324
The Chemistry and Technology of Petroleum
Jeffries-Harris, 1968). This, of course, is analogous to the occurrence of zwitterion (dipolar ion)
structures in proteins and peptides (Lehninger, 1970).
Thus, it is apparent that asphaltene functionality can be defined in terms of several commonly
occurring natural product types that are more in keeping with the recognized maturation pathways
than any of the structures heretofore conceived.
12.4 MOLECULAR WEIGHT
Perhaps the most confusing aspect of the character of the asphaltene fraction is measurement of
the molecular weight. The asphaltene fraction is a complex composition of functional types, and
molecular weight data represent an average number. Furthermore, there are those who will advocate
(with some justification) in one breath that that asphaltene fraction contains 20,000 constituents. In
the next breath, these same advocates will announce that they know the correct and precise molecular weight of the fraction. Such pronouncement border on the ludicrous and nothing can be further
from the truth!
Determining the molecular weights of asphaltenes is a problem, because they have a low solubility in the liquids often used for determination. Also, adsorbed resins lead to discrepancies in
molecular weight determination, and precipitated asphaltenes should be reprecipitated several
times before the determination. Thus, careful precipitation and careful choice of the determination
method are both very important for obtaining meaningful results (Speight et al., 1985).
The molecular weights of asphaltene fractions span a wide range from a few hundred to several million leading to speculation about self-association (Reerink, 1973; Koots and Speight, 1975;
Speight et al., 1985; Speight, 1994; Galtsev et al., 1995).
The tendency of the asphaltene constituents to form aggregates in hydrocarbon solution is one
of their most characteristic features and complicates the determination of asphaltene molecular
weight (Winniford, 1963; Speight and Moschopedis, 1977). The average molecular weights measured by means of vapor pressure osmometry (VPO) or size exclusion chromatography (SEC)
are significantly influenced by the conditions of the analysis (temperature, asphaltene concentration, solvent polarity) (Moschopedis et al., 1976; Speight and Moschopedis, 1980; Speight, 1981b;
Speight et al., 1985). For this reason, molecular weights are reported in the literature as relative
values only, and these values may be quite different from the molecular weights of unassociated
molecules.
The influence of the asphaltene concentration on the values of molecular weight measured is
significant, and it is important to use as high a dilution as possible in order to measure a molecular
weight that at least approaches that of the unassociated asphaltene molecules, remembering that the
asphaltene fraction is a complex mixture of different species and that any molecular weight of unassociated species will be an average molecular weight of different chemical species.
The existence of asphaltenes aggregates in hydrocarbon solvents has been demonstrated by
means of small-angle neutron scattering (SANS) studies. The physical dimensions and shape of the
aggregates are functions of the solvent used and the temperature of the investigation (Ravey et al.,
1988; Overfield et al. 1989; Thiyagarajan et al., 1995). In addition, surface tension measurements
have been used to study the self-association of asphaltenes in pyridine and nitrobenzene (Sheu
et al., 1992; Storm and Sheu, 1994). A discontinuous transition in the surface tension as a function
of asphaltene concentration was interpreted as the critical asphaltene concentration above which
self-association occurs.
The study of asphaltene molecular weights by vapor pressure osmometry shows that the molecular weights of various asphaltenes are dependent not only on the nature of the solvent but also on
the solution temperature at which the determinations were performed (Moschopedis et al., 1976).
However, data from later work involving molecular weight determinations by the cryoscopic method
(Speight and Moschopedis, 1977) indicate that the molecular nature of asphaltenes is not conducive
to the determination of absolute molecular weights by any one method.
The Chemistry and Technology of Petroleum
Jeffries-Harris, 1968). This, of course, is analogous to the occurrence of zwitterion (dipolar ion)
structures in proteins and peptides (Lehninger, 1970).
Thus, it is apparent that asphaltene functionality can be defined in terms of several commonly
occurring natural product types that are more in keeping with the recognized maturation pathways
than any of the structures heretofore conceived.
12.4 MOLECULAR WEIGHT
Perhaps the most confusing aspect of the character of the asphaltene fraction is measurement of
the molecular weight. The asphaltene fraction is a complex composition of functional types, and
molecular weight data represent an average number. Furthermore, there are those who will advocate
(with some justification) in one breath that that asphaltene fraction contains 20,000 constituents. In
the next breath, these same advocates will announce that they know the correct and precise molecular weight of the fraction. Such pronouncement border on the ludicrous and nothing can be further
from the truth!
Determining the molecular weights of asphaltenes is a problem, because they have a low solubility in the liquids often used for determination. Also, adsorbed resins lead to discrepancies in
molecular weight determination, and precipitated asphaltenes should be reprecipitated several
times before the determination. Thus, careful precipitation and careful choice of the determination
method are both very important for obtaining meaningful results (Speight et al., 1985).
The molecular weights of asphaltene fractions span a wide range from a few hundred to several million leading to speculation about self-association (Reerink, 1973; Koots and Speight, 1975;
Speight et al., 1985; Speight, 1994; Galtsev et al., 1995).
The tendency of the asphaltene constituents to form aggregates in hydrocarbon solution is one
of their most characteristic features and complicates the determination of asphaltene molecular
weight (Winniford, 1963; Speight and Moschopedis, 1977). The average molecular weights measured by means of vapor pressure osmometry (VPO) or size exclusion chromatography (SEC)
are significantly influenced by the conditions of the analysis (temperature, asphaltene concentration, solvent polarity) (Moschopedis et al., 1976; Speight and Moschopedis, 1980; Speight, 1981b;
Speight et al., 1985). For this reason, molecular weights are reported in the literature as relative
values only, and these values may be quite different from the molecular weights of unassociated
molecules.
The influence of the asphaltene concentration on the values of molecular weight measured is
significant, and it is important to use as high a dilution as possible in order to measure a molecular
weight that at least approaches that of the unassociated asphaltene molecules, remembering that the
asphaltene fraction is a complex mixture of different species and that any molecular weight of unassociated species will be an average molecular weight of different chemical species.
The existence of asphaltenes aggregates in hydrocarbon solvents has been demonstrated by
means of small-angle neutron scattering (SANS) studies. The physical dimensions and shape of the
aggregates are functions of the solvent used and the temperature of the investigation (Ravey et al.,
1988; Overfield et al. 1989; Thiyagarajan et al., 1995). In addition, surface tension measurements
have been used to study the self-association of asphaltenes in pyridine and nitrobenzene (Sheu
et al., 1992; Storm and Sheu, 1994). A discontinuous transition in the surface tension as a function
of asphaltene concentration was interpreted as the critical asphaltene concentration above which
self-association occurs.
The study of asphaltene molecular weights by vapor pressure osmometry shows that the molecular weights of various asphaltenes are dependent not only on the nature of the solvent but also on
the solution temperature at which the determinations were performed (Moschopedis et al., 1976).
However, data from later work involving molecular weight determinations by the cryoscopic method
(Speight and Moschopedis, 1977) indicate that the molecular nature of asphaltenes is not conducive
to the determination of absolute molecular weights by any one method.