude oil. Pet. Sci. Technol. 2011, 29, 2432−2440. (38) Hernandez-Trujillo, J.; Martinez-Magadan, J. M.; Garcia-Cruz, I. Molecular characterization of p-alkyl phenol-n-heptane interactions and their implication as asphaltene dispersants. Energy Fuels 2007, 21, 1127−1132. (39) Laux, H.; Rahimian, I.; Butz, T. Theoretical and practical approach to the selection of asphaltene dispersing agents. Fuel Process. Technol. 2000, 67, 79−89. (40) Hansen, C. M. The universality of the solubility parameter. Ind. Eng. Chem. Prod. Res. Dev. 1969, 8, 2−11. (41) Annual Book of ASTM Standards, ASTM D2007; American Society for Testing and Materials: Philadelphia, PA, 1991. (42) Goual, L.; Abudu, A. Predicting the adsorption of asphaltenes from their electrical conductivity. Energy Fuels 2010, 24, 469−474. (43) Sharma, A.; Mullins, O. C. Insights into molecular and aggregate structures of asphaltenes using HRTEM. In Asphaltenes, Heavy Oils, and Petroleomics; Mullins, O. C., Sheu, E. Y., Hammami, A., Marshall, A. G., Eds.; Springer: New York, 2007; pp 205−229. (44) Trejo, F.; Ancheyta, J.; Rana, M. S. Structural characterization of asphaltenes obtained from hydroprocessed crude oils by SEM and TEM. Energy Fuels 2009, 23, 429−439. (45) Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. GROMACS 4: Algorithms for highly efficient, load-balanced, and scalable molecular simulation. J. Chem. Theory Comput. 2008, 4, 435−447. (46) Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. Development and testing of the OPLS all-atom force field on conformational energetics and properties of organic liquids. J. Am. Chem. Soc. 1996, 118, 11225−11236. (47) Headen, T. F.; Boek, E. S.; Skipper, N. T. Evidence for asphaltene nanoaggregation in toluene and heptane from molecular dynamics simulations. Energy Fuels 2009, 23, 1220−1229. (48) Sedghi, M.; Goual, L. Effect of asphaltene structure on association and aggregation using molecular dynamics. J. Phys. Chem. B 2013, 117, 5765−5776. (49) Kästner, J. Umbrella sampling. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2011, 1, 932−942. (50) Hub, J. S.; de Groot, B. L.; van der Spoel, D. g_wham−A free weighted histogram analysis implementation including robust error and autocorrelation estimates. J. Chem. Theory Comput. 2010, 6, 3713−3720. (51) Kraiwattanawong, K.; Fogler, H. S.; Gharfeh, S. G.; Singh, P.; Thomason, W. H.; Chavadej, S. Effect of asphaltene dispersants on aggregate size distribution and growth. Energy Fuels 2009, 23, 1575− 1582. (52) Maqbool, T.; Balgoa, A. T.; Fogler, H. S. Revisiting asphaltene precipitation from crude oils: A case of neglected kinetic effects. Energy Fuels 2009, 23, 3681−3686.

J


Asphaltene Aggregation and Impact of Alkylphenols | Langmuir

Recommend Documents