Zeyu Yang

Chemical Fingerprints of Alberta Oil Sands and Related Petroleum Products

Alberta oil sands are known to contain the worlds largest reserves of bitumen. The rapid growth in their production could result in a significant environmental impact. Fingerprinting bitumen and petroleum products from the Alberta oil sands is essential in order to better understand the chemical compositions of oil sands, prepare for potential oil spills, and address the associated environmental problems. This study presents an integrated quantitative chemical characterization of Alberta oil sands bitumen and other related Alberta oils using gas chromatography-flame ionization detection (GC-FID) and gas chromatography-mass spectrometry (GC-MS). The characterized target hydrocarbons include n-alkanes, unsubstituted polycyclic aromatic hydrocarbons (PAHs) and their alkylated homologues (APAHs), biomarker terpanes and steranes, bicyclic sesquiterpanes, and diamondoids. The chemical features of bitumen in oil sands are clearly distinguishable from those of most other conventional crude oils. The chemical fingerprints of diluted oil sands bitumen and Albian Heavy Synthetic crude were significantly altered by either the diluent blended with the former or the upgrading processing of crude bitumen in the latter. A chromatographic hump of unresolved complex mixtures (UCMs) eluting between n-C10 to n-C40 is pronounced and n-alkanes are nearly absent in bitumen extracted from oil sands. Alkylated naphthalenes account for only a small proportion of the total APAHs in Alberta oil sands extracts. The PAH compounds in oil sands extracts and diluted bitumen are dominated by alkylated homologues with the relative distribution of C0– < C1– < C2– < C3– for all five APAH series. Biomarker terpanes and cage-like adamantanes were determined in almost identical abundance and distribution profile in oil sands extracts and diluted crude bitumen, while biomarker steranes and bicyclic sesquiterpanes were removed to varying degrees by physical weathering or biodegradation.

Petroleum biomarker fingerprinting for oil spill characterization and source identification

Biological markers, or biomarkers, are one of the most important hydrocarbon groups in petroleum used for chemical fingerprinting. Biomarkers can be detected in low quantities (ppm and sub-ppm level) in the presence of a wide variety of other types of petroleum hydrocarbons by the use of the gas chromatography–mass spectrometry (GC–MS). Relative to other hydrocarbon groups such as alkanes and most aromatic compounds, biomarkers are highly resistant to degradation in the environment. The chapter provides analytical methodologies for petroleum biomarker fingerprinting, which include: petroleum biomarker families and labeling and nomenclature of biomarkers. The benchtop quadrupole GC–MS systems, although lacking the high-resolution capabilities of larger and more expensive magnetic-sector instruments, have sufficient sensitivity and selectivity for most purposes of biomarker analysis. In addition to petroleum biomarkers, oil-contaminated sediment samples may contain modern plant biomarker compounds. Chemical fingerprinting of source characteristic and environmentally persistent biomarkers generates information of great importance to environmental forensic investigations in terms of determining the source of spilled oil, differentiating and correlating oils, and monitoring the degradation process and weathering state of oils under a wide variety of conditions.

Development of a methodology for accurate quantitation of alkylated polycyclic aromatic hydrocarbons in petroleum and oil contaminated environmental samples

Polycyclic aromatic hydrocarbons (PAHs) are compounds of concern because most of these compounds are toxic, carcinogenic, or mutagenic and are relatively persistent in the environment. Reliable quantitative information of PAHs is important to evaluate the acute and chronic harmful effects of PAHs on the ecosystem. Crude oils and refined petroleum products contain many highly abundant PAHs and heterocyclic PAHs, in particular the alkylated homologues of naphthalene, phenanthrene, dibenzothiophene, fluorene and chrysene (APAH). The alkylated PAH homologues usually occur in significantly higher concentrations than their corresponding unsubstituted parent PAHs. Petrogenic alkylated PAHs generally consist of large numbers of isomers. Unlike those individual unsusbstituted PAHs, most of the APAH isomers are not commercially available. Therefore, historically, the target APAHs are generally quantified using the relative response factors (RRFs) obtained from their respective unsubstituted parent PAH compounds, this inevitably results in the quantitative results of APAH being significantly underestimated. In order to improve the accuracy of the measurement of PAHs and their alkylated homologues in oil and oil related samples, this study measured and compared the response factors of a large number of alkylated PAHs relative to the internal standards in gas chromatography-mass spectrometry (GC-MS) analysis, with the goal of developing a more accurate quantitative methodology for the determination of oil APAHs, and eventually leading to a standardized methodology of quantitative APAH analysis. The PAHs in different oils and related environmental samples collected from oil impacted areas were determined using this GC-MS methodology. Furthermore, the influence of the measurement methodology on the diagnostic ratios of target PAHs was also assessed in this work.

Application of Light Petroleum Biomarkers for Forensic Characterization and Source Identification of Spilled Light Refined Oils

Light petroleum biomarkers such as bicyclic sesquiterpanes and diamondoids are ubiquitous components of crude oils and ancient sediments, and are also widely found in intermediate petroleum distillates and many finished petroleum products. These compounds are relatively resistant to biodegradation and light-to-medium evaporation weathering, thus particularly useful in oil-source correlation and differentiation for those cases where the traditional tri- to pentacyclic biomarkers are absent. This work utilized sesquiterpanes and diamondoids for fingerprinting and identification of light oils spilled on water. The gas chromatography/flame ionization detection (GC/FID) analysis and distribution profiles of polycyclic aromatic hydrocarbon (PAHs) and conventional biomarkers suggest that the spilled oils are mixtures of mainly gasoline and light diesel type fuel. Since potential source oil candidates were not available, and a large part of the hydrocarbons in gasoline and diesel co-eluted in chromatographic analysis, it is a challenge to quantify the gasoline and diesel in spill samples. It has been known from previous studies that the bulk concentrations of C14 to C16 sesquiterpanes are in the range of approximately 6,000 to 9,000 μg/g for many light diesel fuels, while little or no sesquiterpanes were detected in gasoline, light kerosene and heavy-end lubricating oils. The target sesquiterpanes in the spilled oil samples were determined to be in quite high concentrations: approximately 4,000 μg/g oil. Therefore, it was estimated that these spilled oil samples consist of approximately half gasoline and half light diesel. To verify the estimation, spilled samples were simulated by mixing a fresh gasoline and a light diesel with a similar carbon range as the spilled oils. Results from comparison of GC/FID chromatograms of the spilled oils with the simulated spill samples are consistent with that obtained from sesquiterpane analysis.

Determination of polar impurities in biodiesels using solid-phase extraction and gas chromatography-mass spectrometry.

This paper reports on a method for development and validation for simultaneous characterization and determination of oxygenated polar impurities--free fatty carboxylic acids (FFAs), partial glycerides (monoacylglycerides, MGs), residual glycerol and free sterols--in various biodiesels based on the combination of solid-phase extraction (SPE), silylation and GC/MS technologies. The effects of various SPE and silylation conditions on the method recoveries were evaluated. Using this integrated SPE-GC/MS method, 38 target polar compounds (13 FFAs, 17 glycerides and 8 sterols) in 9 biodiesels derived from 4 different feedstocks were successfully separated and quantified. It was found that the carbon chain length of FFAs was ranged from C(6) to C(24), with C(16) and C(18) being the most abundant in all biodiesels. The total FFAs concentration was consistent with the acid values (AVs) measured by standard method ASTM D974-04. MG congeners with carbon number of 18 (mono-C18) were most abundant in the biodiesel samples, followed by mono-C(16) and free glycerol. β-Sitosterol and campesterol were found to be the prevailing phytosterols in all pure vegetable oil-based biodiesels, while brassicasterol and stigmasterol was only significant in the biodiesel from canola oil and soybean oil, respectively, and abundant cholesterol was only detected in animal fat-based biodiesels.

Oil fingerprinting analysis using commercial solid phase extraction (SPE) cartridge and gas chromatography-mass spectrometry (GC-MS)

This study used solid phase extraction (SPE) cartridges for rapid cleanup and fractionation of oil samples in oil fingerprinting analysis. A series of commercially available florisil cartridges, normal phase SPE cartridges, and silica gel/cyanopropyl (SiO2/C3-CN) SPE cartridges was selected for the fractionation of oil into aliphatic and aromatic hydrocarbons. The florisil cartridges and normal phase SPE cartridges can clean up the oil samples but are unable to separate them into two fractions. The SiO2/C3-CN (1 g/0.5 g) SPE cartridge successfully separated oil samples into aliphatic and aromatic fractions by eluting with 4 mL of hexane and 4 mL of dichloromethylene (DCM)/hexane (3 : 1, v:v), respectively. No cross-elution was observed between aliphatic and aromatic fractions when oil loading mass was less than 40 mg on the SiO2/C3-CN SPE cartridge. The relative standard deviation (RSD) of five replicates of SPE-GC-MS analysis of 5 mg of reference oil is 2.8%, 1.2%, and 6.9% for total n-alkanes, polycyclic aromatic hydrocarbons (PAHs), and biomarkers, respectively. The recoveries of six spiked deuterated surrogates were all above 95%. This SPE-GC-MS method was used for the fingerprinting analysis of various crude oils, refined petroleum products, and environmental sediment samples. The characterized target hydrocarbons included n-alkanes, unsubstituted priority PAHs and alkylated homologues, and biomarker terpanes and steranes. The concentration profiles and diagnostic ratios of target compounds are both comparable to those obtained by the conventional silica gel column-GC-MS method.

Aromatic Steroids in Crude Oils and Petroleum Products and Their Applications in Forensic Oil Spill Identification

Aromatic steroids including monoaromatic (MAS) and triaromatic steroids (TAS) are a series of naphthenoaromatic hydrocarbons, which consist of mixed structures of aromatic and saturated 6-carbon or 5-carbon rings. Although these aromatic steroids are in relatively low concentration in oils, their specific fingerprints and high weathering resistance make them desirable biomarkers for the characterization, correlation, differentiation, and source identification in environmental forensic investigations of oil spills. This study presents a quantitative GC/MS analysis of these aromatic hydrocarbons in a number of crude oils and refined petroleum products including light and mid-range distillate fuels, heavy fuels, and lubricating oils collected from various sources. TAS-cholestanes (C26), TAS-ergostanes (C27), and TAS-stigmastanes (C28) are the most distinguishable triaromatic steroids in most oil samples. C26 TAS-cholestane (20R) and C27 TAS-ergostane (20S) are coeluted and often present as the highest peak in m/z 231 chromatograms. A cluster of aromatic steroids were determined in nearly all the studied crude oils at various concentrations. These compounds are generally not detected in light fuel oil like gasoline and light diesel, but at variable abundance in heavy fuels and lubricating oils. In order to better understand the occurrence of aromatic steroids in refined petroleum products, a crude oil was compared with its products of laboratory fluid catalytic cracking (FCC) and hydrotreating processes. The effects of evaporative weathering and biodegradation on these compounds were evaluated with suites of weathered oil samples.

Method development for fingerprinting of biodiesel blends by solid-phase extraction and gas chromatography-mass spectrometry.

A method based on the combination of solid-phase extraction (SPE) with gas chromatography-mass spectrometry (GC/MS) for detailed chemical fingerprinting of biodiesel/petrodiesel blends was developed in the present study. Forensic identification, commonly referred to as chemical fingerprinting, is based on the relative distributions of individual aliphatic hydrocarbons, aromatic hydrocarbons, fatty acid alkyl esters, and free sterols. Fractionation of fuel samples is optimized for the separation of fatty acid esters and free sterols from petroleum hydrocarbons into four fractions: aliphatic, aromatic, fatty acid ester, and polar components. The final recoveries of aliphatic and aromatic hydrocarbons were determined to be in the range of 65-103%, 73-105% for FAMEs, and 78-103% for free sterols in the polar fraction. Excellent separation with negligible crossover of components with different polarities between fractions was observed. Quantitative analysis of blend levels and individual chemical distribution were achieved. The method has great potential for the identification of biodiesel in diesel fuel blends and could form the basis of a method for characterization of biodiesel-contaminated environmental samples.

Method development for forensic identification of biodiesel based on chemical fingerprints and corresponding diagnostic ratios

A forensic identification method based on the chemical fingerprinting of the first generation of biodiesel (fatty acid alkyl esters as effective components), and several corresponding diagnostic ratios was developed and validated. The distribution of major fatty acid methyl esters (FAMEs) and polar compounds (free fatty acids, glycerol, monoacylglycerides, and free sterols) in several representative above biodiesel products commercially available in Canada were positively quantified and compared, a number of cross-plots of diagnostic ratios of target FAMEs and sterols were developed for biofuel correlation and differentiation. It was found that the cross-plots of FAME ratios, for example, the sum of the di-unsaturated relative to saturated homologues of FAMEs (D/S) versus the sum of the mono-saturated to saturated FAMEs (M/S), and the sum of di-unsaturated to mono-saturated FAMEs (D/M) versus the sum of the mono-saturated to saturated FAMEs (M/S), could cluster samples clearly into their individual feedstock. The cross-plots of diagnostic ratios of individual major sterols (cholesterol, brassicasterol, campesterol, β-stiosterol and stigmasterol) to the total sterols were also developed and proved to be effective in identifying biodiesel sources due to their self-normalizing effect on sterol data. The case study of a mystery biodiesel spill using this method showed that the two real samples can be tightly clustered into biodiesel from animal fat (Ban) group. However, the significant discrepancy of free fatty acids, glycerol, monoacylglycerides and sterol concentrations between the two real samples indicated their different producing batches.

Forensic identification of spilled biodiesel and its blends with petroleum oil based on fingerprinting information

A case study is presented for the forensic identification of several spilled biodiesels and its blends with petroleum oil using integrated forensic oil fingerprinting techniques. The integrated fingerprinting techniques combined SPE with GC/MS for obtaining individual petroleum hydrocarbons (aliphatic hydrocarbons, polyaromatic hydrocarbons and their alkylated derivatives and biomarkers), and biodiesel hydrocarbons (fatty acid methyl esters, free fatty acids, glycerol, monoacylglycerides, and free sterols). HPLC equipped with evaporative scattering laser detector was also used for identifying the compounds that conventional GC/MS could not finish. The three environmental samples (E1, E2, and E3) and one suspected source sample (S2) were dominant with vegetable oil with high acid values and low concentration of fatty acid methyl ester. The suspected source sample S2 was responsible for the three spilled samples although E1 was slightly contaminated by petroleum oil with light hydrocarbons. The suspected source sample S1 exhibited with the high content of glycerol, low content of glycerides, and high polarity, indicating its difference from the other samples. These samples may be the separated byproducts in producing biodiesel. Canola oil source is the most possible feedstock for the three environmental samples and the suspected source sample S2.