Energy Fuels 2010, 24, 2369–2375 Published on Web 12/23/2009
: DOI:10.1021/ef901031h
Polycardanol or Sulfonated Polystyrene as Flocculants for Asphaltene Dispersions† Aline F. Lima,‡ Claudia. R. E. Mansur,‡ Elizabete F. Lucas,‡ and Gaspar Gonzalez*,§ ‡ Macromolecules Institute, Federal University of Rio de Janeiro, Av. Hor acio Macedo, 2030 Ilha do Fund~ ao, Rio de Janeiro, RJ 21941-598, Brazil and §Petrobras Research Center, Av. Hor acio Macedo, 950 Cidade Universit aria, Rio de Janeiro, RJ 21941-915, Brazil
Received September 14, 2009. Revised Manuscript Received December 3, 2009
Polycardanol with different molar mass and sulfonated polystyrene with various degrees of sulfonation were prepared by cationic polymerization of cardanol and sulfonation of polystyrene in ethyl sulfate solution, respectively. The molar masses, solution behavior, and effect of these polymers on the stability of asphaltene dispersion were studied as a function of the concentration. The results indicate that, at low concentrations, both sets of polymers behave as flocculants and dispersants at higher concentrations. The flocculation efficiency correlates, within certain limits, with the polymer molar mass and polar groups content.
Asphaltenes are the heaviest, most polar, and nonvolatile components of crude oil, defined as the fraction of petroleum insoluble in light hydrocarbons (pentane, hexane, and heptane) but soluble in benzene and toluene. This fraction is usually composed of condensed polyaromatic rings containing aliphatic and naphthenic side chains and sulfur, oxygen, and nitrogen as heteroelements or functional groups. Metals such as vanadium and nickel are also present in this fraction as part of porphyrinic or nonporphyrinic groups. Resins are defined as the fraction of petroleum that is insoluble in ethyl acetate and soluble in hydrocarbons, such as pentane, heptane, benzene, and toluene. Resins are composed of molecules similar to but less aromatic than asphaltenes. Owing to their greater polarity, which causes them to be less hydrophobic than other petroleum fractions, asphaltenes have a tendency to agglomerate and may undergo phase separation. Variations in crude oil temperature, pressure and/or composition during extraction, transport, or refining can compromise stability and result in phase transitions.1 Previous studies have shown that crude oil stability in relation to changes in asphaltene solubility because of pressure drop under reservoir conditions may been successfully investigated by titration tests with a suitable alkane, normally n-heptane.2,3 In a similar way, the selection of additives to maintain this fraction disperse and avoid its deposition in production flowlines or production facilities is also carried out, comparing the stability of stock tank crude oil/n-heptane mixtures with or without asphaltene dispersants. For characterization purposes, these studies have been extended to model systems
in which asphaltene samples separated and purified following standard procedures4 are dissolved in well-characterized solvents and stability to heptane is evaluated with or without additives.5 In a previous study,5 we reported the effect of various amphiphiles on the stability of asphaltenes dissolved in toluene and it was concluded that solutions in toluene containing alkyl phenols, such as nonyl phenol, were less sensitive to precipitation by the addition of low-molecular-weight alkanes. This effect was ascribed to acid-base interactions between the asphaltenes and the amphiphile and to steric stabilization provided by the hydrocarbon tail. In a subsequent publication,6 it was shown that the liquid extracted from the shell of a cashew nut [cashew nut shell liquid (CNSL) obtained from Anacardium occidentale or Anacardium anum] composed almost completely by phenolic compounds, with some of them also presenting a carboxylic acid group, with long linear alkyl chains containing 15 carbons with variable insaturation degrees, meta-substituted in the aromatic ring, presented a very good performance as an asphaltene stabilizer. A similar result7 was obtained for cardanol, a fraction containing various isomers of 3-(pentadecyl)phenol separated by vacuum distillation of CNSL. In the same study, it was also reported that polycardanol, a polymer obtained by cationic polymerization of cardanol, was not only less efficient as dispersants than its respective monomers but, instead, enhanced asphaltene precipitation. In fact, polymer science has been applied to several steps of petroleum production, and a graduate course in this area has been implemented in Brazil.8,9
† Presented at the 10th International Conference on Petroleum Phase Behavior and Fouling. *To whom correspondence should be addressed. E-mail:
[email protected]. (1) Branco, V. A. M.; Mansoori, G. A.; Xavier, L. C. A.; Park, S. J.; Manafi, H. J. Pet. Sci. 2001, 32, 217–236. (2) De Boer, K. B.; Leerlooyer, K.; Eigner, M. L. P.; van Bergen, A. R. D. SPE Prod. Facil. 1995, 10 (1), 55–61. (3) Wang, J. X.; Buckley, J. S. An experimental approach to prediction of asphaltene flocculation. Presented at the Outer Continental Shelf (OCS), Houston, TX, Feb 13-16, 2001; SPE Paper 64994.
(4) Institute of Petroleum (IP). Standard test method for determination of asphaltenes (heptanes insoluble) in crude petroleum and petroleum products. IP, London, U.K., 2001; IP 145/01. (5) Gonzalez, G.; Middea, A. Colloids Surf. 1991, 52, 207–217. (6) Moreira, L. F.; Gonzalez, G.; Lucas, E. F. Polim.: Cienc. Tecnol. 1998, Jul/Set, 46–54. (7) Moreira, L. F.; Gonzalez, G.; Lucas, E. F. J. Appl. Polym. Sci. 1999, 73, 29–34. (8) Lucas, E. F.; Mansur, C. R. E.; Spinelli, L.; Queiros, Y. G. C. Pure Appl. Chem. 2009, 81, 473–494. (9) Lucas, E. F.; Spinelli, L. S. J. Mater. Educ. 2005, 27, 43–51.
1. Introduction
r 2009 American Chemical Society
2369
pubs.acs.org/EF
Energy Fuels 2010, 24, 2369–2375
: DOI:10.1021/ef901031h
Lima et al.
In the present work, the flocculation of asphaltenes by polycardanol polymers is examined in further detail and the studies are extended to sulfonated polystyrene (PSS). This type of polymer offers a wider range of possibilities because molecules containing different sulfonation degrees may be prepared through controlled substitution reactions. Asphaltene flocculation by polymers has not been studied before because of the simple reason that it does not present interest to the petroleum industry. However, these studies may be useful to obtain some insight on the mechanism of asphaltene precipitation and redispersion and the mechanisms involved in these processes. Furthermore, asphaltene flocculants may find application in processes such as deasphaltation, in which asphaltic material is removed from petroleum residues as part of some petroleum derivates, such as wax or lubricating oils. Another potential application of asphaltene flocculants may be heavy-oil upgrade, in which a partial reduction of the asphaltic fraction would improve the crude oil market price.
The sulfonation of PS was carried out using the following experimental conditions.13 First of all, an acetylsulfate solution was prepared by mixing 10 mL of methylene dichloride and 11.25 mL of acetic anhydride in a beaker maintained at 0 °C in an ice-water bath. After that, 3.75 mL of concentrated (95-97%) sulfuric acid was added very slowly and the reaction mixture was maintained under stirring at room temperature up to the formation of a homogeneous translucent solution. The chemical modification of PS was carried out under stirring, at an inert atmosphere (N2) and low temperature, by mixing 300 mL of dichloromethane and 30 g of PS. After that, the mixture was heated at the solvent reflux temperature (40 °C), to solubilize the polymer, and an appropriate amount of acetylsulfate previously prepared was added, keeping the system at 40 °C for different reaction times: 5, 10, or 15 min. After 50 mL of isopropyl alcohol was added, the reaction conditions were maintained for 30 min more before cooling the system to room temperature. The PSS was obtained by precipitating the solution reaction in 500 mL of hot distillate water. During this process, the solvent is eliminated and the acetylsulfate is hydrolyzed. Asphaltenes, resins, cardanol, and polycardanol were characterized by Fourier transformed infrared spectroscopy (FTIR, Perkin-Elmer, model 1720x). The liquid sample (in the case of asphaltenes and resins, they were solubilized in toluene) was casting on a KBr cell. All analyses were carried out at a resolution of 2 cm-1, 20 scans, and a number wave range of 4000-400 cm-1. The samples of PS and PSS were solubilized, at 1% p/v, in deuterated benzene and dimethylsulfoxide (DMSO), respectively, and characterized by hydrogen nuclear magnetic resonance (1H NMR) in a spectrometer Varian Mercure 300, using tetramethylsilane (TMS) as an internal reference. The cardanol, polycardanol, and PS average molar masses were determined by size-exclusion chromatography (SEC) in a Waters apparatus (Waters detector IR-2414 and columns of Ultrastyragel 500-100 and 50 A˚ of porosity) calibrated with PS standards. Tetrahydrofuran (THF) was used as a solvent. A qualitative estimation of polymer solubility was obtained by measuring the cloud point of 10 g/L polymer solutions in solvent mixtures. The solutions in glass tubes were suspended in a hot water bath at 60 °C and slowly submitted to cooling and heating cycles under moderate stirring. The temperature was measured using a thermometer immersed in the solutions, and the average between the appearance and disappearance of turbidity upon cooling and heating was identified as the cloud point. Toluene, n-heptane, acetone, and a binary mixture of these solvents were used in these solubility studies. The efficiency of the various additives as dispersants or flocculants for asphaltene dispersions was assessed through simple precipitation tests. A total of 10 cylindrical flasks were used, and 1 mL of asphaltene dispersed in toluene was added to each flask. We prepared nine toluene/heptane mixtures with heptane concentrations varying from 0 to 100% (v/v). A total of 9 mL of each mixture was added to each flask already containing 1 mL of asphaltene in toluene. The flasks were sealed, gently homogenized by hand shaking, and left to stand in a temperature-controlled room for 24 h. After this period, the solutions were centrifuged at 3000 rpm for 30 min and the absorbance of the particle-free supernatant measured at 850 nm in a Varian, Cary 50, UV-vis spectrometer using a 5 mm optical path probe. The concentration of asphaltenes was obtained using calibration curves of absorbance versus the asphaltene concentration in toluene. The error of the method was slightly higher for low asphaltene concentrations but did not exceed 5% for three independent measurements. The tests were performed with an
2. Experimental Section 2.1. Materials. The asphaltic residues used were obtained from a local refinery (REDUC, Rio de Janeiro, Brazil). Typically, these residues contain around 14% asphaltenes and 48% resins.10 Two petroleum samples were supplied by Petrobras and named crude oil M and crude oil C. Technical samples of CNSL and cardanol were obtained from RESIBRAS (Companhia Brasileira de Resinas, Brazil). CNSL is a dark, viscous, natural organic fluid obtained from A. occidentale or A. anum.11 The composition of CNSL depends upon the process used to extract the material from the raw cashew nut shell;12 however, phenolic compounds with linear alkyl chains containing 15 carbons with variable degrees of insaturation, meta-substituted in the aromatic ring, and anacardic acid are the major components. The polystyrene (PS) was obtained from Dow Chemical Company, Brazil. Toluene was purchased from Vetec Quı´ mica Fina Ltd., Brazil, and dried by distillation at 110 °C in the presence of metallic sodium. The acetic anhydride, sulfuric acid, dichloromethane, isopropyl alcohol, acetone, and n-heptane (high purity) were also supplied by Vetec Quimica Fina Ltd. and used as received. The tetrahydrofuran (99.5% pure), deuterated benzene, and dimethylsulfoxide were purchased from Tedia Brazil. 2.2. Methods. The asphaltene fraction was separated from the asphaltic residue/petroleum following a procedure similar to the standard method to precipitate and purify this fraction.4 The residue/petroleum was Soxhlet-extracted with n-heptane (C7) to complete removal of the C7-soluble fraction. The solid phase (containing asphaltenes) was dissolved in toluene, filtered, and dried under vacuum to produce the asphaltene fraction. The C7soluble fraction (containing resins) was dried, resolubilized in n-heptane, filtered, and dried to produce the resin fraction. Because we used the asphaltic residue in some tests, this was also purified by Soxhlet extraction with toluene. The solid phase (containing asphaltenes and resins) was dried, resolubilized in toluene, filtered, and dried. Polycardanol was obtained by cationic polymerization of cardanol at 140 °C using boron trifluoride diethyl etherate [BF3 3 O(CH2CH3)2, 1 wt %] as an initiator and 40 min of reaction time.7 The reacting mixture was maintained under nitrogen in an oxygen- and moisture-free atmosphere. Through changes in the initiator concentration, it was possible to obtain polymers with different molar mass. (10) Private Petrobras Databases. (11) Mothe, C. G.; Milfont, W. N. Rev. Quim. Ind. 1994, 695, 15–19. (12) Tyman, J. H. P. Chem. Soc. Rev. 1979, 8, 499–536.
(13) Baigl, D; Seery, T. A. P.; Williams, C. E. Macromolecules 2002, 35, 2318–2326.
2370
Energy Fuels 2010, 24, 2369–2375
: DOI:10.1021/ef901031h
Lima et al.
additive (cardanol or polycardanol) in toluene at a concentration range of 0.05-1.0 wt/vol %. The PSS co-polymers were insoluble in toluene, and an alternative procedure was developed for these chemicals. As result of solubility tests, a 60:40 toluene/acetone mixture was selected for these polymers. The precipitation test for asphaltenes was also carried out using acetone as a flocculant following the same method as described above for n-heptane. Because it was observed that acetone also induced the precipitation of asphaltenes from their toluene solution, by increasing the solution solubility parameter in this case, hence, the following procedure was adopted. PSS solutions were prepared by dissolving the polymer in 9 mL of a toluene/acetone mixture (50:40, v/v) and left for 24 h for complete solubilization. These solutions were added to the cylindrical flasks already containing 1 mL of petroleum fraction in toluene (asphaltenes or resins or asphaltic residue). Therefore, the final toluene/acetone content was 60:40 (v/v). The asphaltene concentrations were 0.1, 0.25, and 0.5 wt/vol %. The concentrations of resins and asphaltic residue were 0.25 and 0.5 wt/vol %. In addition, the concentrations of PSS were varied from 0.1 to 0.5 wt/vol %. The asphaltene solution stabilization test is summarized as follows: (i) All tests were performed using a constant asphaltene concentration in solvent. (ii) Tests varying the heptane concentration were performed with cardanol, PCN1, and PCN5 at 1.0 wt % added to asphaltenes in toluene. (iii) The effect of the type and concentration of additives were studied for (1) asphaltenes in toluene and adding different PCN type (PCN2, PCN3, PCN4, and PCN5) and PCN concentrations in toluene (from 0.0 to 1.0 wt/vol %) and (2) asphaltenes in toluene and adding different PSS type (PSS4a, PSS4b, PSS5a, PSS6b, and PSS6c) and PSS concentrations in toluene/acetone (from 0.0 to 0.5 wt/vol %).
3. Results and Discussion 3.1. Polymers Characterization. The FTIR spectrum of CNSL is rather complex and shows bands and peak characteristics of O-H, C-H, CdC, and C-O bonds that confirm the presence of unsaturated phenolic compounds in the sample, as previously described by other authors.7,14,15 The FTIR spectra obtained for cardanol and polycardanol are shown in Figure 1. For cardanol (Figure 1a), its phenolic structure is confirmed (absorption bands at 3350, 1351, 1265, 1487, and 1457 cm-1), presenting a hydrocarbon tail (bands at 2926 and 2854 cm-1), a meta substitution on the aromatic ring (absorption in a range of 1650-2000 cm-1), and olefinic bonds (bands at 3007, 1612, and 1589 cm-1 and 988, 945, and 912 cm-1). When the FTIR spectra of cardanol and polycardanol are compared (panels a and b of Figure 1, respectively), it is observed that some peaks are similar; however, in the FTIR spectrum of polycardanol, the absorption bands at 988 and 912 cm-1 disappeared and the band at 3007 cm-1 decreased. All of these absorption peaks are related to the unsaturation of the hydrocarbon chain, which suggests that the polymerization took place at the cardanol double bonds. It is also observed that the axial deformation of the hydroxyl group remained the same, which suggests that any reaction took place at the hydroxyl groups of cardanol. Such a result confirms the cardanol addition polymerization through the double bonds of the side alkyl chain of cardanol.7,14,15
Figure 1. FTIR absorption spectra of (a) cardanol and (b) polycardanol.
Five polycardanol samples, hereafter called PCN1, PCN2, PCN3, PCN4, and PCN5, presenting SEC number average molar mass of 2300 (previously determined8), 1500, 2100, 2900, and 4200 Da, respectively, were used in this study. The samples were soluble in toluene and toluene/heptane mixtures up to a 40:60 ratio in volume. In this later medium, the solutions presented a cloud point at 23-29 °C. The PS 1H NMR spectrum presented the characteristic assignment for this polymer: a centered peak at 1.5 ppm corresponding to the linear chain hydrogen and two centered peaks at 6.5 and 7.0 ppm (A) corresponding to the aromatic ring hydrogen. The spectra corresponding to PSS samples presented, beside the PS peaks, a centered peak at 7.4 ppm (B) ascribed to the aromatic ring hydrogen neighboring the sulfonic group. From these spectra, the sulfonation degree (mol %) obtained for the various polymers may be calculated by eq 1 ð1Þ mol % ¼ 3AB =ð2AA þ 3AB Þ where AA and AB are the peak areas at 7.0 and 7.4 ppm, respectively. The results presented in Table 1 indicate that, as expected, longer reaction times result in polymers with a higher degree of sulfonation. It is important also to note that, even though the reaction conditions and experimental procedures were rigorously maintained, some differences in sulfonation are still observed for samples prepared in different runs.
(14) Anthony, R.; Pillai, C. K. S. J. Appl. Polym. Sci. 1990, 41, 1765– 1775. (15) Tyman, J. H. P. Chem. Soc. Rev. 1979, 8, 499–536.
2371
Energy Fuels 2010, 24, 2369–2375
: DOI:10.1021/ef901031h
Lima et al.
report that, independent of their origin or concentration, asphaltenes precipitate at solution solubility parameters between 16.4 and 16.8 MPa1/2.19,20 As expected, resins that are soluble in heptane remain stable in the whole composition range. The asphaltic residue precipitates at 30% heptane, equivalent to a solubility parameter of 16.1 MPa1/2. This effect may be ascribed to the resins, and it indicates that this fraction indeed stabilizes asphaltenes. However, considering that the resin/asphaltene ratio in these type of residues is on the order of 4-6, it may be inferred that it is, in fact, a minor effect probably related to the formation of less polar asphaltene-resin aggregates. The hypothesis that resins present a dispersing effect to maintain asphaltenes in solution used by some authors to develop asphaltene depositional models21 seems to be inaccurate because effective dispersants show good performance at concentrations on the order of parts per million (ppm) in relation to the asphaltenes. 3.3. Effect of PCN and PSS on Asphaltene Solution Stability. The effect of PCN and PSS on the stability of asphaltene solutions was assessed using solubility diagrams similar to those shown in Figure 2 or directly measuring the amount of asphaltenes remained in solution by the addition of solvent mixtures with or without the dispersant/flocculant agent. The solvent mixture for PCN was toluene/heptane, and the solvent mixture for PSS was 60:40 toluene/acetone. Figure 3a presents the effect of PCN1 on the precipitation of asphaltenes separated from the asphaltic residue. For comparison, the effect of cardanol is also shown in the graph. It is clear that, while cardanol appears as a good dispersant, increasing the heptane/toluene ratio for asphaltene precipitation, polycardanol induces precipitation at low heptane concentrations and increases the precipitated amount for all solvent ratios. The same flocculant effect is observed for asphaltenes separated from a crude sample (Figure 3b). In Figure 4, the precipitation of asphaltenes initially dissolved in toluene by various polycardanol samples is shown. These tests were carried out in the absence of heptane, and considering that the nanoaggregate formation concentration for asphaltenes has been reported in the range of 50-150 mg/L,22 the asphaltene concentration of 60 mg/L was used to ensure the absence of large aggregates and maximum solubility. The four polymers induced asphaltene precipitation, and the effect was more significant for low polymer concentrations. The flocculation efficiency of the polymers correlates with the molar mass as it increases in the order PCN2, PCN3, PCN4, and PCN5. Mechanisms of flocculation by polymers related to polymer molar mass and concentration have been explained in the literature.23,24 The results confirm that polycardanol behaves as a flocculant for asphaltene dispersions, either enhancing the precipitating effect of aliphatic hydrocarbons or, even in dilute solutions, not containing a non-solvent as a destabilizing agent. Originally, it was assumed that, considering that molecules containing acidic phenol groups, such as cardanol,
Table 1. Sulfonation Degree of Modified PSs sample
sulfonation degree (mol %)
PSS4a PSS4b PSS4c PSS5a PSS5b PSS5c PSS6a PSS6b PSS6c
4.0 5.0 5.0 7.0 7.0 7.0 10.0 12.0 13.0
Figure 2. Precipitation of petroleum fraction diagrams for asphaltic fractions in toluene/n-heptane mixtures: (9) asphaltic residue, (2) asphaltenes, and (b) resins. Initial petroleum fraction concentrations = 0.1 wt/vol %.
The reproducibility of the reaction carried out at the same conditions was (3.0%, in terms of sulfonation degree. The number and weight average molar masses (M n and M w) obtained by SEC for the PS used in the sulfonation reactions were, respectively, 87 300 and 234 200, correspond ing to a polydispersity (M w/M n) of 2.7. This PS sample presented a limited solubility in toluene, and it was insoluble in toluene/heptane mixtures but soluble in all proportions of toluene/acetone. The PSSs were insoluble in toluene and toluene/heptane but presented good solubility in toluene/ acetone mixtures containing more than 30% acetone. 3.2. Asphaltene Characterization and Solubility Behavior. The asphaltene samples correspond in all cases to the fraction separated from asphaltic residues or crude oil, which is soluble in toluene and insoluble in n-heptane. The FTIR spectra registered for these samples presented the characteristic bands described for the asphaltene fraction in previous works.16-19 Figure 2 presents typical solubility diagrams for an asphaltic residue and its corresponding asphaltene and resin fractions in toluene/n-heptane mixtures. Asphaltene is the less soluble fraction, becoming insoluble at a heptane concentration of around 45%, which corresponds to a solvent mixture solubility parameter of 16.8 MPa1/2. This value is in good agreement with the literature results that
(20) Wiehe, I. A. American Institute of Chemical Engineers (AIChE) Spring National Meeting, Houston, TX, March 14-18, 1999. (21) Leontaritis, K. J. Presented at the Society of Petroleum Engineers (SPE) Production Operation Symposium, Oklahoma City, OK, March 13-14, 1989; SPE Paper 18892. (22) Andreatta, G.; Brostrom, N.; Mullins, O. C. Lagmuir 2005, 21 (7), 2728–2736. (23) Brostow, W.; Pal, S.; Singh, R. P. Mater. Lett. 2007, 61, 4381– 4384. (24) Tuinier, R.; Rieger, J.; de Kruif, C. G. Adv. 2003, 103, 1–31.
(16) Yen, T. F.; Erdman, J. G. Prepr.-Am. Chem. Soc., Div. Pet. Chem. 1962, 7, 5. (17) Jada, A.; Ait Chaou, A. J. Pet. Sci. Eng. 2003, 39, 287–296. (18) Gonz alez, G.; Souza, M. A.; Lucas, E. F. Energy Fuels 2006, 20, 2544–2551. (19) Souza, M. A.; Oliveira, G. E.; Lucas, E. F.; Gonzalez, G. Prog. Colloid Polym. Sci. 2004, 128, 283–287.
2372
Energy Fuels 2010, 24, 2369–2375
: DOI:10.1021/ef901031h
Lima et al.
Figure 3. Asphaltene precipitation diagrams in toluene/heptane mixtures in solutions with or without polycardanol. (a) Asphaltenes separated from the asphaltic residue. Initial asphaltene concentration = 100 mg/L. (þ) No additives, (b) 1.0 wt % cardanol, and () 1.0 wt % polycardanol PCN1. (b) Asphaltenes separated from crude oil sample M. Initial asphaltene concentration = 1000 mg/L. (þ) No additives and () 1.0 wt % polycardanol PCN5.
Figure 4. Asphaltene precipitation by various polycardanol samples in the absence of heptane. Asphaltenes separated from crude oil sample C. Initial asphaltenes concentration = 60 mg/L. (9) PCN2, (b) PCN3, ([) PCN4, and (2) PCN5.
results in less asphaltene-asphaltene interactions to form aggregates susceptible to flocculate. To guarantee the complete solubilization of the polymers, the effect of the PSS on asphaltene solubility was studied in 60:40 toluene/acetone mixtures. Asphaltenes are insoluble in acetone and present a limited solubility in toluene/acetone mixtures. At this solvent ratio, nearly 30% of its initial asphaltene concentration precipitates. This effect is probably related to the formation of dimers by the acetone through dipole-dipole interactions,25 because for this solution solubility parameter, asphaltenes are expected to be soluble.18 Figure 5 presents the asphaltene solubility diagrams in the presence of PSS polymers presenting various levels of sulfonation. PS and 3.8% PSS did not modify asphaltene solubility. Polymers with sulfonation levels between 5 and 10% initially reduce the asphaltene concentration in solution and, subsequently, increase it again for polymer concentrations beyond 0.2%. The first branch of the curves, ascribed to flocculation, becomes less important for 12% sulfonation and disappears completely for 13% sulfonation, which presents a moderate dispersant effect.
were capable of stabilizing asphaltenes, the incorporation of several of these groups susceptible to interact with asphaltene molecules in a polymeric structure would result in an efficient asphaltene dispersant. The flocculation caused by these polymers may be ascribed to the formation of aggregates presenting a higher polarity than the original asphaltenes because of the interaction of a limited number of the polymer phenol groups with the asphaltenes, leaving the rest of them exposed to the periphery of the aggregates. One of the reasons for this behavior is that the double bonds through which the polymerization occurs are relatively close to the pendant phenolic groups, and consequently, these groups are also very close in the polymer. Therefore, when a phenolic group interacts with a polar group of asphaltene molecules, the interaction of PCN neighbor phenolic groups with another asphaltene polar group becomes difficult, so that few phenolic groups remain free to increase the aggregate polarity. The results in Figure 4 indicate that, for all PCN samples, flocculation initially increases with concentration, goes through a maximum, and decreases at higher concentrations. This behavior may be ascribed to the amount of phenol groups available to interact with the asphaltene polar groups. A large number of asphaltene-polymer interactions
(25) Elias, H.-G. Macromolecules 1: Structure and Properties; Plenum Press: New York, 1977.
2373
Energy Fuels 2010, 24, 2369–2375
: DOI:10.1021/ef901031h
Lima et al.
Figure 5. Precipitation test of asphaltenes in the presence of PSS. Asphaltenes separated from the asphaltic residue. Initial asphaltene concentration = 0.25 wt/vol %. (9) PSS4a, (b) PSS4b, (2) PSS5a, (1) PSS6b, and (left-facing triangle) PSS6c.
The poor solubility and precipitation of asphaltenes in toluene/acetone mixtures is caused by the increase in the solution solubility parameter caused by the acetone (δ = 20.05 MPa1/2), and at the precipitation onset, the less polar components of the asphaltene fraction become insoluble. Therefore, at a particular toluene/acetone ratio, the components that remain dissolved correspond to a more polar subfraction of the asphaltenes. In this context, the effect of PSS may be ascribed to asphaltene-PSS interactions that at low polymer concentrations or for low sulfonation levels conduce less polar aggregates susceptible to precipitation. For higher concentrations or higher sulfonation levels instead, these interactions are insufficient to cover all of the polymer sulfonic acid groups and the aggregates are sufficiently polar to maintain or even increase their solubility in the solvent mixture. The curves presented in Figure 5 follow a pattern similar to the curves in Figure 4, and the considerations asserted for PCN in relation to the interaction between the polymer and asphaltene polar groups may be considered applicable to PSS. In comparison to both sets of polymers, some PSS samples were more effective as flocculants. In principle, it may be assumed that this difference correlates with the distribution of the polar group in the polymer. For PCN, these groups are closer to each other than in PSS, where they may distribute along the polymer chain. The estimated average distance between the sulfonic groups, for instance, for the PSS 13%, is about 10 times higher than the distance between the phenols in PCN, and the proximity of these groups in this latter case may prevent the interaction of asphaltenes with neighboring polar groups of the polymer. In PSS molecules, the distance of the polar groups is related to their content in the polymer composition. Although it is well-known that the reaction is random, it is possible to consider that, the higher the content of polar groups, the shorter the average distance between them. Because a relationship between the ratio of polar group/ asphaltene concentration in solution and stabilization behavior was observed, a representative scheme is suggested in Figure 6. Three different situations are analyzed: (1) a relatively high concentration of polymers presenting a low
Figure 6. Schematic representation of the interactions between PSS and asphaltenes. (a) Relatively high concentration of polymers presenting a relatively low content of polar groups. (b) Relatively low concentration of polymers presenting a relatively high content of polar groups. In both cases, the systems present a low concentration of PSS polar groups in solution in relation to the asphaltene interaction site concentration, provoking aggregation. (c) High concentration of PSS polar groups in solution in relation to the asphaltene interaction site content, provoking dispersion.
content of polar groups and a high asphaltene interaction site concentration in relation to the PSS polar groups (Figure 6a), (2) a relatively low concentration of polymers presenting a high content of polar groups and a high asphaltene interaction site concentration in relation to the PSS polar groups (Figure 6b) (these two situations conduce flocculation), and (3) a high concentration of PSS polar groups in solution in relation to the asphaltene interaction site content (in this case, stabilization rather than flocculation takes place). 2374
Energy Fuels 2010, 24, 2369–2375
: DOI:10.1021/ef901031h
Lima et al.
fraction are related to the content of polar groups in solution, which is selected by varying polymer concentration and/or polymer type, which means the content of polar groups in the polymer structure. This implies that further developments would require optimization procedures to fit the products for one or another application.
4. Conclusions It is possible to prepare polymeric compounds to flocculate asphaltenes dissolved in organic media. Suitable polymers must contain polar groups capable of undergoing strong interactions with asphaltenes and a solution behavior similar to asphaltenes. The two sets of polymers studied presented flocculation and dispersion effects depending upon molar mass and concentration. For the polymer systems studied in this work, it seems that, better than concentration, the polymer effects on asphaltene
Acknowledgment. The authors thank Petrobras, the Petroleum Brazilian Agency (ANP), and the National Scientific and Technological Research Council (CNPq) for supporting this study.
2375
