3.1. Experiment Process
Most crude oils or asphalts are composed of stable colloids with asphaltene as the core of the dispersed phase and resins as the solvation layer [
17]. Dilution with a large amount of low-molecular-weight n-alkanes can reduce the aromaticity and viscosity of the dispersion system and destroy the colloidal stability, so that the asphaltenes flocculate and precipitate into a separate phase [
18]. To solve problems in previous schemes such as heating required, safety risks and time consumed, etc., this paper designs a simple scheme to extract asphaltenes from asphalt at room temperature, using the solubility difference of asphaltenes in different solvents.
As shown in
Figure 1, the asphalt is dissolved in a solvent to obtain a solution containing asphaltenes at first, and the inorganic substances and residual carbon are removed by filtration. Then, the abovementioned solution containing asphaltenes is dropped into n-alkanes to precipitate asphaltenes and obtain suspensions. Finally, the above suspension is allowed to stand and then filtered, and the filter cake is dried to obtain asphaltene solids. When the standing time was in the range of 30 min to 24 h, no significant change in the yield was found, so standing for 30 min was considered to be sufficient. The drying time usually needs only 4 h at room temperature to achieve constant weight, and if vacuum drying at room temperature is used, it only takes 2 h to achieve constant weight. The room temperature mentioned in this paper is generally defined as 25 °C, but this scheme can still be used to extract asphaltenes within the range of 10 °C above room temperature or 30 °C below room temperature.
As shown in
Figure 2h, the asphaltenes extracted from asphalt with different solvents are all brittle solids with coal-like luster. No apparent adhesion to glass, skin or plastics was found at room temperature. Therefore, the relation between asphaltene content with the adhesion strength and coating ability of asphalt to aggregates may need further studying.
One advantage of this scheme is that it saves time and money. Asphaltene can be separated and extracted at room temperature, avoiding the economic cost of purchasing specialized equipment such as reflux devices, electric heating plates or heating jackets in traditional methods as well as energy costs during the heating and refluxing process. As no heating and cooling processes are required, this method consumes less time compared to traditional methods. Another advantage is the ability to avoid safety hazards caused by heating. Since the n-alkanes used (such as n-heptane) have a low flash point, the organic vapor generated during refluxing may leak and easily catch fire, which constitutes safety hazards to the laboratory and operators.
3.2. Characterization of Extracted Asphaltenes
Elemental and chemical structural characterizations were performed on the asphaltenes extracted from the first group of experiments as well as on the matrix asphalt. The experimental results of elemental analysis in
Table 2 were obtained using an organic elemental analyzer. C and H elements had the largest fractions, with a total amount around 90%, consistent with previous experimental results [
9]. C/H atomic ratio is defined as
q. Cimino et al. analyzed the mass percentages of CHN elements for extracted asphaltenes from refluxing and found that C/H mass ratio was as high as 13.38 for nC
5 and 14.21 for nC
7 solvents [
16]. According to
Table 2, the C/H mass ratio of the extracted asphaltene was 11.34 (
q = 0.94), which is higher than the C/H mass ratio of the base asphalt of 9.40 (
q = 0.78), so the unsaturation degree of the extracted asphaltene is higher than the base asphalt. The
q values calculated according to
Table 2 are listed in
Table 3. Compared with the base asphalt, the content of C and H elements in the extracted asphaltene decreased, while the ratio of N and S elements increased. The N and S elements could introduce polar groups into the asphaltene molecules, increasing their molecular polarity, the aggregation ability, and the coating ability [
19].
The molecular structures of the extracted asphaltenes, as well as the base asphalt, are characterized by NMR in
Figure 3. The hydrogen atomic fractions calculated from
Figure 3 are listed in
Table 3. The aromatic carbon ratio
fA was calculated according to: [
20]
in which,
hA,
hα,
hβ,
hγ are respectively the hydrogen atomic fractions of aromatic hydrocarbons
HA; α aromatic hydrocarbons
Hα; β aromatic hydrocarbons
Hβ, and γ aromatic hydrocarbons
Hγ.
As shown in
Figure 4, the hydrogen atoms of petroleum samples are divided into four categories [
9,
20]: aromatic carbon hydrogens
HA (hydrogen directly attached to aromatic carbons, chemical shift 6.0–9.0 ppm); α aromatic carbon hydrogens
Hα (hydrogen attached to α carbons of aromatic rings, chemical shift 2.0–4.0 ppm); β aromatic carbon hydrogens
Hβ (hydrogen connected to β carbon of aromatic ring and hydrogen on CH
2 and CH farther from β, chemical shift 1.0–2.0 ppm) and γ aromatic carbon hydrogens
Hγ (hydrogen attached to the γ position and the hydrogen on CH
3 farther from the γ position, chemical shift 0.5–1.0 ppm). MestReNova software was used to integrate the absorption peak areas in each component, and the hydrogen atomic fractions for the above four categories were calculated, as shown in
Table 3. Combining this data with a modified Brown–Ladner method, the structures of asphaltenes and their subcomponents can be inferred.
Table 3 shows that the atomic fraction of aromatic carbon hydrogens
HA contained in the extracted asphaltenes is
hA = 0.12, which is higher than that of the asphalt 0.04, indicating that the asphaltenes have more aromatic structures and are more unsaturated than the asphalt. This is consistent with the C/H mass ratio results in
Table 2. For both asphaltene and base asphalt, the proportion of
HA hydrogen atoms is relatively small, and the highest proportion is β carbon hydrogens
Hβ. The aromatic carbon ratio
fA of asphaltenes is around 0.50, indicating that the aromatic ring structure in the molecular structure is well developed, consistent with the previous reports [
20,
21]. Asphaltenes have higher
hA and
hα and smaller
hβ than asphalt, indicating that asphaltenes have larger aromatic rings, more carbon chains connected with the aromatic ring but shorter side chains than asphalt, consistent with the previous reports [
9,
22]. The above results are consistent with the elemental and structural characteristics of asphaltenes, which proves that the extracted materials are asphaltenes.
3.3. Choice of Solvent and Its Effect on Asphaltene Yield
Solvents that can dissolve asphaltenes include aromatic hydrocarbons, carbon disulfide and some halogen-containing organic solvents [
23]. It is generally believed that asphaltenes have archipelagic and continental molecular structures with alkanes and aromatic parts linked by chemical bonds [
24]. Asphaltenes have a high degree of unsaturation. Toluene and xylene have similar molecular structures to asphaltenes, so they are used as good solvents in this paper. According to
Table 4, the C/H mass ratios of toluene and xylene are 10.51 and 9.61, respectively, and their atomic ratios are 0.88 and 0.80, respectively. Therefore, toluene has a higher degree of unsaturation than xylene and greater solubility for asphaltenes. However, after being dropped into the poor solvent, the residual toluene has a stronger inhibitory effect on the precipitation of asphaltenes, thus causing more asphaltenes to remain in the solution, and resulting in a lower yield of asphaltenes as shown in
Table 5.
When the poor solvents are n-heptane, the yield of asphaltenes with xylene is 13.2 ± 0.4% while that with toluene is 11.3 ± 0.4%. Guo et al. also extracted asphaltenes at room temperature by using n-heptane as poor solvent without using good solvent [
9]. Their asphaltene yield was 17.5 ± 1.5%, higher than the yields in
Table 5. According to Sakib and Bhasin’s results [
15], the yield of extracted asphaltenes at room temperature was in range of 15.9–26.6%, which was also higher than yields in
Table 5. These higher yields may arise from differences in the source of asphalt or from absence of good solvent. As good solvent was not used, the solubility for asphaltenes would be reduced and the yield should rise.
The poor solvent should have little solubility to asphaltenes. In this paper, three organic solvents, n-pentane, n-hexane and n-heptane, are used as poor solvents. According to
Table 4, from n-pentane to n-heptane, the C/H mass ratio and the C/H atomic number ratio increase sequentially. The discrepancies decrease between these solvents and asphaltenes, as asphaltenes have C/H mass ratio 11.34 and the C/H atomic number ratio 0.94. Therefore, the ability to flocculate asphaltenes increases with the carbon chain length. When all good solvents are xylene, as in
Table 5, the yields of asphaltenes with alkane carbon numbers of five to seven decrease from 16.8 ± 0.4% to 13.2 ± 0.4%. Previous studies have also found that the precipitation ability of n-pentane to asphaltenes is stronger than that of n-heptane [
25]. Some asphaltenes can be dissolved in n-heptane but not in n-pentane. This part of asphaltenes has a lower molecular weight and a higher degree of saturation, does not form a stacking structure, and its molecular structure is closer to resins [
25].
Compared with toluene, the yield of xylene is higher, and xylene is cheap and not a controlled solvent, so it is more suitable to use as a good solvent for extracting asphaltenes. Compared with n-pentane and n-hexane, n-heptane has a higher boiling point and is easier to recover. We therefore recovered the solvent from the first group of experiments and used it directly for the re-extraction of asphaltenes. The yields are listed in
Table 6. A high asphaltene yield of 10.8 ± 0.4% can still be obtained by utilizing the recovered solvent in the two-time extraction. Thus, the solvents used in previous extractions can be recycled to a certain extent, reducing the extraction cost of asphaltenes. With increasing extraction times, however, the yield of asphaltenes decreased gradually from 13.2 ± 0.4% to 8.3 ± 0.4%. This is because xylene and n-heptane can form an azeotrope and can be distilled out together. With the increase of extraction times, the purity of n-heptane and xylene decreases. Since the poor solvent n-heptane has more volume, the xylene dissolved in n-heptane can hinder the precipitation of partial asphaltenes and reduce the yield of extracted asphaltenes.
The simple method for extracting asphaltenes from asphalt proposed in this paper can be used for extracting asphaltenes from asphalt in laboratories or industry. It avoids the risk of fire and explosion caused by heating and refluxing. As shown in
Table 7, it has the advantages of simple and fast operation, lower energy consumption and more safety, compared to previous methods with heating or refluxing procedures [
12,
13]. Compared to other extraction methods at room temperature [
9,
14,
15], the method proposed in this study could remove inorganic impurity and residual carbons from asphaltenes by the first filtering in
Figure 1, resulting in a purer asphaltene product.
One drawback of our scheme is that good solvent is needed, and the separation between good and poor solvent is still energy-consuming. Fortunately, the solvent can be recovered and reused to lower partial costs for asphaltenes. To increase asphaltene yield, further research is needed to improve the separation efficiency between good and poor solvents. When the amounts of inorganic impurity and carbon residues in asphalt sources are acceptably small, the use of good solvent could be omitted and asphaltene can be extracted by using poor solvent only. The flowchart would then change from
Figure 1 to
Figure 5. The proportion of inorganic impurity and carbon residues of the asphalt in this study is only around 0.1% wt. Following the flowchart in
Figure 5, 2.5 g asphalt and 250 mL n-heptane were stirred for 1 h and then filtered and dried. The asphaltene yield was 19.7 ± 0.4%, higher than the yields in
Table 5 and
Table 6. As good solvent causes some asphaltenes to remain in the solution, some asphaltenes remain rather than flocculate, thereby reducing the yield. For other sources of asphalt, a procedure following the flowchart in
Figure 1 is suggested, unless the amount of the inorganic impurity and residual carbons are identified and sufficiently small.