Inhibiting asphaltene precipitation from Iranian crude oil using various dispersants: Experimental investigation through viscometry and thermodynamic modelling

Inhibiting asphaltene precipitation from Iranian crude oil using various dispersants: Experimental investigation through viscometry and thermodynamic modelling

Fluid Phase Equilibria 442 (2017) 104e118 Contents lists available at ScienceDirect Fluid Phase Equilibria j o u r n a l h o m e p a g e : w w w . e...

2MB Sizes 3 Downloads 38 Views

Fluid Phase Equilibria 442 (2017) 104e118

Contents lists available at ScienceDirect

Fluid Phase Equilibria j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / fl u i d

Inhibiting asphaltene precipitation from Iranian crude oil using various dispersants: Experimental investigation through viscometry and thermodynamic modelling Amir Hossein Saeedi Dehaghani a, *, Mohammad Hasan Badizad b a b

Department of Petroleum Engineering, Faculty of Chemical Engineering, Tarbiat Modares University, Tehran, Iran Department of Chemical and Petroleum Engineering, Sharif University of Technology, Tehran, Iran

a r t i c l e i n f o

a b s t r a c t

Article history: Received 16 November 2016 Received in revised form 22 March 2017 Accepted 24 March 2017 Available online 29 March 2017

Asphaltene precipitation is a major assurance problem posing significant technical and economic loss on petroleum industry. To tackle this issue, various treatments have been proposed and applied by industry. Amongst, inhibiting or retarding the asphaltene precipitation has been understood as the most efficient approach. In this regard, blending crude oil with chemical additives could appreciably heighten its stability. Surfactants, owing to amphiphilic nature, could keep asphaltene dissolved in crude oil by precluding self-tendency of those particles to making agglomerates. Despite importance of this subject, there is still lack of sufficient experimental data to evaluate effectiveness of different inhibitors for oils of various regions, in particular Iran. Therefore, our research group made effort to investigate performance of different oil soluble additives for enhancing stability of asphaltene in Iranian crude oil through measuring onset of asphaltene precipitation using viscometry method. This study presents complementary data to our previous works as well as providing a deep discussion to compare influence of toluene, linear and branched dodecyl benzene sulfonic acid (DBSA) and cocamide diethanolamine (CDEA). It was realized that inhibition characteristic is driven by the synergy between strength of polar head and tail length of chemicals, as well as inhibitor concentration in the oil phase. In this respect, this work suggests the following priority order for selecting inhibitors: LinearDBSA > CDEA > Branched-DBSA > Toluene. In the second part of this work, the measured onset points were predicted through coupling PengRobison (PR) with Dehaghani association equation of state (DA-EOS). The applied model gave successful prediction via taking asphaltene self-association and asphaltene-inhibitor cross association into account. © 2017 Elsevier B.V. All rights reserved.

Keywords: Asphaltene Precipitation Inhibitor Viscometry Amphiphile Thermodynamic modeling

1. Introduction Asphaltene precipitation is a major disaster occurring almost in all parts of oil industry, from reservoir to refinery unit [1]. It has been well recognized that asphaltene deposition is prone to plug rock's pore throats, resulting in formation damage and subsequently losing well deliverability [2]. Also, deposition in drill string or well head facilities could lead to technical and economic issues and even, in worst case, unexpected shutdown [3]. In past decades, researchers and industrial specialists have made attempts to devise efficient preventive treatments for asphaltene precipitation [4].

* Corresponding author. E-mail address: [email protected] (A.H. Saeedi Dehaghani). http://dx.doi.org/10.1016/j.fluid.2017.03.020 0378-3812/© 2017 Elsevier B.V. All rights reserved.

Amongst, using inhibitors has been proved to be the best remedy for avoiding or postponing asphaltene drop out [5]. Generally, asphaltene cannot be defined precisely [6]. In fact, asphaltene is classified as the heaviest fraction of crude oil possessing polynuclear (condensed) aromatic rings with alkyl (aliphatic) side chains and polar functional groups [7]. From a practical point of view, the term asphaltene is defined as that portion of a crude oil being precipitated once diluting with normal alkanes (mostly n-pentane or n-heptane); while aromatic solvents such as toluene and xylene improve its stability [8]. In accord with this operational definition, traditionally, asphaltene content is reported as amount of hydrocarbon precipitation after mixing a given oil sample with normal alkane, usually n-heptane [9]. To date, there is no consensus over microscopic distinction of

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

asphaltene molecules, yet [6]. No universal criterion has ever offered to distinguish asphaltene molecules individually [10]. However, as macroscopic behavior of a given crude oil is in directly associated with state of its constituting species, so one needs at least a conceptual model to relate asphaltene stability to its interaction with other species in hydrocarbon media [11]. In this way, colloidal theory has found great attention [12]. According to this theory, asphaltene is described as heavy organic components which are likely to agglomerate through their lyophobic (polar) sites [12,13]. In this sense, asphaltene are naturally distributed as small agglomerates, analogous to what happens in aqueous colloidal solutions [14]. Further, to maintain their stability, resin plays the role of a stabilizer adsorbing on periphery of asphaltenes and impeding their further self-association [15]. Upon this theory, any process disturbing (or weakening) resin task might lead to asphaltene precipitation. By contrast, adding resin-liked compounds to crude oil, could improve asphaltene stability, which is the focus of this paper. Till now, various methods have been proposed to measure onset of asphaltene precipitation (in short OAP) [10]. Escobedo and Mansoori devised a viscometry technique being able to identify OAP for both types of light and heavy (black) crude oils [16]. In their approach, an oil sample is gradually blended with a precipitant, e.g., n-heptane, while simultaneously measuring its viscosity. According to their philosophy, asphaltene flocculation is associated with producing big agglomerates (flocs) which renders oil a sharp increase in viscosity. In this sense, one could simply ascertain OAP as the first point of deviation (sharp break) in the trend of oil viscosity versus solvent concentration added to the crude. This method is advantageously simple and could be easily utilized for different kinds of crude oil [17]. Shadman et al. applied viscometry to investigate inhibition of asphaltene-induced precipitation by different amphiphiles, including branched- and linear-dodecyl benzene sulfonic acid (DBSA) [18]. They pointed out asphaltene stability is directly proportional to amphiphile concentration mixed with crude oil. Most importantly, it was realized that specification and characteristics of the given oil, for example amount of different fractions in particular resin, and also chemical structure of amphiphile, control effectiveness of chemical additives. In past years, many studies were devoted to establishing an interrelationship between oil and amphiphile properties with inhibition efficiency based on measurement of onset point [13,19e21]. Leon et al. studied adsorption of alkylbenzene-derived amphphiles on asphaltene particles [22]. They suggested a two-stage process for this phenomenon. In their observations, first, additive particles adsorb individually on asphaltene surface. Then, at the second step, amphiphiles self-interaction leads to further accumulation of peptizing agents at the asphaltene covered area. They came to the conclusion that inhibition strength largely depends on maximum amount of stabilizer adsorbed on asphaltene aggregates. In another study, to improve stability of Brazilian crude oil, Gonzalez and Middea assayed peptizing ability of alkyl-benzene derived materials with focus on functionality of hydroxyl and amide polar groups [23]. In this case, nonyl phenol was observed to be the best dispersing agent. They attributed asphaltene dispersion to direct interaction between polar head of amphiphiles and polar sites located on asphaltene molecule. Besides this direct conclusion, they inferred asphaltene stabilization to be the result of amphiphile association with their molecules. In this scheme, aromatic moieties of asphaltene could act as electron donors to neutralize polar head of chemical additives [23]. In a comprehensive study, Al-Sahhaf et al. investigated inhibition efficiency of deasphalted oil (DO), extracted resin, toluene and three types of surfactant while adding to Kuwaiti crude oil [19]. In their study, DO and toluene proved to be inefficient needing

105

extraordinary concentration to give appreciable inhibition effect. By contrast, owing to their polar head acidity, surfactants retarded OAP at lower concentration. It was recognized that, for Kuwaiti crude oil, peptizing strength is directly associated with acidity (nature and number of functional groups) of the amphiphile used. In a similar investigation on Kuwaiti oil, Ghloum et al. utilized three types of inhibitors (both commercial and non-commercial) and came to similar conclusions [24]. Through measuring zeta potential and electrophoretic mobility of two kinds of Chinese residue oil, Wang et al. observed vital role of the asphaltene electric property [25]. To obtain optimum inhibition performance in terms of amphiphile-asphaltene interaction, they asserted negatively charged asphaltenes should be peptized with cationic inhibitors and vice versa. This would partly neutralize asphaltene polarity, which makes it more compatible with nonpolar media of hydrocarbon phase [25]. Junior et al. made use of commercial surfactants as well as vegetable oils and their derivatives to increase onset of asphaltene flocculation in Brazilin oil while being diluted with aliphatic solvents (n-heptane and n-pentane) [21]. They pointed out stabilizing capacity of an amphiphile is controlled by balance between polarity (lyophobic) and hydrocarbon chain length. In other words, having sole strong polar heads does not essentially correlate with amphiphile strength to disperse asphaltene particles. For example, as observed by Junior et al. salicylic acid even resulted in asphaltene destabilization [21]. Until now, different experimental studies have been conducted in an attempt to establish a generic correlation for interrelashionship between characteristics of asphaltene and inhibitor, and also composition of a given oil sample. In spite of discussed works, there is a huge gap to thoroughly identify effect of chemical inhibitors on asphaltene stabilization, yet. After an extensive literature survey, authors realized a conspicuous paucity of released information on the asphaltene stabilization of Iranian crude oils using chemical inhibitors. In this regard, our research group is investigating interaction between different oil soluble inhibitors and Iranian crude oil. This work presents complementary data to out earlier works [18,26e28] along with deep discussion on effectiveness of various chemicals (toluene, linear and branched- DBSA, and CDEA) for peptizing asphaltenes in Iranian crude oil. In remainder of this paper, first, we explain the experimental technique of our work and then, the measured onset points will be compared based on characteristics of additives. 2. Experiment 2.1. Materials Two dead oil samples, hereafter called Oil I and Oil II were used to analysis their asphaltene stability. Tables 1 and 2 present identifications and composition of the oil samples, respectively. Also, Fig. 1 shows viscosity and density of the crude oils at varying temperatures. To verify our measurements, asphaltene content was obtained through SARA test and IP 143 standard method; both

Table 1 SARA (saturate, aromatic, resin, and asphaltene) fraction of oil samples used in experiments. Crude Oil Density (API) SARA test

IP143

Saturates Aromatics Resins Asphaltene Asphaltene Oil I Oil II

17.96 17.62

30.79 46.63

42.10 34.30

13.36 4.77

13.75 14.30

13.75 14.30

106

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

Table 3 presents the chemical structure of stabilizers used in this study.

Table 2 Composition (mole%) of oil samples. Oil I

Oil II

Component

Mole %

Component

Mole %

C7 C8 C9 C10 C11 C12þ

4.31 20.85 16.87 13.89 10.02 34.06 e

C6 C7 C8 C9 C10 C11 C12þ

9.80 6.97 7.82 5.66 5.65 5.10 59.00

e

2.2. Method At the beginning, a known amount of inhibitor (1000, 2000, 10,000, or 20,000 ppm), was mixed with an oil sample and then, the mixture was stirred rigorously for 2 h at 50  C in a closed container to attain a homogeneous mixture. The mixing was conducted at higher temperature (50  C) compared to that of measuring viscosity (20  C) to ensure complete dissolution of added inhibitors. At this point, the oleic mixture (oil þ inhibitor) was diluted in various ratios with n-heptane and then immediately was injected to the viscometer to measure its viscosity. It should be emphasized that all measurements were taken at atmospheric pressure. A Stabinger Viscometer Anton Parr SVM3000 was employed to measure dynamic viscosity in the range of 0.2e20,000 (mPa s) with reproducibility of 0.35%, together with specific gravity in the range of (0.65e3) gr/cm3 with precision 0.0002 gr/cm3. It should be noted that all measurements were run at constant temperature 20  C. 3. Results and discussion

Fig. 1. Viscosity and density of oil samples measured by viscometer at varying temperatures and atmospheric pressure (note some of data were also presented in our previous work [27]).

indicating equal amounts, as presented in Table 1. Both crudes are of same specific gravity. Normal heptane, with purity >99% purchased from MERCK company, was utilized as precipitating agent. To select chemical inhibitors, several points were factored in, such as: being environmental friendly, commercial availability and compatibility with oil fractions. In this respect, toluene (a conventional aromatic solvent) as well as three organic acids, linear and branched dodecyl benzene sulfonic acid (in short linear- and branched-DBSA, respectively) and Cocamide diethanolamine (CDEA), were employed as asphaltene precipitation inhibitors. All chemicals were used as received without further purification.

Apart from different SARA fraction and composition of oil samples reported in Tables 1 and 2, we shall compare their bulk properties while responding to varying temperature. As shown in Fig. 1, both crudes exhibit linear density reduction once increasing temperature, whereas their viscosities display an exponential decreasing trend. As seen, temperature could markedly influence on viscosity (~1300 cp viscosity reduction just by increasing Oil II's temperature from 20 to 60  C). On the other hand, owing to higher saturate contents, Oil II is more sensitive to heating (steep viscosity reduction versus temperature). To evaluate asphaltene stability, diverse measures have been proposed. Generally, amount of asphaltene is the prime indicator to judge over stability of an oil sample. However, it is well recognized that there is no such direct proportionality [29]. In other words, the asphaltene stability is mostly controlled by molecular interactions in oil and almost always light oils (with lower asphaltene content) are more likely to precipitate asphaltene [6]. As mentioned earlier, according to colloidal theory, asphaltene is regarded as

Table 3 Specifications of chemical additives used as stabilizing agents. Chemical name

Molecular structure

n-heptane

Molecular formula

Supplier

Purity (%weight)a

C7H16

Merck

>99%

C6H5CH3

Merck

>99.9%

C18H30O3S

Behdash Chemical Company

>96%

C18H30O3S

Behdash Chemical Company

>96%

e

Behdash Chemical Company

>90%

Toluene

Linear DBSA

Branched DBSA

CDEA

a

As reported by the supplier.

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

107

agglomerating species in hydrocarbon media which is stabilized by resin particles adsorbed on its surface [13]. In this sense, resins surround an asphaltene agglomerate and at the same time, stretching out their lipophilic hydrocarbon tails. Naturally, asphaltene is inclined to precipitation and resins preclude this process by acting as natural stabilizers in crude oil. From this viewpoint, researchers have tried to justify effectiveness of amphiphilic stabilizers analogous to resin action [30], which will be discussed later. As a result, asphaltene-resin (AR) ratio is a sensible measure for asphaltene stability in a given oil sample [31]. However, this measure does not include relative amount and contribution of other oil constituents, i.e. saturates and aromatics, and might give misleading prediction [32]. The colloidal instability index (CII) is another measure suggested to quantize asphaltene stability in crude oil, defined by [31]:

Asphaltene þ Saturate CII ¼ Aromatic þ Resin

(1)

With the denominator indicates those SARA fractions playing as peptizer and the numerator includes total amount of asphaltene plus saturate fraction which acts as flocculant. Table 4 compares oil samples used in this study based on different measures for their asphaltene stability. As reported, Oil II is more likely to form asphaltene precipitation compared to Oil I, due to higher asphaltene/resin ratio, and CII. As stated by Asomaning, CII greater than 0.9 and below 0.7 shows oil with unstable and stable asphaltene, respectively [31]. Therefore, Oil II should be regarded as potentially unstable while Oil I's stability is indeterminate (CII lies between 0.7 and 0.9). Further, based on empirical evidence, Asomaning suggested a cutoff value of 0.35 for asphaltene stability based on AR ratio. In this respect, both Oil I and II, with AR ratios of 1 and 2, respectively, should be regarded unstable crudes, while latter demonstrates more instability tendency. Figs. 12 and 3 viscosity variation of oil samples while diluting with n-heptane in the absence of any inhibitor. It should be emphasized that viscosities were measured with reproducibility ±0.35%. The viscosity measurement technique employed in our study indicates precipitation point. However, some other methods, such as measuring IFT reduction, indicates flocculation point [33], when larger agglomeration takes place in oil phase which could be detected even by microscopic observation. In this view, each technique specifies a different step of asphaltene separation process. To detect onset point, one should seek for the first hump in the viscosity diagram. Of course, in contrast to figures presented in present study, it would be a simple task to discern even small humps thanks to available modern spreadsheets. The onset is ascribed to the point at the start of humping or, in other words, unusual ascending trend in viscosity curves. As shown in Figs. 2 and 3, the black arrows indicate onset point of Oil I and II in terms of lowest n-heptane concentration inducing drastic fluctuation in viscosity curve, respectively, 12.4 and 9.2 vol%. This observation is in agreement with oils specification reported in Tables 1 and 4 because Oil II with higher CII, needs lower amount of n-heptane compared to Oil I for asphaltene precipitation. Fig. 4 through 10 specify OAP of oil samples containing different

Table 4 Stability parameters for oil samples used in this work (CII was defined through Eq. (1)). Oil Sample

Asphaltene content (wt.%)

Asphatene/Resin

CII

OI OII

13.75 14.30

1.0 3.0

0.8 1.5

Fig. 2. Onset of asphaltene flocculation for oil I in absence of inhibitor, indicated by arrow (These data were also presented in our previous work [18]).

Fig. 3. Onset of asphaltene flocculation for oil II in absence of inhibitor, indicated by arrow.

inhibitors. In fact, first point at sharp deviation from the smooth trend of viscosity was considered as the incipient of asphaltene precipitation in the oil sample, as indicated by an arrow. As mentioned earlier, adding n-heptane could trigger asphaltene precipitation and this directly influences on the viscosity of oil sample. As seen in figures, one could simply detect onset point by means of viscometry, which signifies a practical advantage of this method. In following discussion, n-heptane concentration at OAP is a measure to judge over effectiveness of different chemicals. In this regard, higher amount of n-heptane at onset point corresponds to higher inhibition strength of an additive (toluene or amphiphile). Before proceeding further, it should be pointed out that in some figures, for instance Fig. 4(a), one could notice an erratic behavior (ascending and descending fluctuations) in the viscosity curves beyond the onset point. In fact, onset point means formation of asphaltene agglomerates in the bulk of oil phase and such ordered structures increase viscosity. Recall that our measurements were carried out via a rotary instrument which imposes shear force on a given sample. This shear stress could break asphaltene aggregates apart and bring about re-dispersion (or dissociation) of precipitated asphaltenes. Consequently, due to shear effect, oil viscosity will be reduced. This periodic process (induced agglomeration-shear dispersion) gives rise to alternating humps in the viscosity diagram. In what follows, the effect of each inhibitor will be discussed in detail.

108

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

Fig. 4. Determining onset of asphaltene flocculation, indicated by arrow, by detecting sharp deviation in oil I viscosity versus n-heptane vol% used for dilution at varying toluene concentrations of: (a) 1000 ppm; (b) 2000 ppm; (c)10,000 ppm; and (d) 20,000 ppm (Figure “a” was also presented in our previous work [18]).

3.1. Toluene Fig. 4 show viscosity variation in presence of toluene. Also, Fig. 11 compares inhibition strength of toluene against other inhibitor (all of amphiphilic nature) at varying concentrations. Notice, as shown through error bars in Fig. 11, the onset points were measured with uncertainty 0.2 vol%. As seen, among different inhibitors used in this study, toluene was of lowest inhibition strength with OAP up to 16.2 vol% of n-heptane at 20,000 ppm. Although toluene is a conventional solvent for asphaltene, but, it lacks any polar part to make strong interaction with asphaltene particles. In accord with this study, Taher et al. made similar observation for Kuwaiti oil while inhibiting with both toluene and deasphaltened oil. They pointed out toluene inhibition effect becomes significant once diluting with at least 50 mass% [19]! Certainly, such extraordinary concentration is too much to be used in practical applications. Let imagine one continued our experiment for much higher concentrations of toluene (>20,000 ppm), thus, we would possibly observe pronounced asphaltene stabilization at toluene concentrations over 500,000 ppm! In fact, at such concentration, toluene re-dissolves precipitated asphaltenes. For toluene, asphaltene stabilization, indeed, is due to sole effect of aromaticity which allows toluene molecules to disturb asphaltene entanglement process through interfering into p  p interactions

between condensed aromatic rings of overlapped sheets of asphaltene aggregates [1]. In this sense, toluene action radically differs from aphiphiles' functionality, discussed in next subsections. Generally, role of aromaticity is significantly weaker than acid-base interaction and it is directly proportional to amount of toluene used for dilution. A seen in Fig. 11, toluene exhibits highest inhibition effect at 20,000 ppm. Simply put, toluene inhibitory relies mainly on its concentration used for dilution. From economic point of view, a useful additive must necessarily improve asphaltene stability at lowest possible concentration, and accordingly, toluene fails this task.

3.2. DBSA As shown in Figs. 11 and 12, DBSA could effectively improve asphaltene stabilization, up to n-heptane 21.4 and 22.2 vol% necessary to induce precipitation in oil I and II, respectively. Sulfonyl hydroxide (SO2-OH) gives strong acidity to DBSA such that being irreversibly attached on asphaltene cores [21,25,34]. Notably, as seen in Figs. 11 and 12, linear and branched form of DBSA are of distinct inhibition performance rendering different onset point for asphaltene precipitation. According to colloidal theory, asphaltene should be regarded as polar particles dispersed in a non-polar media mostly made up aliphatic components [13,35]. Hence, due to lyophobic nature, asphaltenes are prone to agglomerate via self-

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

109

Fig. 5. Determining onset of asphaltene flocculation, indicated by arrow, by detecting sharp deviation in oil I viscosity versus n-heptane vol% used for dilution at varying linearDBSA concentrations of: (a) 1000 ppm; (b) 2000 ppm; (c)10,000 ppm; and (d) 20,000 ppm (Figure “a” was also presented in our previous work [18]).

association. In this event, analogous to resin action, amphiphile (herein DBSA) could enhance asphaltene stability by forming steric-stabilized layer around the asphaltene micelle [20,27]. In this fashion, amphiphile is attached on the asphaltene micelle through interactions (some authors use term acid-base interaction) between their head-groups with polar functionalities of asphaltene [25]. Amphiphile's alkyl chain (as straight and/or branched) will contact with non-polar species of hydrocarbon phase, maintaining asphaltene stability by hindering further association through neutralizing attractive interactions. Leon et al. pointed out that amphiphile effectiveness is in direct relation to maximum amount of chemical adsorbed on the asphaltene core [20,22]. In this sense, an efficient inhibitor should necessary make a shield layer around asphaltene particles by covering all active (polar) parts of asphaltene molecules. As seen in Figs. 11 and 12, branched DBAS is constantly of lower inhibition effect (according to lower onset point) compared to linear one. Additionally, this difference becomes greater at higher inhibitor concentration. Branching reduces length of hydrocarbon chain, that is, lipophilic portion of DBSA molecule, weakening hydrocarbon-affinity of surfactant molecule [36]. Therefore, branched DBAS forms weaker shield around asphaltene particles than the linear one. In this regard, Chang and Fogler asserted that balanced strength of polar head and alkyl chain governs inhibiting

performance. They observed insignificant inhibition power for chemicals with short alkyl chain. Further, they suggested only amphiphiles with tail composed of carbon number larger than six could make an appreciable contribution to asphaltene peptization [37,38]. Besides rendering lower dispersivity in non-polar media, as stated by Shadman et al. branching brings about a steric hindrance effect which impedes adsorption of further amphiphile molecules on asphaltene surface owing to methyl interference of alkyl chains, leaving some polar sites of asphaltene uncovered [27]. Once being adsorbed, DBSA stands perpendicularly on the asphaltene surface [22], thus, steric hindrance of methyl groups acts as the major cause of reducing coverage of asphaltene particles by further DBSA molecules. As a result, most of branched DBSA remains in monomeric state (individual molecules) through the bulk oil phase. As branched-DBSA concentration increases, role of branching becomes more pronounced, as seen in Figs. 11 and 12, at 10,000 and 20,000 ppm. For instance, Oil I containing branched DBSA at 1000 ppm made asphaltene precipitation at n-heptane 0.2 vol% sooner than the case diluted with linear DBSA. But, at highest amphiphile concentration, 20,000 ppm, the difference of onset points reached 2.4 vol%, indicating larger effect of branching at higher DBSA concentration. Based on this observation, once could infer that beyond an optimum concentration, branched DBSA could

110

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

Fig. 6. Determining onset of asphaltene flocculation, indicated by arrow, by detecting sharp deviation in oil I viscosity versus n-heptane vol% used for dilution at varying branchedDBSA concentrations of: (a) 1000 ppm; (b) 2000 ppm; (c)10,000 ppm; and (d) 20,000 ppm (Figure “a” was also presented in our previous work [18]).

not be adsorbed on asphaltene core surface while there is still some polar sites uncovered by amphiphile molecules. This argument is in agreement with observation made by previous investigations [18,27,37,38]. In this respect, Leon et al. suggested a two-step process for amphiphile adsorption on asphaltene surface [20,22]. In this scheme, first, chemical additives adsorb individually on the asphaltene core, and thereafter, interaction between adsorbed amphiphile controls extend of adsorption process to form a hemimicelle around the asphaltene. 3.3. CDEA CDEA possesses two hydroxyl head-groups, as presented in Table 3, to interact with asphaltene in crude oil. Previous studies demonstrate lower inhibition strength of nonylphenol (having a single OH group) compared to DBSA [19,24]. In fact, as pointed out by Al-Sahhaf et al., sulfonyl hydroxide (SO2-OH) is of stronger acidity than hydroxyl (OH) and could form stronger attachment on asphaltene surface [19]. Despite this observation, it should be emphasized that CDEA possesses two hydroxyl groups and this double functionality renders stronger inhibitory strength of CDEA in comparison to DBSA. As shown in Figs. 11 and 12, CDEA is the strongest asphaltene inhibitor for both oil samples studied in this work up to concentration 10,000 ppm. However, beyond this concentration, CDEA loses its strength and linear DBSA takes over it. Interestingly, CDEA exhibits equal inhibition effect (regarding equal n-heptane concentration at onset point) at both 10,000 and 20,000 ppm, as seen in Figs. 11 and 12. In this respect, one might

infer that CDEA could not maintain its effectiveness at concentrations above 10,000 ppm. By contrast, DBSA and particularly linear one, still tends to enhance onset point. It is not a fortuitous observation! This apparent stationary state could be ascribed to the amount of CDEA adsorbed on asphaltene. On the one hand, it has been recognized that phenol-derived surfactants tend to lie parallel on the asphaltene surface making a dense pack of monolayers [20,22,36]. On the other hand, recall that strong polar interactions through double hydroxyl group accelerate coverage of polar sites on asphaltene by CDEA molecules. In conclusion, beyond a socalled saturation point (herein 10,000 ppm), adding more CDEA into the oil promotes a thick layer of monomeric amphiphiles adsorbed axially on the asphaltene surface. In this circumstance, adsorbed molecules of CDEA cannot stretch their alkyl chains and the protective layer around asphaltene micelle reaches its maximum lipophilic affinity. Besides, to make a vigorous steric stable layer around asphaltene, CDEA necessarily must satisfy both its hydroxyl groups through attachment on asphaltene micelles, otherwise, the unadsorbed hydroxyls is likely to preserve micelle polarity to some extent by acting similar to the original polar sites on asphaltene molecule. In this way, there is a parallelisms between steric hindrance of alkyl chains in branched DBSA and double OH groups of CDEA. According to foregoing discussion, regarding molecular structure of amphiphile, there is an optimum concentration beyond which no appreciable inhibitory effect could take place. For instance, in this work, branched DBSA and CDEA exhibit their

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

111

Fig. 7. Determining onset of asphaltene flocculation, indicated by arrow, by detecting sharp deviation in oil I viscosity versus n-heptane vol% used for dilution at varying CDEA concentrations of: (a) 1000 ppm; (b) 2000 ppm; (c)10,000 ppm; and (d) 20,000 ppm (Figure “a” was also presented in our previous work [18]).

maximum strength at 10,000 ppm. It should be emphasized that typical range of additives concentration studied in all previous studies is 500e3000 ppm, thus, 20,000 ppm might seem rather inordinate. It should be pointed out that for some practical purposes, particularly in upstream part of petroleum industry, engineers must inevitably utilize extraordinary concentration of inhibitors to ensure their ultimate effectiveness. For example, to avoid asphaltene precipitation during solvent injection into hydrocarbon reservoirs, making use of peptizing agents is a matter of choice. But, due to dispersion and adsorption taking place in porous structure of underground sediments, amphiphile is subject to severe dilution, losing its effectiveness while propagating throughout reservoir [39]. At this point, we shall examine how effectively applied dispersants could retard asphaltene precipitation of Iranian crude oil. In this respect, one should compare this work with similar studies conducted on Middle-East crude. It should be mention again that the onset point is described by smallest n-C7 concentration inducing asphaltene precipitation in presence of an inhibitor. Indeed, by comparing effectiveness of an inhibitor for different crudes, one could realize that the lower inhibitor concentration at a given onset point, i.e., equal amount of n-C7, implies higher strength of that dispersant to retard asphaltene precipitation. Based on this criterion, the onsets measured in our work are comparable to results presented in literature [19]. For instance, Al-sahhaf et al. pointed out onset of Kuwait crude at ~13 cm3/gr oil n-C7 in presence of DBSA with maximum concentration 0.1 mass fraction, i.e., 100,000 ppm. Certainly, DBSA is quite effective for maintaining stability of our oil samples and in general middle east heavy crude.

Also, it should be emphasized that performance of a dispersant strongly depends on the particular oil sample under consideration. As a point of novelty, this study made attempts to provide insight into variation of inhibitors effectiveness in a broad range of concentrations. In most previous studies, most researchers have narrowly focused on asphaltene stabilization at constant concentration of inhibitor. In this manner, they proposed general orders for inhibition strength of different amphiphiles. For example, AlSahhaf suggested DBSA > nonylphenol > resin > toluene [19,37]. Nevertheless, our experiments carried out at varying concentrations, contradicts such general trend. In other words, the effectiveness of any amphiphile is in relation to its adsorption capacity on asphaltene micelle which is directly proportional to amount of chemical added to the crude oil. For practical purposes, this work suggests evaluating inhibition of different additives at varying concentrations before making hasty decisions just upon acidity of head-groups. In addition to commercial inhibitors, some researchers have made use of special dispersants to render stability of asphaltene. For instances, Firoozabadi's group applied proprietary nonionic dispersants to investigate the kinetic of asphaltene aggregation in presence of that inhibitors with probing time taken to observe onset point via dynamic light scattering and electrophoretic mobility [40e43]. They concluded that electrostatic repulsion could stabilize asphaltene against aggregation as sufficient as steric stabilization. Also, they proposed a molecular model for inhibition process of DBSA in which dispersant contributes to protonation of heteroatomic moieties of asphaltene to completely cover all polar parts of asphaltene.

112

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

Fig. 8. Determining onset of asphaltene flocculation, indicated by arrow, by detecting sharp deviation in oil II viscosity versus n-heptane vol% used for dilution at varying linearDBSA concentrations of: (a) 1000 ppm; (b) 2000 ppm; (c)10,000 ppm; and (d) 20,000 ppm.

To conclude this part, we shall highlight accuracy of our measurements compared to the earlier but successful works conducted by other researchers. In their original work, Escobedo and Mansoori made use of a modified Ostwald viscometer (glass viscometer) to measure oil viscosity during dilution [16]. This kind of viscometer is a simple, old instrument measuring viscosity via recording the time takes a given fluid to pass a capillary tube driven by the gravity force. Despite being quite convenient, there is no control on the shear experienced by the sample. In spite of this potential drawback, Escobedo and Mansoori could accurately obtain the asphaltene precipitation onset using their simple instrument. Based on this argument, certainly, one could expect obtaining much more accurate data while employing Viscometer SVM™ 3000 which is an accurate for device measuring dynamic viscosity. Besides being convenient, viscometry allows obtaining asphaltene precipitation point which is equivalent to its solubility in maltene. By contrast, some techniques such as IFT measurement, determine flocculation point which occurs after precipitation. Through viscometry, advantageously, one could simultaneously obtain solubility data and use them for thermodynamic modelling which is the subject of next part. 3.4. Thermodynamic modeling To design a field scale project, an engineer needs using

appropriate simulators and modelling approaches to evaluate effect of different parameters on inhibitory strength of a given dispersant at different thermodynamic conditions. For example, recently, Nasrabadi et al. studied asphaltene precipitation induced by CO2 injection in reservoir. For the first time, they developed a unified compositional reservoir model which couples three-phase flash calculation with fluid flow equations [44]. Certainly, to expand such studies, one needs a knowledge of capability of different thermodynamic models to represent particular behavior of asphaltene in presence of dispersants. At this stage, a new thermodynamic model has been employed to give prediction for the measured onsets. In this study, finding the onset point corresponds to obtaining minimum amount n-heptane which leads to asphaltene precipitation. To this end, we followed the traditional procedure by first establishing two-phase VLE equilibrium and then calculating amount of precipitate through LLE calculation [45]. The VLE calculation was carried out to obtain composition of saturated liquid (at bubble point). In this respect, Peng-Robinson equation of state (PR-EOS), detailed in Appendix, was applied to calculate fugacity coefficient of components. Afterwards, by supposing no associative component (asphaltene and inhibitor) in the vapor phase, the LLE calculation was performed to determine amount of asphaltene precipitated. In this manner, another liquid phase, namely, l2 was assumed to be composed of precipitated asphaltene along with adsorbed amphiphiles.

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

113

Fig. 9. Determining onset of asphaltene flocculation, indicated by arrow, by detecting sharp deviation in oil II viscosity versus n-heptane vol% used for dilution at varying branchedDBSA concentrations of: (a) 1000 ppm; (b) 2000 ppm; (c)10,000 ppm; and (d) 20,000 ppm.

Therefore, for LLE calculation, one necessarily should merely consider two equilibrium constants (K) for pseudo-components, that is, asphaltene and inhibitor. Mathematically [33]:

Ki ¼

xli2

(2)

xli1

plus two additional components (asphaltene and inhibitor). Due to associative nature of asphaltene precipitation and inhibition, a proper model should be able to handle such interactions. The non-ideality pertaining to such mixture is due to the chemical bond between associated compounds and physical interactions owing to van der Waals forces [46]. As a result, as stated by Lambert, the second virial coefficient should be splitted as [47]:

Based on material balance, one will obtain:

zi ¼ l2 xli2 þ l1 xli1

(3)

Here zi represents total mole fraction of component i in the liquid phase. Combining preceding expression results in the following relations:

xli1 ¼

zi l1 þ ð1  l1 ÞKi

(4)

xli2 ¼

zi Ki l1 þ ð1  l1 ÞKi

(5) PNþ2

l1 i¼1 xi

PNþ2

l2 i¼1 xi

According to identities ¼ 1 and ¼ 1, the working equation analogous to Rachford-Rice expression was derived:

f ðl1 Þ ¼

Nþ2 X i¼1

zi ðKi  1Þ ¼0 l1 þ ð1  l1 ÞKi

(6)

Bearing in mind that summation runs over liquid composition

B ¼ Bche þ Bphy

(7)

where superscripts che and phy account for chemical and physical contribution to non-ideality of a given associated fluid, respectively. According to definition of compressibility factor, the preceding equation takes the following form [48]:

Z ¼1þ



Bche þ Bphy ; V



Bche V

(8.a)

! þ



Bphy V

! 1

(8.b)

In which first parenthesis represent chemical part of the compressibility factor and the second one accounts for physical effect which normally is calculated a cubic EOS. Up to now, numerous expressions have been suggested to incorporate effect of association into phase equilibrium calculation. Generally, they fall in three categories: (1) chemical theory, (2) perturbation theory, and (3) lattice theory [49]. The chemical AEOS

114

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

Fig. 10. Determining onset of asphaltene flocculation, indicated by arrow, by detecting sharp deviation in oil II viscosity versus n-heptane vol% used for dilution at varying CDEA concentrations of: (a) 1000 ppm; (b) 2000 ppm; (c)10,000 ppm; and (d) 20,000 ppm.

Fig. 11. Comparing onset of asphaltene flocculation for oil I versus concentration of various inhibitors (data was also presented erroneously in our previous work [18]).

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

115

Fig. 12. Comparing onset of asphaltene flocculation for oil II versus concentration of various inhibitors.

considers association as hypothetical reactions taking place between similar and dissimilar components leading to non-ideality of a fluid phase. Recently, Dehaghani et al. proposed a new AEOS, called DA-EOS, which has demonstrated to give accurate prediction for the coprecipitation of asphaltene and resin [33]. Also, Dehaghani and Badizad pointed out rigorous prediction of DA-EOS for inhibiting hydrate formation by alcohols [50]. On contrary to previous equations, DA-EOS was developed based on taking finite extend of association between components, given by [51]:

Z chem ¼

C C þ PK0

(9)

RT v

where C and K represent extend of association and equilibrium constant, respectively; and both depend on the nature of associative components. Table 6 presents parameters used in this study. To obtain inhibitor associating parameter, due to the paucity of data points, we made use of first two data points at each diagram, corresponding to lowest amphiphile concentration to regress the unknown parameters. Then, having all parameters, DA-EOS was used to predict onset point at higher inhibitor concentrations. For mixtures containing n associative components along with m non-associative ones, Eq. (9) could be generalized as follows [51]:

Z ch ¼

n X i¼1

Ci xi

0 @Ci þ 10 P

RT v

Pn j¼1

1þ Kij xj A

m X

xn

(10)

n¼1

Which for our case, takes the following form:

Z ch ¼

CA þ P10

RT v



CA xA þ CI þ P10 KA;A xA;A þ KA;I xI

RT v



C I xI  KI;I xI þ KI;A xA

þ 1  xA  xI (11)

where subscripts A and I stand for asphaltene and inhibitor species, respectively. Fig. 13 compares measured n-heptane vol% at onset points (presented in Table 5), with those predicted by DA-EOS. As seen, the proposed model gives satisfactory estimation with AAD% for linear-, branched-DBSA and CDEA equals to 8.8, 6.6 and 7.8, respectively. It should be noted that by increasing amount of amphiphiles in the crude oil, as shown in Fig. 13, the model tends to deviate from the measured values, which could be attributed to the underlying assumption of applied model. As explained above, in line with almost all chemical association theories, DA-EOS assumes a finite number of hypothetical linear associating reactions accounting for formation of complexes due to hydrogen bonding. Evidently, the actual structure of asphaltenes coated by resin could not fully imagined as such supposed reactions. In fact, asphaltene surrounded by adsorbed molecules of dispersant, is of a three dimensional structure. As a result, at higher inhibitor concentrations, the oleic phase demonstrates to be more colloidal in conflict with assumption of DA-EOS.

4. Conclusions Regarding vital role of inhibiting asphaltene precipitation, present study took effort to unify (and complete) our previous works, providing an insight into stabilizing Iranian crude oil by commercial chemical inhibitors including toluene, linear and branched DBSA, and CDEA. Also, our research group pioneers extending viscometry as a fast and rigorous method for detecting onset of asphaltene precipitation with or without inhibitor. It was realized that to maintain asphaltene stability in Iranian oil, linear DBSA is the best chemical of, regarding benign environmental effects, inhibition strength and commercial availability. Also, It was observed that inhibition strength is controlled by molecular structure of a given amphiphile as well as its concentration in bulk oil phase. Multi-polar functionality or branched alkyl chains could reduce inhibitor effectiveness. We deduced that amphiphile attachment on asphaltene surface is an equilibrium reaction

116

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118 Table 5 Onset points (n-heptane vol%) for oil samples containing various additives at 20  C and atmospheric pressure. Note some of data in this table were already presented erroneously in our previous work [18]. Inhibitor, ppm

Toluene

Linear DBSA

Branched DBAS

CDEA

Oil I

Oil II

Oil I

Oil II

Oil I

Oil II

Oil I

Oil II

1000 2000 10,000 20,000

12.6 12.8 14.2 16.2

e e e e

14.0 14.4 18.4 21.4

11.8 14.0 19.4 22.2

13.8 14.0 18.2 19.0

11.2 13.0 17.2 20.6

14.4 15.0 20.4 20.6

13.0 15.8 21.0 21.6

Standard uncertainties: u(T) ¼ 0.02  C, u(n-heptane vol %) ¼ 0.2 vol %.

Table 6 Parameters of DA-EOS used for predicting onset of asphaltene precipitation. Parameter

Inhibitor Linear-DBSA 1.2 5.3 3.5 1.0 3.8

KA,I KA,Aa KI,I CAa CI a

    

7

10 107 107 106 106

Branched-DBSA 1.5 5.3 3.2 1.0 3.1

    

7

10 107 107 106 106

CDEA 1.1 5.3 2.1 1.0 4.5

    

107 107 107 106 106

Estimated from our previous study [33].

Iranian crude oil, this work suggests the following general priority order for selecting inhibitors: Linear-DBSA > CDEA > BranchedDBSA > Toluene.

Appendix A The Peng-Robinson equation of state (PR-EOS) is expressed as follows:



RT a  v  b vðv þ bÞ þ bðv  bÞ

(A.1)

The exact thermodynamic expression, given below, was used to calculate fugacity coefficient associated with physical interactions:

lnðfi ZÞ ¼

1 RT

ZV  ∞

 RT vP  dV V vni

(A.2)

By some mathematical manipulation, one could obtain following expression: ph

ln fi

Fig. 13. Predicting the n-heptane vol% needed to induce asphaltene precipitation at various inhibitor concentrations: (a) linear DBSA; (b) branched DBSA; and (c) CDEA.

¼

   bi  ph ph ZPR  1  ln ZPR  B bm 2

pffiffiffiffiffiffiffiffi 6 PN A 62 j¼1 xi ai aj 1  dij  pffiffiffi 6 am 2 2B 6 4 3

governed partly by additive concentration in bulk oil phase. Other factors, such as structure of asphaltene and distribution of active polar sites on its surface is the ongoing subject followed by our research group. Last but not least, to retard asphaltene precipitation from

2 ph  pffiffiffi 3 7 ZPR þ 1 þ 2 B bi 7 7ln4   pffiffiffi 5 ph bm 7 5 Z þ 1 2 B PR

(A.3)

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

Nomenclature a b C f K R T v x y Z f

Parameter of PR-EOS Parameter of PR-EOS Association-length Fugacity chemical association constant Universal gas constant Temperature Molar volume Mole fraction in liquid phase Mole fraction in gas phase Compressibility factor Fugacity coefficient

Superscript che Chemical phy Physical o Standard state

[18]

[19]

[20]

[21]

[22]

[23] [24]

[25] [26]

Supplementary data [27]

Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.fluid.2017.03.020.

[28]

References

[29]

[1] S. Ilyin, M. Arinina, M. Polyakova, G. Bondarenko, I. Konstantinov, V. Kulichikhin, A. Malkin, Asphaltenes in heavy crude oil: designation, precipitation, solutions, and effects on viscosity, J. Pet. Sci. Eng. 147 (2016) 211e217. [2] M. Daaou, A. Larbi, B. Martínez-Haya, M. Rogalski, A comparative study of the chemical structure of asphaltenes from algerian petroleum collected at different stages of extraction and processing, J. Pet. Sci. Eng. 138 (2016) 50e56. [3] A.H. Saeedi Dehaghani, M.H. Badizad, Experimental study of iranian heavy crude oil viscosity reduction by diluting with heptane, methanol, toluene, gas condensate and naphtha, Petroleum 2 (2016) 415e424. [4] H. Pan, A. Firoozabadi, Thermodynamic micellization model for asphaltene precipitation inhibition, Aiche J. 46 (2000) 416e426. [5] A. Yen, Y.R. Yin, S. Asomaning, Evaluating Asphaltene Inhibitors: Laboratory Tests and Field Studies. In Spe International Symposium on Oilfield Chemistry, Spe, 2001. [6] E. Buenrostro-Gonzalez, H. Groenzin, C. Lira-Galeana, O.C. Mullins, The overriding chemical principles that define asphaltenes, Energy fuels. 15 (2001) 972e978. [7] T.J. Kaminski, H.S. Fogler, N. Wolf, P. Wattana, A. Mairal, Classification of asphaltenes via fractionation and the effect of heteroatom content on dissolution kinetics, Energy fuels. 14 (2000) 25e30. [8] Q. Shi, D. Hou, K.H. Chung, C. Xu, S. Zhao, Y. Zhang, Characterization of heteroatom compounds in a crude oil and its saturates, aromatics, resins, and asphaltenes (sara) and non-basic nitrogen fractions analyzed by negative-ion electrospray ionization fourier transform ion cyclotron resonance mass spectrometry, Energy fuels. 24 (2010) 2545e2553. [9] A. Hemmati-Sarapardeh, R. Alipour-Yeganeh-Marand, A. Naseri, A. Safiabadi, F. Gharagheizi, P. Ilani-Kashkouli, A.H. Mohammadi, Asphaltene precipitation due to natural depletion of reservoir: determination using a sara fraction based intelligent model, Fluid Phase Equilib. 354 (2013) 177e184. [10] M.M. Shadman, A.H.S. Dehaghani, M.H. Badizad, How much do you know about the methods for determining onset of asphaltene precipitation? Petroleum (2016) (Accepted), 10.1016/j.petlm.2016.08.011. [11] J. Murgich, J. Rodríguez, Y. Aray, Molecular recognition and molecular mechanics of micelles of some model asphaltenes and resins, Energy fuels. 10 (1996) 68e76. [12] G. Porte, H. Zhou, V. Lazzeri, Reversible description of asphaltene colloidal association and precipitation, Langmuir 19 (2003) 40e47. [13] K. Leontaritis, G.A. Mansoori, Asphaltene Flocculation during Oil Production and Processing: a Thermodynamic Collodial Model. In Spe International Symposium on Oilfield Chemistry, SPE, 1987. [14] O.P. Strausz, P.A. Peng, J. Murgich, About the colloidal nature of asphaltenes and the mw of covalent monomeric units, Energy & Fuels 16 (2002) 809e822. [15] S.I. Andersen, J.G. Speight, Petroleum resins: separation, character, and role in petroleum, Pet. Sci. Technol. 19 (2001) 1e34. [16] J. Escobedo, G.A. Mansoori, Viscometric determination of the onset of asphaltene flocculation: a novel method, Spe Prod. Facil. 10 (1995) 115e118. [17] S.A. Mousavi-Dehghani, M.R. Riazi, M. Vafaie-Sefti, G.A. Mansoori, An analysis

[30]

[31] [32] [33]

[34]

[35]

[36] [37]

[38]

[39]

[40]

[41] [42]

[43]

[44]

[45] [46]

117

of methods for determination of onsets of asphaltene phase separations, J. Pet. Sci. Eng. 42 (2004) 145e156. M.M. Shadman, M. Dehghanizadeh, A.H. Saeedi Dehaghani, M. Vafaie Sefti, N. Mokhtarian, An investigation of the effect of aromatic, anionic and nonionic inhibitors on the onset of asphaltene precipitation, J. Oil Gas. Petroche. Tech. 1 (2014) 17e28. T.A. Al-Sahhaf, M.A. Fahim, A.S. Elkilani, Retardation of asphaltene precipitation by addition of toluene, resins, deasphalted oil and surfactants, Fluid Phase Equilib. 194 (2002) 1045e1057. O. Leon, E. Contreras, E. Rogel, G. Dambakli, J. Espidel, S. Acevedo, The influence of the adsorption of amphiphiles and resins in controlling asphaltene flocculation, Energy fuels. 15 (2001) 1028e1032. L.C.R. Junior, M.S. Ferreira, A.C. Da Silva Ramos, Inhibition of asphaltene precipitation in brazilian crude oils using new oil soluble amphiphiles, J. J. Pet. Sci. Eng 51 (2006) 26e36. n, E. Rogel, A. Urbina, A. Andújar, A. Lucas, Study of the adsorption of O. Leo alkyl benzene-derived amphiphiles on asphaltene particles, Langmuir 15 (1999) 7653e7657. lez, A. Middea, Peptization of asphaltene by various oil soluble G.G. Gonza amphiphiles, Colloids Surf. 52 (1991) 207e217. E.F. Ghloum, M. Al-Qahtani, A. Al-Rashid, Effect of inhibitors on asphaltene precipitation for marrat Kuwaiti reservoirs, J. Petroleum Sci. Eng. 70 (2010) 99e106. J. Wang, C. Li, L. Zhang, G. Que, Z. Li, The properties of asphaltenes and their interaction with amphiphiles, Energy fuels. 23 (2009) 3625e3631. A.H. Saeedi Dehaghani, M.M. Shadman, S. Ahmadi, an investigation of stabilization and inhibition strength of the asphaltene inhibitors by viscometric method, Energy Sources, Part A 38 (2016) 2860e2866. M.M. Shadman, A.H. Saeedi Dehaghani, M. Vafaie-Sefti, The effect of inhibitors on asphaltene precipitation in crude oil using the viscometric method, Energy Sources, Part A 34 (2012) 827e838. M. Vafaie-Sefti, M.M. Shadman, A.H. Saeedi Dehaghani, M. Dehghanizadeh, An experimental study on strength of inhibitors stabilizing asphaltene flocculates using a viscometric method, Pet. Res. 25 (2015) 71e81. H.H. Ibrahim, R.O. Idem, Interrelationships between asphaltene precipitation inhibitor effectiveness, asphaltenes characteristics, and precipitation behavior during N-Heptane (light paraffin hydrocarbon)-induced asphaltene precipitation, Energy fuels. 18 (2004) 1038e1048. pez, R. Salas, F. Silva, C. Ferna ndez, C. Urbina, J.C. Lo pez, Resins: J.C. Pereira, I. Lo the molecules responsible for the stability/instability phenomena of asphaltenes, Energy fuels. 21 (2007) 1317e1321. S. Asomaning, Test methods for determining asphaltene stability in crude oils, Pet. Sci. Technol. 21 (2003) 581e590. E. Rogel, O. Leon, Y. Espidel, Y. Gonzalez, Asphaltene stability in crude oils, Spe prod. Facilities 16 (2001) 84e88. A.H. Saeedi Dehaghani, M. Vafaie-Sefti, A. Am erighasrodashti, The Application of A New association equation of state (AEOS) for prediction of asphaltenes and resins deposition during Co2 gas injection, Pet. Sci. Technol. 30 (2012) 1548e1561. Y.F. Hu, T.M. Guo, Effect of the structures of ionic liquids and alkylbenzenederived amphiphiles on the inhibition of asphaltene precipitation from Co2Injected reservoir oils, Langmuir 21 (2005) 8168e8174. K.J. Leontaritis, G.A. Mansoori, Asphaltene Flocculation during Oil Production and Processing: a Thermodynamic Collodial Model. In Spe International Symposium on Oilfield Chemistry, Spe, 1987. E. Rogel, Effect of inhibitors on asphaltene aggregation: a theoretical framework, Energy fuels. 25 (2010) 472e481. C.L. Chang, H.S. Fogler, Stabilization of asphaltenes in aliphatic solvents using alkylbenzene-derived amphiphiles. 1. Effect of the chemical structure of amphiphiles on asphaltene stabilization, Langmuir 10 (1994) 1749e1757. C.L. Chang, H.S. Fogler, 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 10 (1994) 1758e1766. P. Permsukarome, C. Chang, H.S. Fogler, Kinetic study of asphaltene dissolution in amphiphile/alkane solutions, Industrial Eng. Chem. Res. 36 (1997) 3960e3967. S.M. Hashmi, K.X. Zhong, A. Firoozabadi, Acidebase chemistry enables reversible colloid-to-solution transition of asphaltenes in non-polar systems, Soft Matter 8 (2012) 8778e8785. S.M. Hashmi, A. Firoozabadi, Controlling nonpolar colloidal asphaltene aggregation by electrostatic repulsion, Energy & Fuels 26 (2012) 4438e4444. S.M. Hashmi, A. Firoozabadi, Effect of dispersant on asphaltene suspension dynamics: aggregation and sedimentation, J. Phys. Chem. B 114 (2010) 15780e15788. S.M. Hashmi, L.A. Quintiliano, A. Firoozabadi, Polymeric dispersants delay sedimentation in colloidal asphaltene suspensions, Langmuir 26 (2010) 8021e8029. H. Nasrabadi, J. Moortgat, A. Firoozabadi, New three-phase multicomponent compositional model for asphaltene precipitation during CO2 injection using CPA-EOS, Energy & Fuels 30 (2016) 3306e3319. G.A. Mansoori, Modeling of asphaltene and other heavy organic depositions, J. Pet. Sci. Eng. 17 (1997) 101e111. Y. Zhou, G. Stell, Chemical association in simple models of molecular and ionic fluids. Ii. Thermodynamic properties, J. Chem. Phys. 96 (1992) 1504e1506.

118

A.H. Saeedi Dehaghani, M.H. Badizad / Fluid Phase Equilibria 442 (2017) 104e118

[47] J. Lambert, Association in polar vapours and binary vapour mixtures, Discuss. Faraday Soc. 15 (1953) 226e233. [48] A. Anderko, A simple equation of state incorporating association, Fluid Phase Equilib. 45 (1989) 39e67. [49] B. Shirani, M. Nikazar, A. Naseri, S.A. Mousavi-Dehghani, Modeling of asphaltene precipitation utilizing association equation of state, Fuel 93 (2012) 59e66.

[50] A.H.S. Dehaghani, M.H. Badizad, Thermodynamic modeling of gas hydrate formation in presence of thermodynamic inhibitors with a new association equation of state, Fluid Phase Equilib. 427 (2016) 328e339. [51] A.H.S. Dehaghani, M. Vafaie Sefti, M. Emami Meibodi, M. Fallahnezhad, Modeling of coprecipitation of resin and asphaltene in crude oil by association equation of state, Pet. Sci. Technol. 29 (2011) 1932e1947.