298
The Chemistry and Technology of Petroleum
Infrared spectroscopy has also been used extensively in the elucidation of the structural entities
present in asphaltenes and resins (Monin and Pelet, 1984; Tripathi et al., 1984; Hasan et al., 1988).
For example, the bands at about 3030 cm −1 have been assigned to aromatic carbon–hydrogen
stretching and have demonstrated that this band may shift to values as high as 3052 cm −1 with
decreasing ring number. In addition to the carbonyl functions, the infrared spectra of asphaltenes
show a strong carbon–hydrogen band at 2940 cm −1 , a peak characteristic of CH 2 groups at 1470 cm −1 ,
and peaks for methyl groups at 1380 cm −1 . The group of peaks at about 760, 814, and 870 cm −1 are
believed due to substituted aromatic ring structures. The aromatic ring band at 1600 cm −1 is preeminent, while the broad and strong absorption between 1330 and 1110 cm −1 are assigned to oxygencontaining species. Peaks at 1300 and 1135 cm −1 have been attributed to sulfur–oxygen bonds but a
characteristic peak at 1040 cm −1 has not been adequately diagnosed. The band at 725 cm −1 , usually
attributed to methylene chains having four or more units, is often present, but varies in intensity
depending upon the petroleum source. The band at 740 cm −1 has been suggested to arise from alkylbenzenes, the band at approximately 1600 cm −1 has been assigned to aromatic carbon–carbon bonds
(Nagy and Gagnon, 1961; Millson, 1967).
In addition, the aromaticity (1600 cm −1 band) of the asphaltenes is significantly greater than
that of the maltene fraction but, from infrared spectroscopy, increases only slightly with molecular
size. This is in general agreement with the data from elemental analysis and from carbon-13 resonance spectroscopy where the aromaticity approaches an asymptotic limit for decane-precipitated
asphaltenes (Speight, 1994), although the distribution of aromaticity appears to be dependent upon
the functional types within an asphaltene (Gould and Long, 1986).
The first use of infrared spectroscopy in a quantitative approach to the structural analysis of
petroleum occurred in 1956 (Brandes, 1956) in which the distributions of aromatic, naphthenic, and
paraffinic carbon were calculated for a series of gas oils. Following from this, the 1610 and 720 cm −1
infrared absorption bands have also been used to calculate the carbon distribution (Fischer and
Schram, 1959).
The conclusion was that asphaltenes consist of condensed aromatic ring systems substituted by
paraffinic chains, with naphthenic rings also present, but only to a minor extent. However, use of
the 1600 cm −1 band for estimation of weight percent aromatic carbon has, in some instances, led
to values in excess of 100% and any discrepancy may arise from interference by carbonyl groups
attached to the aromatic system.
The application of applied infrared spectroscopic techniques to the elucidation of the structures
of petroleum resins and asphaltenes has illustrated the presence of many structural groups and
their relative positions to one another. These studies also were suggestive of a different type of
ring condensation for petroleum asphaltenes to that found in coal and it was concluded that petroleum asphaltenes were primarily peri-condensed, like phenanthrene, while the coals were katacondensed, like anthracene.
Further development of the method for structural group analysis, using a large number of hydrocarbons, allowed three chemical and two physical properties to be expressed in terms of five structural
groups. The chemical properties included the carbon and hydrogen content as well as the number of
aromatic carbons present per molecule, and the physical properties required for the analysis are the
molar volume and molar refraction. However, it was concluded that the 1600 cm −1 absorption band
was not capable of producing reliable estimates of aromatic carbon in the heavier feedstocks because
the fractions are of such complex composition. The results did indicate that the infrared data were
self-consistent for the estimation of methylene and methyl groups in the bituminous fractions.
Infrared spectroscopy has also been used to identify various functional groups including
hydroxyl (3450 cm −1 ) and carbonyl (1695–1725 cm −1 ) (McKay et al., 1978). In combination with
nitrogen titration data (Jacobson and Gray, 1987), better representations of the structural groups can
be made (Petersen, 1967).
Conventional infrared spectroscopy also yields information about the functional features of various petroleum constituents. For example, infrared spectroscopy aids in the identification of N–H
The Chemistry and Technology of Petroleum
Infrared spectroscopy has also been used extensively in the elucidation of the structural entities
present in asphaltenes and resins (Monin and Pelet, 1984; Tripathi et al., 1984; Hasan et al., 1988).
For example, the bands at about 3030 cm −1 have been assigned to aromatic carbon–hydrogen
stretching and have demonstrated that this band may shift to values as high as 3052 cm −1 with
decreasing ring number. In addition to the carbonyl functions, the infrared spectra of asphaltenes
show a strong carbon–hydrogen band at 2940 cm −1 , a peak characteristic of CH 2 groups at 1470 cm −1 ,
and peaks for methyl groups at 1380 cm −1 . The group of peaks at about 760, 814, and 870 cm −1 are
believed due to substituted aromatic ring structures. The aromatic ring band at 1600 cm −1 is preeminent, while the broad and strong absorption between 1330 and 1110 cm −1 are assigned to oxygencontaining species. Peaks at 1300 and 1135 cm −1 have been attributed to sulfur–oxygen bonds but a
characteristic peak at 1040 cm −1 has not been adequately diagnosed. The band at 725 cm −1 , usually
attributed to methylene chains having four or more units, is often present, but varies in intensity
depending upon the petroleum source. The band at 740 cm −1 has been suggested to arise from alkylbenzenes, the band at approximately 1600 cm −1 has been assigned to aromatic carbon–carbon bonds
(Nagy and Gagnon, 1961; Millson, 1967).
In addition, the aromaticity (1600 cm −1 band) of the asphaltenes is significantly greater than
that of the maltene fraction but, from infrared spectroscopy, increases only slightly with molecular
size. This is in general agreement with the data from elemental analysis and from carbon-13 resonance spectroscopy where the aromaticity approaches an asymptotic limit for decane-precipitated
asphaltenes (Speight, 1994), although the distribution of aromaticity appears to be dependent upon
the functional types within an asphaltene (Gould and Long, 1986).
The first use of infrared spectroscopy in a quantitative approach to the structural analysis of
petroleum occurred in 1956 (Brandes, 1956) in which the distributions of aromatic, naphthenic, and
paraffinic carbon were calculated for a series of gas oils. Following from this, the 1610 and 720 cm −1
infrared absorption bands have also been used to calculate the carbon distribution (Fischer and
Schram, 1959).
The conclusion was that asphaltenes consist of condensed aromatic ring systems substituted by
paraffinic chains, with naphthenic rings also present, but only to a minor extent. However, use of
the 1600 cm −1 band for estimation of weight percent aromatic carbon has, in some instances, led
to values in excess of 100% and any discrepancy may arise from interference by carbonyl groups
attached to the aromatic system.
The application of applied infrared spectroscopic techniques to the elucidation of the structures
of petroleum resins and asphaltenes has illustrated the presence of many structural groups and
their relative positions to one another. These studies also were suggestive of a different type of
ring condensation for petroleum asphaltenes to that found in coal and it was concluded that petroleum asphaltenes were primarily peri-condensed, like phenanthrene, while the coals were katacondensed, like anthracene.
Further development of the method for structural group analysis, using a large number of hydrocarbons, allowed three chemical and two physical properties to be expressed in terms of five structural
groups. The chemical properties included the carbon and hydrogen content as well as the number of
aromatic carbons present per molecule, and the physical properties required for the analysis are the
molar volume and molar refraction. However, it was concluded that the 1600 cm −1 absorption band
was not capable of producing reliable estimates of aromatic carbon in the heavier feedstocks because
the fractions are of such complex composition. The results did indicate that the infrared data were
self-consistent for the estimation of methylene and methyl groups in the bituminous fractions.
Infrared spectroscopy has also been used to identify various functional groups including
hydroxyl (3450 cm −1 ) and carbonyl (1695–1725 cm −1 ) (McKay et al., 1978). In combination with
nitrogen titration data (Jacobson and Gray, 1987), better representations of the structural groups can
be made (Petersen, 1967).
Conventional infrared spectroscopy also yields information about the functional features of various petroleum constituents. For example, infrared spectroscopy aids in the identification of N–H