Asphaltene Adsorption and Desorption From Mineral Surfaces

Abstract

Summary This paper reports results of asphaltene adsorption/desorption on clayminerals, silica, and carbonates. It also describes the effect of adsorbedasphaltenes on rock wettability and a screeningpyrolysis-flame-ionization-detection (P-FID) test to evaluate the abilitypyrolysis-flame-ionization-detection (P-FID) test to evaluate the ability of solvents to remove asphaltene from kaolin and formation core material. Introduction Reservoir wettability is a major factor controlling the location, fluiddistribution, and flow properties of the system. Wettability conditions affectformation capillary pressure and relative permeability behavior, electricalproperties, and residual oil saturations. The wet-tability of originallywater-wet mineral surfaces may be reverseby adsorption of polar organiccompounds in crude oils. The highest concentration of polar organic compoundsgenerally is found in the heavy ends of crudes, particularly in the asphalteneand resin fractions. Wettability alterations of oil-bearing formations, particularly those containing clay minerals, have been attributed toparticularly those containing clay minerals, have been attributed to adsorption of these compounds onto mineral surfaces. Because of the high molecular weightsand multifunctional character of asphaltenes and resins, their adsorption byspecific minerals is a major element in wettability changes and was thereforeselected for study. Significant factors that control adsorption of asphaltenes and resins onmineral surfaces are (1) the presence, thickness, and stability of water filmson the mineral surface; (2) the chemical and structural nature of the mineralsubstrate; (3) asphaltene and resin contents of the crude; (4) the presence of asphaltenes and resins in crude oils in the form of colloidal micelles oraggregates; and (5) the ability of the hydrocarbon fraction of the crude tostabilize these colloidal aggregates in the oil or even to dissolve them intotrue solution. Further, specific asphaltene/mineral interactions control thedegree to which such adsorption is irreversible with respect to various organicsolvents and, hence, may determine optimum laboratory corecleaningprocedures. A variety of physical measurements, including molecular-weightdeterminations in various solvents, have deduced that asphaltenes associate orform aggregates even in dilute solution. Moschopedis et al. showed that intermolecular hydrogen bonding is involvedin asphaltene association and is reflected in the observed molecular weights. Vapor pressure osmometry determinations in nitrobenzene (epsilonr=34.8) producemolecular weights of 1,650 to 2,100 compared with 5,000 to 6,700 in benzene(epsilonr= 2.3). These nitrobenzene-derived molecular weights may be themolecular weights of the individual asphaltene particles. The power of various solvents has been expressed in terms of Hildebrandsolubility parameters, delta. The relation between delta and asphaltenesolubility was confirmed by Mitchell and Speight. They compared the weight of asphaltenes separated from Athabasca bitumen with a series of solvents(solvent/bitumen volume ratio of 40: 1) with the maximum asphaltene precipitateobtained by addition of an excess of n-pentane. The weight percent of precipitated asphaltenes decreased linearly with increasing delta. Completeprecipitated asphaltenes decreased linearly with increasing delta. Completesolubilization of the asphaltenes in the bitumen was obtained for solvents withdelta greater than = 8.4 cal 1/2 × cm / . For our initial adsorption experiments, the effect of solvent variation forasphaltene adsorption on the clay mineral kaolin was examined. The solventseries toluene/n-dodecane at 1.75 to 1.00 wt/wt toluene and chloroform was chosen primarily because of its increasing delta. The toluene/n-dodecanemixture of aromatic/aliphatic solvents has a delta close to the limiting valuerequired for complete solubilization of our asphaltene sample. Chloroform is aproton donor in hydrogen-bond formation and has a somewhat higher dielectricconstant and dipole moment than the other two solvents. Asphaltene adsorptionstudies were also carried out with a wide variety of other mineral and clayadsorbents from toluene solution. Collins and Melrose, Clementz, and Cuiec, indicated that the presence of a thin film of water on the mineral surfacereduces presence of a thin film of water on the mineral surface reducesasphaltene adsorption and can affect the kinetics of adsorption. To establishbase cases for further work, the presence of water, as well as resins, wasexcluded from all systems. The effects of various solvents on the desorption process were also examinedto evaluate their effectiveness in core-cleaning operations. Solvents werechosen on the basis of their delta values, polar character, andhydrogen-bonding capabilities. Experimental Asphaltene Adsorption on Kaolin From Solution. Adsorption studies werecarried out with a tar-sand-derived asphaltene (npentane insolubles) sample indifferent solvents and kaolin clay mineral (from J.T. Baker Chemical Co.) asthe adsorbent. Table 1 gives the asphaltene elemental analysis. Elementalanalysis and X-ray diffraction (XRD) indicate that the clay is predominantly inthe sodium form, consisting of 15% illite. The Brunauer-Emmett-Teller (BET)surface area, with N gas, was 11.9+/-0.4 m/g (five determinations). Thecation-exchange capacity, measured by Ba/Mg conductimetric titration, was 4.572meq/100 g. The solvents chloroform and toluene were analytical reagent grade(Mallinckrodt Co.) and contained less than 400 and 200 ppm water, re-spectively, according to Karl Fisher titration. n-Dodecane (Aldric Chemical Co.) was 99% pure (with less than 10 ppm water). Analytical reagent gradenitrobenzene (Baker Chemical Co.) had appx. 300 ppm water. Solvents were dried by storing over 0. 4-nm molecular sieves for at least 48hours before use. Kaolin mineral was dried at 110 deg.C for 14 hours and cooledin a desiccator. Glassware was oven-dried; solvents and asphaltene solutionswere manipulated under dry nitrogen atmosphere. Initial asphaltene concentrations were varied from 300 to 2.500 ppm. Mineral/asphaltene solutions in the ratio of 1 : 100 were shaken ppm. Mineral/asphaltene solutions in the ratio of 1 : 100 were shaken mechanicallyfor 48 hours at ambient temperatures; liquid was separated from solid bycentrifugation. Both initial and equilibrium asphaltene solution concentrations weredetermined spectrophotometrically with a Bausch and Lomb Spectronic 1001 TMspectrophotometer. Asphaltene concentrations were determined from calibrationcurves of absorbance vs. concentration at 450 nm for all solvents exceptnitrobenzene, for which 600 nm was used. Practical delta values for the solvent/asphaltene systems were obtained by avariant of the Bichard test and compared with literature values. Asphaltenesolutions in the respective solvents at various concentrations were eachtitrated against n-dodecane, a reference precipitant for asphaltenes. Titrationendpoints were marked by the first appearance of asphaltene precipitate, determined by microscopic examination. Plots of solvent/asphaltene ratio(milliliters per gram) vs. the n-dodecane/asphaltene ratio (milliliters pergram) produced a series of straight lines. The cotangents of per gram) produceda series of straight lines. The cotangents of the angles, theta, formed bythese lines with the x-axis have been used as practical solubility parametersfor the individual solvents.