Simulations of aerobic biodegradation have been widely employed to investigate the mechanisms of crude oil biodegradation in geological environments. In this study, a simulated biodegradation experiment was performed with crude oil under aerobic conditions, in which n-alkanes were nearly depleted, thus providing an opportunity to study the biodegradation mechanisms of n-alkanes in crude oils. The sequences of biodegraded oils with a slight to moderate degree of biodegradation were characterized by negative-ion electrospray (ESI) Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) and gas chromatography (GC). Semiquantitative results on the molecular compositions of heteroatom classes were obtained by coinjection of internal standards. The biodegradation mechanisms for n-alkanes and n-fatty acids, as well as some other heteroatomic compounds, are discussed. Evidence from FT-ICR MS and GC analyses of biodegraded oils indicates that n-alkanes can be progressively biodegraded to n-fatty acids through β-oxidation, or to hydroxycarboxylic and dicarboxylic acids though ω-oxidation. The O3 class species which have a high relative abundance in the carbon number range of C33–C38 with a Double Bond Equivalent (DBE) of 1–3 were assigned and speculated to be bacterial metabolites, which could be a conspicuous indicator of bacterial activity. Neutral nitrogen compounds, such as carbazoles, exhibited a very slight decrease in the stage of biodegradation that was investigated.

•Flash pyrolysis was employed to release the bound biomarkers from asphaltenes.•Bound biomarkers in asphaltenes show little alteration during biodegradation.•Bound biomarker profiles can represent oils at an early and low maturity stage.•Correlation studies can be made by comparisons of asphaltene pyrolysates.Previous studies have showed that the hydrocarbon biomarkers bound in asphaltenes can preserve original geochemical information during subsequent thermal alteration and biodegradation relative to their free counterparts in crude oil. In the present study, asphaltenes separated from a set of heavy oils which possess an identical source origin and similar thermal maturity, but have experienced different degrees of biodegradation, were analysed to study the influence of biodegradation on the bound biomarkers of asphaltenes. On-line flash pyrolysis directly coupled with gas chromatography–mass spectrometry (Py–GC–MS) was used on these asphaltenes to release the hydrocarbon biomarkers bound within the asphaltene structure. Additionally, the free biomarker profiles in these oils were investigated by GC–MS for a comparison with the bound phase in the corresponding asphaltenes. The results reveal that most of the free biomarkers in the reservoired oils were significantly altered at heavy to severe biodegradation stages, and thus lost much original geochemical information. However, the biomarkers bound in the asphaltenes show very little alteration due to biodegradation, and they also appear to be less mature than their free counterparts in the corresponding oils. This indicates that the bound biomarker profiles in asphaltene flash pyrolysates can represent “original oils” at an early and low thermal maturity stage. Therefore, the biomarker parameters on the basis of asphaltene flash pyrolysates can still be used as indices for correlation studies even though the oils were heavily to severely biodegraded.

The compositions of crude oils can vary significantly during thermal maturation through cracking and aromatization. In this study, an immature high sulfur crude oil was pyrolyzed in a closed gold-tube system with heating rates of 20 °C/h and 2 °C/h, respectively, to simulate the thermal maturation process of crude oil. The molecular compositions of heteroatom-containing compounds in crude oil were investigated by Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) while the impact of thermal maturation on migration and maturity geochemical parameters were investigated by gas chromatography–mass spectrometry (GC-MS). Based on the analysis of pyrolysis products, the compositional variations of heteroatom-containing aromatic compounds and the validity of the aromatic geochemical parameters during thermal maturation were investigated. When the equivalent vitrinite reflectance value (Easy %Ro) was greater than ca. 0.85, alkyl chain cracking was the major reaction and led to the producing of a large amount of ca. 1.18) due to further secondary cracking. Polycyclic aromatic compounds (PAHs) gradually became the dominant compounds in pyrolysis products; the carbon number of alkyl chains attached to aromatic core decreased while the aromaticity increased. Simultaneously, polar heteroatom-containing species became more dealkylated and aromatic with the increasing of maturity. The valid maturity range of geochemical parameters relevant to oil migration and maturity based on heteroatom-containing aromatic compounds were also discussed.

•Py-GC with external standard can be a reliable method to quantify C1–C5 yields.•The yields of C1–5 gases released from various kerogens were studied by Py-GC.•Kerogen types control the compositions of C1–C5 generated by primary cracking.•For type III kerogens, aromatization and condensation can yield > 25% of total C1.Flash pyrolysis (Py) in combination with gas chromatography (GC) and mass spectrometry are often used to elucidate the structure of geomacromolecules such as kerogens, coals, asphaltenes and humic acids by analyzing their low molecular weight pyrolysis products. Most previous work restricts itself to C6+ pyrolysis products with only a few reports on the quantitative measurement of individual C1–C5 gaseous hydrocarbons. This is because of the difficulty in simultaneously separating C1–C5 efficiently. By using a GasPro capillary column to separate C1–C5 gaseous hydrocarbons and using polystyrene as external standard, we explore a quantitative Py-GC flame ionization detector (FID) method to study the C1–C5 pyrolysis products released from kerogens (coals) by primary cracking. Our study indicates that there is a good linear relationship between peak area of C1–C5 on Py-GC FID and kerogen sample weight. Based on the above methods, the yields of C1–C5 gaseous hydrocarbons released from various kerogen types were quantitatively studied. The results indicate that kerogen type plays a key role in controlling the compositions of C1–C5 gaseous hydrocarbons released by kerogen primary cracking under pyrolysis conditions. Type III kerogens (vitrinite rich coals) seem to have more abundant short alkyl chains linked to aromatic nuclei. Sequential flash pyrolysis at 800 °C, 1000 °C and 1200 °C suggests the generation mechanism of C1 is different from that of C2–C5. C2–C5 are mainly generated by the release of alkyl precursors while aromatization and condensation of the kerogen structure may also be an important source of C1 at high maturity, especially for type III kerogens (vitrinite rich coals). It seems to be a reliable way to study the generation mechanisms of C1–C5 by kerogen primary cracking and also the distribution of short alkyl chains within kerogen structure.

•Both the kerogens and O1h solid bitumens are of high maturity and low in EOM.•Biomarkers are released from kerogens and solid bitumen by catalytic hydropyrolysis.•Catalytic hydropyrolysis has higher yield of EOM than Soxhlet extraction.•Those biomarkers can provide dependable information in oil–source correlation.•The O1h solid bitumens are sourced from the Lower Cambrian mudstone.Large amounts of solid bitumen occur in the Honghuayuan Formation of the Lower Ordovician (O1h) within the Majiang paleo-reservoir. Most of the potential source rocks and solid bitumens in the Southern Guizhou Depression are of high maturity (%Ro > 2). Consequently, the yields of extractable organic matter (EOM) from the potential source rocks and solid bitumens are too low to satisfy the requirement of instrumental analysis. Moreover, the reliability of oil–source correlations based on biomarkers in EOM may be questionable because the routine biomarkers obtained by Soxhlet extraction may have been severely altered by secondary alterations. In this research, catalytic hydropyrolysis (HyPy) is used to release covalently bound biomarkers from highly overmature kerogens and solid bitumens. We analyzed the covalently bound biomarkers and n-alkanes released from the O1h solid bitumens and potential source rock kerogens via catalytic hydropyrolysis by using gas chromatography–mass spectrometry and gas chromatography–isotope ratio mass spectrometry to clarify the oil–source correlation of the Majiang paleo-reservoir. Compared with free biomarkers in EOM, the covalently bound biomarkers released by HyPy exhibit low-maturity characteristics and they retain inherited geochemical information. The oil–source correlation indicates that the Lower Cambrian marine mudstones are the main source for the O1h solid bitumens of the Majiang paleo-reservoir. Furthermore, this research also suggests that some of the O1h solid bitumens had suffered severe biodegradation ahead of thermal alteration.

Five heavily biodegraded tar sand bitumens from an oil column were separated into maltene and asphaltene fractions for analysis by negative-ion electrospray (ESI) Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS). These bitumens have an identical source, which have experienced a natural sequence of biodegradation. The polar NSO compounds in maltene fractions contain O1, S1O1, O2, S1O2, S2O2, O3, S1O3, O4, S1O4, N1, N1O1, N1O2, N1S1, and S1 classes, while the polar NSO compounds in asphaltene fractions contain O1, S1O1, S2O1, O2, S1O2, S2O2, O3, S1O3, S2O3, O4, S1O4, S2O4, O5, S1O5, S2O5, O6, N1, N1O1, N1O2, N1O3, N1S1, and N2 classes. Polar NSO compounds with stronger molecular polarity and larger molecular weight are readily fractionated into asphaltene fractions. The O2 class is prevalent in polar NSO compounds of both maltene and asphaltene fractions of all bitumen samples. The N1 class in maltene fractions is dominated by compounds with DBE values of 9, 10, 12 and 13, while the N1 class in asphaltene fractions is dominated by compounds with a DBE of 15. Most of these N1 compounds are likely pyrrolic compounds with various numbers of aromatic rings. The biodegradation pathways of nitrogen-containing compounds are also explored in this study. N1 species are likely converted to N1O1 and N1O2 species following biodegradation pathways such as ring-opening reaction or carbazole dioxygenase (CARDO) catalytic oxidation reaction. S1O2–5 classes are identified as the dominant sulfur-containing compound classes under negative-ion ESI mode. These classes are considered to contain acid functionalities with higher polarity because the sulfur-containing compounds without oxygen are difficult to analyze by negative-ion ESI in which acids can be ionized by deprotonation. Both progressive oxidation and sulfuration may be involved in the anaerobic biodegradation of sulfur-containing acidic compounds.

Solid bitumen occurs extensively in the paleo-reservoirs of marine sequences in southern China. The fluids in these paleo-reservoirs have usually experienced severe secondary alteration such as biodegradation and/or thermal maturation. The concentrations of extractable organic matter (EOM) in the resulting solid bitumens are too low to satisfy the amount required for instrumental analysis such as GC–MS and GC–IRMS. It is also difficult to get enough biomarkers and n-alkanes by dry pyrolysis or hydrous pyrolysis directly because such solid bitumens are hydrogen poor due to high maturities. Catalytic hydropyrolysis (HyPy) can release much more EOM from solid bitumen at mature to highly over-mature stages than Soxhlet extraction, dry pyrolysis and hydrous pyrolysis. However, whether the biomarkers in hydropyrolysates can be used for bitumen-source or bitumen–bitumen correlations has been questionable. In this study, a soft biodegraded solid bitumen sample of low maturity was thermally altered to various maturities in a closed system. HyPy was then employed to release bound biomarkers and n-alkanes. Our results show that the geochemical parameters for source and maturity based on biomarkers released from these thermally altered bitumen residues by HyPy are insensitive to the degree of thermal alteration. Furthermore, the maturity parameters are indicative of lower maturity than bitumen maturation products at a corresponding temperature. This suggests that biomarker source and maturity parameters, based on the products of HyPy, remain valid for bitumens which have suffered both biodegradation and severe thermal maturation. The distributions of δ13C of n-alkanes in hydropyrolysates are also insensitive to the temperature used for bitumen artificial maturation. Hence, the δ13C values of n-alkanes in hydropyrolysates may also provide useful information in bitumen–bitumen correlation for paleo-reservoir solid bitumens.Highlights► A soft solid bitumen was heated to simulate maturation of bitumen in reservoir. ► HyPy was engaged to release biomarkers and n-alkanes from bitumen residues. ► Biomarker parameters in HyPy pyrolysates were more stable than in maturation products. ► δ13C of n-alkanes in HyPy pyrolysates of various maturities are similar. ► HyPy provides a way to study solid bitumen at mature to high-over mature stages.

Seven reservoir core (tar sand) bitumens of identical source and similar maturity from the Liaohe Basin of northeast China possess a natural sequence of increasing severity of biodegradation. This set of samples provides us an opportunity to study the change in oil composition or compound class distributions with biodegradation severity by negative ion electrospray Fourier transform-ion cyclotron resonance mass spectrometry (FT ICR-MS). The bitumen extracts from two columns (Es3 and Es1) were separated into maltene and asphaltene fractions for analysis of heteroatomic species by ESI FT-ICR MS. The maltene fractions were found to mainly contain N1, N1O1, N1O2, N1O3, O1, O2, O3 and O4 classes, while the asphaltene fractions mainly contain N1, N2O1, N1O1, N1O2, N1O3, N1O4, O2, O3, O4 and O5 classes. These species identified by FT-ICR MS in asphaltene fractions are likely to be chemisorbed/coprecipitated compounds, or the species precipitated due to high polarity during deasphaltene process. The susceptibility of compound classes and homologous series to biodegradation was studied based on the relative abundances. The results indicate that microorganisms alter the distribution of acids and nitrogen-containing compounds by selective removal and preservation of certain classes of compounds according to their susceptibility to biodegradation. For example, O2 and N1O2 classes increase significantly while N1 and N1O1 classes decrease with biodegradation. The differences in the susceptibility to microbial alteration within acyclic acids, 4–5 ring acids and 1–2 ring acids are discussed and the differences in the susceptibility of homologous series of heteroatom-containing polycyclic aromatic hydrocarbons are also discussed in this work.Highlights► Tar sand bitumens within PM level 2–8 were separated into maltenes and asphaltenes. ► Both maltenes and asphaltenes were studied by FT-ICR MS. ► The compounds identified in asphaltenes were of high aromaticity and/or polarity. ► Susceptibilities in various classes, and within homologous series were discussed.

Asphaltenes are traditionally considered to be recalcitrant to microbial alteration. Resins and asphaltenes of seven biodegraded oils extracted from reservoir cores of two columns (Es3 and Es1) of the Lengdong oilfield in the Liaohe Basin, NE China, were studied to test this hypothesis. Elemental (C, H, N, O, S) and isotopic compositions (δ13C and δ15N) were measured, FT-IR was used to study the oxygenated functionalities of both resins and asphaltenes, and Py–GC–MS was used to elucidate how alkyl side chains of asphaltenes were altered during biodegradation. We conclude that the products of biodegradation, such as carboxylic acids, phenol and alcohols, may not only contribute to the resin fraction of crude oils, but also are linked with functionalities of resins and asphaltenes. The amount of asphaltenes increases because some resin molecules are enlarged and their polarity increased such that they can be precipitated by hexane as newly generated asphaltenes. Thus, the hydrocarbons that are progressively consumed during biodegradation can pull the δ13C of asphaltene fraction closer to the δ13C of the altered resins and hydrocarbons that were consumed.

•Kerogens and solid bitumens are of high maturity in the Nanpanjiang Depression.•Catalytic hydropyrolysis is used to release biomarkers from kerogens and bitumens.•Bitumen in the north of the depression is sourced from the Permian Source rock.•Bitumens in the middle and midwest are mainly from the middle Devonian source rock.•Oil-source correlation results are consistent with petroleum system analysis.Solid bitumens occur extensively in Permian coral reefs of the Nanpanjiang Depression. Both potential source rocks and solid bitumens in the study area are highly overmature and have similar bulk carbon isotope values. It is difficult to perform an oil–source rock correlation study in this area based on only regular molecular geochemical methods and bulk carbon isotope values. Thus the covalently bound biomarkers released from solid bitumens and source rock kerogens by catalytic hydropyrolysis (HyPy), together with the geological settings, were taken into account in this oil–source rock correlation study. The distribution characteristics of covalently bound biomarkers suggest that the major source rock of the Longlin paleo-reservoir (in the midwest of the depression) solid bitumen should be the Middle Devonian mudstone, whereas the source rock of the Ziyun paleo-reservoir (in the north of the depression) solid bitumen should be the Lower Permian source rock. However, solid bitumens in the Ceheng and Wangmo paleo-reservoirs (in the middle of the depression) may be mainly sourced from the Middle Devonian source rock, but partly from the Permian source rock. Our bitumen–source rock correlation results are also supported by the petroleum geological settings of the study area, which indicate that the filling of those paleo-reservoirs was controlled by the matching of hydrocarbon generation and trap formation. Basically, the timing of hydrocarbon generation of the Middle Devonian source rocks matches well with the formation of Permian coral reef traps in the middle and midwest portions of the depression, but it is earlier than the formation of the Permian coral reef trap in the north of the depression. We show that our oil–source rock correlation study based on covalently bound biomarkers can provide reliable information for petroleum system analysis when highly overmature strata in South China are involved.