3.1. Use of minerals

Among the materials that have been used for adsorption of asphaltenes, mainly minerals found in nature have been investigated because of their low cost and high availability. However, one of the problems in the oil fields is that clays and rocks, sometimes found there, adsorb asphaltenes causing some losses in resulting crude oil extraction. One of the oldest records (González and Middea, 1987) in this area is the feldspar, which demonstrated an adsorption capacity of 0.94 mg/cm2. Studying the mechanism of its adsorption of asphaltenes, it was indicated that fine particles of feldspar with polar fractions of adsorbed oil play an important role in stabilizing water-in-oil emulsions formed in some secondary processes of oil recovery. In a related research using quartz (González and Middea, 1987), its adsorption capacity does not alter its surface charge or the superficial aspects after being desorbed. Therefore, this mineral is not affected by the presence of adsorbed organic species.

The important materials to be analyzed here are also minerals such as Bedford limestone calcite (97.3% SiO2, 1.7% Al2O3, 0.5% MgCO3), Berea sandstone (93.1% Al2O3, 3.86% FeO, 0.54% MgO, 0.25% Fe2O3, 0.11%), and those on the basis of dolomite CaMg(CO3)2 (Castro et al., 2009). These materials are the most common in nature and present a good adsorption capacity for heavy hydrocarbons, using standard conditions (pressure and temperature). In some studies, the adsorption isotherms of heavy hydrocarbons are reported (Castro et al., 2009). The following adsorption capacities for asphaltenes were noted: the dolomite having highest adsorption capacity with 13.1 mg/g followed by Bedford limestone calcite with an adsorption capacity of 12.8 mg/g and Berea sandstone with 6.8 mg/g. Main influencing factors on adsorption capacity are the morphology and major surface area of the Berea sandstone compared with two other aforementioned materials.

Other important and frequently used natural materials are siltstone and montmorillonites (Na,Ca)0.33(Al,Mg)2(Si4O10)(OH)2·nH2O (López-Linares et al., 2009). Two montmorillonites, one with calcium atoms and another with sodium atoms, were compared with synthetic microporous kaoliniteAl2Si2O5(OH)4, revealing that these materials, after being submerged for 300 min, have the order in their adsorption capacity as shown below: macroporous kaolinite (2.30 mg/g) > Ca-montmorillonite (1.2 mg/g) > Na-montmorillonite (0.80 mg/g) > siltstone (0.75 mg/g). These capacities are obviously much lower than those analyzed by Castro et al. (2009).

3.2. Main important factors

López-Linares et al. (2009) indicated that some of the most important variables are the textural properties of these materials and also properties of natural asphaltenes used for analysis of their adsorption. The most important factor is the surface area of solids. However, it mainly depends on the pore volume and pore diameter of the adsorbent, possibly indicating that adsorbed asphaltenes provide a substrate followed by the adsorption of molecules to reach saturation, either pore volume and chemical equilibrium. These last two variables limit the adsorption of asphaltenes on the material surface. It is also mentioned that size of asphaltene molecule adhered on the surface does not limit the ability of the substrates. In a related report, Cosultchi et al. (2005) studied Na-montmorillonite with a solution of known asphaltene concentration of 2.5 wt.% and 6.4 wt.% of polar species containing metals in their structures. The micro and macroscopic diffusion data indicate that it is a slow and continuous process. Due to the macroscopic morphology of clays and viscosity of microemulsions (Hannisdal et al., 2006), the process tends to be continuous but slow. The viscosity does not limit the adsorption capacity, but directly influences the time needed for clays to reach its maximum adsorption capacity. Maximum adsorption was observed after 40 h; however, in the first 2 h, 70% of the total adsorption capacity of the material is obtained.

Jada et al. (2006) reported the effects of asphaltenes in dependence on particle size and the surface charge of two montmorillonite clays in an aqueous mediumin order to establish any correlation among these variables and their adsorption capacity. The effects of pH in the applied water were also examined, confirming that the reaction mechanism of absorption of asphaltenes involves interactions between surface groups of asphaltenes such as carboxylic and carboxylate ones to the silanol and aluminol groups of the mineral. The pH in the water contributes to the saturation effect of the material surface thus diminishing its ability to adsorb asphaltenes, consistent with the above explanations by Gonzalez and Taylor (2016). The adsorption capacity was shown to be greatly affected by the pH more than by moisture.

On the other hand, referring specifically to the kaolinite, better results of its ability on adsorption by surface area were reported by Wang et al. (2016)showing an adsorption capacity of 3 mg/m2 giving a more accurate value. However, this report notes that the adsorbed amount depends not only on the variables mentioned above, but also on the asphaltene/kaolinite ratio. Another less-impact factor was found to be heterogeneity of asphaltene solutions. It was also mentioned that the repeated contact of asphaltene solutions with kaolinite caused the adsorption of asphaltenes in solution of up to 98%. This seems like a great result; it was indicated that the kaolinite, after being immersed several times, does not show a selective adsorption in respect to any particular species. However, the repeated exposure of kaolinite to the solution of asphaltenes greatly increased its adsorption capacity after each immersion. The surface composition data, analyzed by XPS, show that the kaolinite surface remained without alterations despite adsorptions. Further, strong adsorption of asphaltenes was found to occur preferentially on asphaltenes already adsorbed on the kaolinite surface. This proves a great ability of asphaltenes tending to have strong interactions with other asphaltenes and producing agglomerations contributing to a double layer on the surface. Finally, Wang et al. (2013) showed that the maximum thickness of asphaltenes adsorbed on kaolinite, with a total surface coverage, is estimated to be 11 nm based on the XPS depth profile. It was also indicated that no preferential adsorption of nitrogen or rich sulfur species at the interface between the asphaltenes and kaolinite was registered.

Both kaolinite and illite (K,H3O)(Al,Mg,Fe)2(Si,Al)4O10[(OH)2,(H2O)] were studied in a related report (Wang et al., 2013). Having as main target the surface charges and wettability, those interactions that occur between surface groups of asphaltenes and silanol (-(Si-OH)-)/aluminol (-(Al-OH)-) mineral groups, were studied in respect to their mechanisms. Another factor, not mentioned in other works, is the hydrophobicity of the observed clay materials which leads to a modification of properties of the mineral surface, such as surface charge, composition, and surface energy (Gan and Liu, 2008). The asphaltenes containing carboxylic and carboxylate groups have greater adsorption of asphaltenes by interaction of these groups and those of silanol and aluminol in the mineral. The data indicate that the maximum adsorption of asphaltenes depends on the surface charge in the clay; the latter contradicts the findings of Wang et al. (2013) because of non-consideration of the interaction of adsorbed and free asphaltenes.

In some cases, we have observed that some variables for these minerals predominate over other variables. Jada et al. (2009) raised the possibility of mineral saturation with water, which improves their adsorption capacity. Meanwhile, Wang et al. (2016) noted that the clay particles, as kaolinite, in contact with water, change the state from hydrated to dehydrated. This could be explained why in the report mentioned above (Wang et al., 2016) no morphological change is seen on the surface, if the effect was caused by water absorbance in the surface of kaolinite, which plays an important factor for adsorption of asphaltenes. These data, observed for wetting of materials, were thoroughly studied by Gonzalez and Taylor (2016) who indicated that water is involved in the behavior of asphaltenes. For example, deposition delays of asphaltene aggregates were noted, meaning that their agglomeration is avoided, producing micro-emulsions instead. It was also established that the adsorption of asphaltenes on mineral surfaces is reduced in the presence of water.

In the processes, where steam is used, the adsorption is affected by this condition; the adsorption of organic molecules in vapor phase is reduced in the presence of pre-adsorbed water (Rankin et al., 2012), so it is interested to get to know and determine if the same effect is applicable for liquid phase adsorptionof asphaltenes by materials. For some materials, it has been reported to reduce even four times its adsorption capacity. The water adsorbed in the surface of a material generate a monolayer reducing the ability to adsorb asphaltenes in the material. It is believed to be the first demonstration of the effect of pre-adsorbed water on the adsorption of material in the solution. The highest adsorption of asphaltenes occurs at very low relative humidity suggesting that its presence is inversely proportional to the adsorption of asphaltenes on different mineral surfaces. In addition, surface water films have been observed in samples; this may suggest that the adsorbed water inside the material generates interactions leading to an improved adsorption. However, in this case a possibility exists that asphaltenes and water are competitive to the same (silanol) surface sites. These results added a support for the existence of films separating water from asphaltenes on the solid surfaces of natural sands and can also serve to explain the differences between data adsorption materials for asphaltenes reported in the literature.

6.1. Nanostructured iron oxides

The nanomaterials, applied for asphaltene treatment and viscosity decrease, are generally nanostructurized metals in distinct forms (in a lesser degree), which can be free or on such supports as, for example, alumina or graphene, and metal oxides (widely represented, especially those on the basis of iron oxides). Thus, asphaltenes found to form aggregates on the iron oxide surface with a diameter of a few hundred nm and about 50 nm height. No clear concentration dependency of aggregate size was observed (Balabin et al., 2011). Iron oxide nanorods were used both as a catalyst as an adsorbent (Al-Ruqeishi et al., 2016). Small nanorods are important to increase the chemical reactivity and heat conductivity in hydrocarbons when heated from an external source of radiation, in order to reduce the viscosity of crude oil. The viscosity was shown to be reduced by 10% when 0.2 g of nanorods were added to 1 L of heavy oil at 30 °C. This reduction increased to 38% and 49% when 0.4 and 0.6 g/L of additives, respectively, at the same temperature were added. On the other hand, when 0.8 g of heavy oil nanorods were added, no appreciable change in viscosity was found, indicating the achieved saturation point of the nanorods, delimited by their surface area.

Among recent studies on distinct Fe2O3 modifications, nanoparticles of iron oxides, magnetite (γ-Fe2O3) and hematite (α-Fe2O3), are investigated (Shayan and Mirzayi, 2015). The effect of the structural differences between these forms and their effect on the asphaltene adsorption, as well as the possibility of formation of monolayers or multilayers on the nanoparticle surfaces, were elucidated (Abu Tarboush and Husein, 2015). In particular, the self-association of asphaltenes and multilayer adsorption of nanoparticles were confirmed. The optimum value of the nanoparticle dose (mass of nanoparticles/volume of solution) was found to be of 10 g/L. The adsorption includes two steps: the fast and the slow steps, adsorbing the greatest amount of asphaltenes in the first hour (90%). Crystalline iron(III) oxide nanoparticles (10 nm) were synthesized (Mirzayi and Naghdi Shayan, 2014) with maghemite structure, showing that their oxidative capacities on asphaltenes start at 203 °C. Their resulting adsorption capacity was found to be 9.09 mg/g, similar to the reported above. These nanoparticles could be used as catalysts to improve the process of refining asphaltenes (Nassar, 2010). Behavior of asphaltenes on the surface of Fe2O3 nanoparticles were also studied by Mejía et al. (2012), revealing the formation of asphaltene aggregates of 10–20 nm in size.

For the iron(II,III) oxide Fe3O4 nanoparticles, an unusual strategy was applied based on their injection in the CO2 flow in order to reduce interfacial tension (Kazemzadeh et al., 2015). Solutions of n-heptane and toluene with two types of asphaltenes were used. It was shown that the higher mass fraction of Fe3O4nanoparticles the lower the intensity of asphaltene precipitation for the mass fractions attempted. The iron system promotes precipitation of asphaltenesindependently on the pressure (Khoudiakov et al., 2005). The oil (with 5% of sulfur content) in these experiments contained heavy hydrocarbons in its composition, possessing high affinity with iron oxide. Fe3O4 nanoparticles were unable to eliminate the asphaltenes completely. Adsorption of the studied types of asphaltenes on the nanoparticle surface was found to be enhanced by the low ratio of H/C and high nitrogen content. In a related report (Nassar and Hassan, 2012), it was shown that Fe3O4 nanoparticles can serve as an excellent adsorbent/catalyst for heavy oil upgrading.

6.2. Nickel-containing nanomaterials

In addition to iron oxide, certain attention is paid to nickel-based nanomaterials. Thus, the mixture of cobalt and molybdenum supported on alumina, was prepared by pyrolysis (Choi et al., 2004), resulting particles with a diameter in the range from 10 to 20 nm. This composite does not have a great capacity for adsorption of asphaltenes, but it is useful as a selectively catalyst in processes with sulfur-containing compounds, as well as in thermal cracking (Carbognani et al., 2008). Nickel and palladium composites in the form of nanoparticles are also known (Franco et al., 2015) and used as bifunctional, taking into account the adsorption capacity of nickel and catalytic capabilities of palladium. Main idea of this work was the evaluation of the rate of adsorption/oxidation and their best conditions. The adsorption capacities for these composites were the following: silice (0.23 mg/m2), Ni silice (0.45 mg/m2), silice palladium (0.43 mg/m2), silice Pd 0.66 Ni 0.66 (0.46 mg/m2), silice with 2% of palladium (0.54 mg/m2), silice with nickel (2%) (0.46 mg/m2). The advantage of the nanoparticles of this type is its ability to disperse and stabilize the aggregates of asphaltenes (Guzman et al., 2016). Avoiding aggregates is the result of the adsorption capacity of the nanoparticles which prevents their flocculation and simultaneous precipitation when the asphaltene content is large. In addition, several methods are being investigated using these metals for in situapplications in deep reservoirs, mostly containing heavy oil and asphaltenes, to reduce viscosity. The microemulsions containing metal nanoparticles of W, Ni, and Mo in the form of ultradispersed colloids (Hashemi et al., 2013). They can even traversing the porous material thus helping to heavy hydrocarbons not to be trapped in the rocks.

The presence of nickel was found to promote multi-layer formation (Franco et al., 2013a, Franco et al., 2013b, Franco et al., 2013c). This fact could be considered as a great help for asphaltene inhibition and removal. It was found (Abu Tarboush and Husein, 2012) that the amount of adsorbed asphaltenes on nickel oxidenanoparticles is 2.8 g of asphaltene per gram of NiO for 2 h. In addition to treating nickel oxide as it is, its adsorption capacities could be improved by generating composites on a matrix of silicon oxide. The following adsorption capacities were observed: 5% Ni silice (322 mg/g) and Ni 15% silice (384 mg/g) at 25 °C. In a related report (Hosseinpour et al., 2014), adsorption of asphaltenes on NiO and/or PdO supported on fumed silice was found to be much higher than for fumed silica in the range of tested concentrations. Nanoparticles significantly decreased decomposition temperature of asphaltenes; as a consequence, use of these nanoparticles significantly improves the thermal decomposition of asphaltenes, decreasing viscosity of heavy oils in situ. In case of other supporting materials, the adsorption capacity of the composite of nickel oxide particles on alumina was found to achieve high efficiency in only 2 min (Franco et al., 2013a, Franco et al., 2013b, Franco et al., 2013c). This nanomaterial could serve as a commercial adsorbent. In particular, it was noted that the nickel oxide in the presence of carbon dioxide in porous media favors adsorption kinetics of alumina.

Comparing the ability of nickel oxide in inert substrates above (alumina and silicon dioxide) in similar conditions, interesting results were obtained. Thus, using silica gel (amorphous), silica gel (commercial), nanoparticles of composite nickel alumina 15% (AlNi15), nanoparticles of alumina with 15% of nickel oxide NiO (AlNi5), nanoparticles of alumina with 5% of nickel oxide (AlNi5), nanoparticles of silice with 15% of nickel oxide l5% (SNi15), nanoparticles of silice with 5% of nickel oxide (SNi5), aluminum oxide, silica gel (crystalline), the following order for the asphaltene adsorptivity was observed: AlNi15 > SNi15 > SNi5 > AlNi5 ≈ silica gel (amorphous) > silica gel (crystalline) > zeolite > alumina > silica gel (commercial). A complete asphaltene adsorption in these materials especially of nickel composites was quickly reached, making this material a candidate for inhibiting the precipitation and deposition of asphaltenes. However, for more concentrated environments, it is possible multilayer-type adsorption. Nanoparticles adsorbing strongly more polar compounds are capable of neutralizing the polar forces that remain active during the weak adsorption between asphaltenes (Iman and Babak, 2015). The tendency to form multiple layers seems to be depended primarily on the concentration and nature of the adsorbent.

6.3. Mixed oxides

Recently studied some mixed metal oxides (MMO) on the basis of aluminum, manganese and copper were studied in combination with graphene oxide (GO). Adsorption capacity of graphene oxide was recently observed (Menzel et al., 2016). It was shown that the graphene oxide may be used as an excellent support because of its geometry and other outstanding well-known characteristics, being a hybrid material with high adsorption capacity. It seems to be more selective to heavy thiophenic hydrocarbon compounds. Its addition improves by 5% the adsorption capacity of the crude oil system, but its strongest impact arose due to the adsorption of thiophenic organosulfur compoundsincreasing up 170%. Graphene oxide can separate adsorbent particles generating more active sites, thus being a better remover than nickel to remove sulfur compounds; however, graphene oxide does not present such strong catalytic ability as nickel (Ancheyta et al., 2005). This work shows an importance of graphene oxide among many others, which displays a wide range of novel materials for the removal of heavy sulfur-containing hydrocarbons and which can be applied in different process steps. In addition to pure GO, its metal-containing composites were also studied: MgAl-MMO (0.987 mg S/g of adsorbent), MgAl-MMO/GO (1.449 mg S/gads.), CuAl-MMO (0.209 mg S/gads.), CuAl-MMO/GO (0.598 mg S/gads.), coal-MMO (0.133 mg S/gads.), coal-MMO/S GO (0.344 mg/gads.) (Menzel et al., 2016).

6.4. Comparative studies

To get a better panorama about the influence nanoparticles' nature on asphaltenes, the conditions in which they are measured, need to be recreated. As noted in the sections above, all variables such as pH, temperature, moisture in the system, solvents, oil type, etc. need to be strictly controlled. So, a series of comparative studies of the nanoparticles began to emerge after the year 2010 on this subject, in particular in several reports of Nassar. In one of them, the nanoparticles of three oxides, NiO, Co3O4 and Fe3O4, were found to present a peculiar behavior, being used in solutions with asphaltene concentrations about 200 mg/L. Their adsorption capacities are 22.3 mg/g (for NiO), 12.4 mg/g (for Co3O4) and 15.1 mg/g (for Fe3O4), meanwhile if the concentration of asphaltenes solutions is 2000 mg/L, higher capacities were observed: 60.1 mg/g (for NiO), 63.3 mg/g (for Co3O4) and 62.0 mg/g (for Fe3O4) (Nassar et al., 2011b). The affinity for surface adsorption by nanoparticles is NiO > Co3O4 > Fe2O3(amorphous) (Nassar et al., 2011a). Later on, on the basis of results the capacities for asphaltene oxidation using the same oxides (NiO, Fe3O4 and Co3O4) (Nassar et al., 2011b), it was shown that a correlation exists between the affinity constant and the catalytic activity: the greater the catalytic activity, constant affinity appears.

In another report (Nassar et al., 2011a), the behavior of six different metal oxides (Fe3O4, Co3O4, TiO2, MgO, CaO, and NiO) was comparatively studied. The adsorption type was compared with a Langmuir model, resulting the following order CaO > Co3O4 > Fe3O4 > MgO > NiO > TiO2, using the same composition of asphaltene solutions. It was also established that the adsorption capacities are as follows 2.7 mg/m2 (CaO) > 1.76 mg/m2 (Co3O4) > 1.7 mg/m2(Fe3O4) > 1.35 mg/m2 (MgO) > 0.58 mg/m2 (NiO) > 0.54 mg/m2 (TiO2). As noted, CaO possesses highest adsorption capacity, but it has a small selectivity and adsorbs both heavy and light hydrocarbons, which is not helpful for asphaltenes isolation. This shows that the quality of adsorption and degree of adsorption are not always related. These studies aimed at analyzing the effect of chemical nature of the nanoparticles on the adsorption of asphaltenes in conditions that can be found in the deposits. Therefore, the delay or inhibition of deposition of asphaltenes and transport of nanoparticles were investigated in a porous medium tank at standard pressures and temperatures.

In a recent work (Hosseinpour et al., 2014), three different types of metal salts and oxides were studied (acidic: WO3 and NiO; amphoteric: Fe2O3 and ZrO2; basic: CaCO3 and MgO (Fig. 3)). It was found that the ability of asphaltene adsorption by another series of nanoparticles decreases in the order NiO > Fe2O3 > WO3 > Mn2O3 > CuO > silica Co3O4 (Fig. 4). In this case, we note that nickel oxide has greater adsorptivity than iron compared with that reported above by Nassar. However, this is because Hosseinpour used petroleum coke and not virgin oil, so this allowed the nickel an increasing capacity to generate multilayers by 11%. The catalytic oxidation of asphaltenes by metal oxides moves from 100 to 150 °C to lower temperatures. This latter is reflected if the activation energy for each material, using them as catalyst for asphaltenes oxidation, is observed as Co3O4 < NiO < CuO ≈ Mn2O3 < Fe2O3 < WO3. As it is seen, the higher is the redox activity of metal oxide, the greater the degree of conversion is achieved. In addition, as interesting observation, Nassar concluded that basic oxides such as MgO and CaO, and amphoteric oxides, such as Fe3O4 and Co3O4, demonstrate a greater adsorption capacity compared to acidic oxides, such as TiO2 and NiO (Franco et al., 2013a, Franco et al., 2013b, Franco et al., 2013c). However, the quality of adsorption is higher for basic oxides except CaO, as already mentioned above. The authors suggested that strong interactions present between the asphaltenes and nanoparticles, likely increasing catalytic effect of nanoparticles for various reactions.

1. Download : Download high-res image (612KB)
2. Download : Download full-size image

Fig. 3. HRTEM images of MgO nanoparticles after the asphaltene adsorption at 25 °C and the 2000 mg/L initial asphaltene concentration. Scale bars are (a) 50 and (b) 5 nm. Reproduced with permission of the American Chemical Societyfrom the Hosseinpour et al., 2014.

1. Download : Download high-res image (1016KB)
2. Download : Download full-size image

Fig. 4. FESEM micrographs of (a) NiO, (b) Fe2O3, (c) WO3, (d) MgO, (e) CaCO3, and (f) ZrO2 samples before and after asphaltene adsorption. Left, bare nanoparticles; right, samples after asphaltene adsorption at 25 °C and 2000 mg/L initial asphaltene concentration. Scale bars are 500 nm except in panel e, left (scale bar = 1 μm). Reproduced with permission of the American Chemical Society from the Hosseinpour et al., 2014.

6.5. Use of supporting nanomaterials

The nanomaterials, described above as supports, were also applied alone for asphaltene treatment. Thus, the use of nanoparticles of γ-Al2O3, as typical catalysts/supports, commonly present in heavy oil upgrading, for the adsorption of asphaltenes was reported as a first step in improving heavy oil in situ (Nassar, 2010). It was shown that the effect of the concentration in the adsorption of asphaltenes is smaller in case of the titanium oxide in comparison with other metal oxides such as nickel(II) oxide. This type of technology is dependent largely on the pH in the reservoir (Mohammadi et al., 2011). The titanium dioxide is known to form colloidal dispersions at pH < 4; in basic conditions it can be used without any problem to reducing adsorption and subsequent precipitation of asphaltenes for removal. However, at lower pH nanoparticles of titanium dioxide can serve as a precipitation inhibitor in crude. So, TiO2 can be effectively used to enhance stabilization of asphaltenes, as well as some other materials, such as ZrO2, SiO2.

As noted in this section, the applications of nanomaterials combined with new technologies have great potential to solve main problems caused by asphaltenes. However, in order to compare all these materials, it is necessary to take into consideration the reported adsorption conditions. On the other hand, certain possibilities to misinterpret results exist, as there are a number of factors that were observed which could change in a greater or lesser extent the adsorption capacity of the nanomaterials. These variables are discussed below.