Comprehensive Chemical Composition of Gas Oil Cuts Using Two

Article pubs.acs.org/EF

Comprehensive Chemical Composition of Gas Oil Cuts Using Two-Dimensional Gas Chromatography with Time-of-Flight Mass Spectrometry and Electrospray Ionization Coupled to Fourier Transform Ion Cyclotron Resonance Mass Spectrometry Bárbara M. F. Á vila,*,† Boniek G. Vaz,*,‡ Ricardo Pereira,§ Alexandre O. Gomes,∥ Rosana C. L. Pereira,∥,⊥ Yuri E. Corilo,⊥ Rosineide C. Simas,⊥ Heliara D. Lopes Nascimento,⊥ Marcos N. Eberlin,*,⊥ and Débora A. Azevedo*,† †

LAGOA-LADETEC, Instituto de Química, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Rio de Janeiro (RJ) 21941-909, Brazil ‡ Instituto de Química, Universidade Federal de Goiás, Goiás, Goiânia (GO) 74001-970, Brazil § Departamento de Geologia, Centro de Tecnologia e Geociências, Universidade Federal de Pernambuco, Recife, Pernambuco (PE) 50740-530, Brazil ∥ Ilha do Fundão, CENPES, Petrobras, Rio de Janeiro, Rio de Janeiro (RJ) 21941-909, Brazil ⊥ Laboratório ThoMSon de Espectrometria de Massas, Instituto de Química, Universidade Estadual de Campinas, Campinas, São Paulo (SP) 13083-970, Brazil ABSTRACT: Seven gas oil (GO) cuts from the same atmospheric petroleum residuum were obtained by molecular distillation at final temperatures of 490.0 and 503.2 °C (medium GO), 522.5 and 549.5 °C (heavy GO), and 583.7, 622.4, and 662.2 °C (extra heavy GO). The detailed chemical composition of these samples was investigated using comprehensive two-dimensional gas chromatography with time-of-flight mass spectrometry (GC × GC−TOFMS) and electrospray ionization ultrahighresolution and -accuracy Fourier transform ion cyclotron resonance mass spectrometry (ESI FT-ICR MS). Many compound classes were identified by GC × GC−TOFMS, such as tri-, tetra-, and pentacyclic terpanes, steranes, and secohopanes; several polycyclic aromatic hydrocarbons (PAHs), such as fluorene, phenanthrene, pyrene, and benzo[g,h,i]perylene; sulfur compounds, e.g., alkylbenzothiophenes, alkyldibenzothiophenes, and alkylbenzonaphthothiophenes; and alkylphenols. With ESI FT-ICR MS in both the positive- and negative-ion modes, many polar components that cannot be directly characterized via GC were identified. These included quinolines, benzoacridines, carbazoles, benzocarbazoles, thiophenequinolines, and also furanacridines and trends in carbon number and unsaturation as measured by the double-bond equivalent number (DBE) were monitored. Comprehensive characterization of the chemical composition was therefore possible using these two powerful and complementary techniques, enabling detailed monitoring of the complex chemical composition of GO cuts and its variation as a function of distillation temperatures.

■

saturated and aromatic compounds.3,4 Recently, speciation of nitrogen- and sulfur-containing compounds in heavy petroleum cuts by high-temperature GC × GC with a nitrogen chemiluminescence detector and a sulfur chemiluminescence detector has been reported.5,6 The petroleum industry is facing increasing demands to convert heavy oil fractions into valuable products. Molecular distillation7,8 of vacuum residues from petroleum has offered an attractive alternative, yielding extra heavy gas oil (EHGO) as an important heavy feedstock. The molecular composition of EHGO has been analyzed using GC × GC coupled to time-offlight mass spectrometry (GC × GC−TOFMS), and a complex mixture was revealed.9,10 More than a technique for group-type identification, GC × GC−TOFMS allowed for individual identification of several saturated9 and aromatic10 compounds,

INTRODUCTION A petrochemical sample can be considered well-defined when the composition of its constituents is comprehensively known. Methods applied to compound classes and individual compound characterization have been mainly based on separation methods, commonly employing gas and liquid chromatography. However, the extreme complexity of the chemical composition of petrochemical samples with a great number of individual components from different classes and contrasting chemical properties has led to limitations of chromatographic techniques to perform proper characterization of these valuable materials. Comprehensive two-dimensional gas chromatography (GC × GC) has been offered as a powerful alternative to solve this limitation for the investigation of light to heavy petrochemical samples.1,2 There are a few studies reporting GC × GC analysis of heavy petrochemical samples. High-temperature GC × GC with flame ionization detection (FID) was applied to the analysis of vacuum gas oil for group-type separation and semi-quantification of © 2012 American Chemical Society

Received: April 13, 2012 Revised: June 27, 2012 Published: July 9, 2012 5069

dx.doi.org/10.1021/ef300631e | Energy Fuels 2012, 26, 5069−5079

Energy & Fuels

Article

Table 1. Saturated Hydrocarbons and Aromatic and Sulfur Compounds Detected in the GO Cuts by GC × GC−TOFMS substances

carbon number

Aromatic Compounds C4- to C20-alkylbenzenes C10−C26 naphthalene and C1- to C13-alkylnaphthalenes C10−C24 C1- to C11-alkyltetrahydronaphthalenes C21−C31 acenaphthlene C12 acenaphthene C12 fluorene and C1- to C5-alkylfluorenes C13−C18 phenanthrene and C1- to C8C14−C22 alkylphenanthrenes anthracene C14 fluoranthene C16 pyrene and C1- to C4-alkylpyrenes C16−C20 benzo[a]anthracene C18 chrysene and C1- to C4-alkylchrysenes C18−C22 benzo[a]pyrene and C1- to C3-alkylbenzo[a] C20−C23 pyrenes benzo[g,h,i]perylene and C1- to C3C22−C25 alkylbenzo[g,h,i]perylenes benzothiophene and C1- to C6C8−C14 alkylbenzothiophenes dibenzothiophene and C1- to C4C12−C16 alkyldibenzothiophenes benzonaphthothiophene and C1- to C3C16−C18 alkylbenzonaphthothiophenes C3- and C9-alkylphenols C10, C15 Saturated Hydrocarbons n-alkanes C12−C40 C8- to C33-alkylcyclohexanes C14−C39 C17- to C31-alkylcyclopentanes C22−C36 tricyclic terpanes C23−C39

MM range (Da) 134−358 128−310 146−286 152 154 166−236 178−290 178 202 202−258 228 228−284 252−294

Figure 2. GC × GC−TOFMS extrated ion chromatogram (m/z 137) for the C490 GO sample. Note the separation achieved for tricyclic terpanes and alkyldecalins. m/z 137 is not the base peak but a common peak for the classes.

276−318 134−218 184−240 234−276 150, 220 170−562 196−546 308−504 318−542

Figure 3. GC × GC−TOFMS extrated ion chromatogram (m/z 137) for the C490 GO sample. Note the separation achieved for tri-, tetra-, and pentacyclic terpanes and secohopanes. m/z 137 is not the base peak but a common fragment for the terpanes.

of these polar components, electrospray ionization (ESI) coupled to Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) offers a powerful technique able to perform direct and highly selective classification.12,13 In the positive-ion mode (ESI+), basic polar heteroatomic compounds, such as pyridines and analogues, are selectively ionized, whereas in the negative-ion mode (ESI−), acidic polar heteroatomic compounds, such as aromatic carboxylic acids, phenols, and carbazoles, are also selectively ionized. ESI FT-ICR MS has become widely used to characterize crude oils and petroleum fractions using the polar components as natural markers for geochemical parameters, such as biodegradation, thermal evolution, and origin.14−19 However, because ESI response of different polar compound classes may change upon changes in sample and/or additive concentrations, experimental relative abundances cannot be directly related to the relative abundances of their precursor neutrals in the original sample. Ultimately, for quantitative purposes in ESI(±), it is necessary to compensate for the relative

Figure 1. GC × GC−TOFMS extrated ion chromatogram (m/z 83) for the C662 GO sample. Note the separation achieved for n-alkanes, alkylcyclopentanes, and alkylcyclohexanes. m/z 83 is not the base peak but a common fragment for these classes.

such as n-alkanes, terpanes, steranes, polycyclic aromatic compounds, and triaromatic steroids, as well as a quantitative analysis of its aromatic composition based on external standards.10 Petrochemical samples are mainly constituted (80−90%) of relatively volatile hydrocarbons suitable for GC analysis, but ca. 5−10% of crude oil composition is made of less volatile, polar components dominated by heteroaromatics,11 which are unsuitable to direct GC analysis. For the detailed characterization 5070


Energy & Fuels

Article

Figure 4. (A) GC × GC−TOFMS chromatograms highlighting the sulfur compounds identified in all samples. Alkylbenzothiophenes are identified by peak bubbles. (B) EI mass spectra of components identified as C3- and C5-alkylbenzothiophenes.

■

ionization responses by spiking the mixture with a standard with known ionization efficiency for all different chemical functionalities present in the mixture. Petroleomic MS approaches have therefore been solely based on trends arising from comparative analysis of biomarker profiles.20−22 The knowledge of the chemical composition of gas oil (GO) samples is crucial for the petrochemical industries, guiding proper treatment and use, as well as actions of environmental concerns. A quite comprehensive petroleomic12 investigation of the chemical composition of GO samples targeting both polar and nonpolar components would therefore be provided by the combined use of both GC × GC−TOFMS and ESI FT-ICR MS. Recently, Pomerantz et al.23 and Juyal et al.24 have successfully used such a powerful combination to characterize crude oils. In this work, we applied this comprehensive petroleomic approach covering both nonpolar and polar components to better understand the evolution of the molecular distillation process and its influence on the GO chemical composition. Seven GO cuts obtained at different final atmospheric temperatures from the same atmospheric petroleum residuum were analyzed. The seven cuts comprise medium, heavy, and extra heavy gas oil samples. Characterization and semi-quantification of saturated hydrocarbons and aromatic compounds were therefore performed by GC × GC−TOFMS, whereas a detailed composition in terms of the more polar components was performed by ultrahigh-resolution and -accuracy ESI(±) FTICR MS.

EXPERIMENTAL SECTION

Samples. Seven GO cuts from the same atmospheric petroleum residuum were obtained by molecular distillation at final temperatures of 490.0, 503.2, 522.5, 549.5, 583.7, 622.4, and 662.2 °C. These cuts comprise medium gas oils (490.0 and 503.2 °C), heavy gas oils (522.5 and 549.5 °C), and extra heavy gas oils (583.7, 622.4, and 662.2 °C). The samples were supplied by CENPES/PDEDS/TAP, Petrobras (Brazil), and named C490, C503, C522, C549, C583, C622, and C662, respectively. The GO samples were fractioned into saturated (n-hexane), aromatic (n-hexane/dichloromethane, 8:2, v/v), and polar compounds (dichloromethane/methanol, 9:1, v/v) by liquid chromatography using activated silica gel (Merck), for the GC × GC analyses. The standard of n-tetracosane-d50 and pyrene-d10 was acquired from Cambridge Isotopes Laboratories (Andovar, MA), being used as an internal standard for semi-quantification of saturated hydrocarbons and aromatic compounds. GC × GC−TOFMS. The GC × GC system used was a Pegasus 4D (Leco, St. Joseph, MI), which consists of an Agilent Technologies 6890 gas chromatograph (Palo Alto, CA) equipped with a secondary oven and a nonmoving quad-jet dual-stage modulator. Data acquisition and processing was carried out using ChromaTOF software, version 4.21 (Leco Corp., St. Joseph, MI). The GC column set consisted of a DB-5 (30 m, 0.25 mm inner diameter, 0.25 μm df) as the first dimension (1D) and a BPX-50 (Austin, TX) (1.3 m, 0.1 mm inner diameter, 0.1 μm df) as the second dimension (2D). The second column was connected to TOFMS by an empty deactivated capillary (0.5 m × 0.25 mm inner diameter). The columns and the empty deactivated capillary were connected by SGE unions using SilTite metal ferrules (Austin, TX) for 0.10−0.25 mm inner diameter GC columns. For the GC conditions used to the saturated fraction analysis, the primary oven temperature program was 70 °C for 1.00 min, ramp at 20 °C min−1 to 170 °C, and then ramp at 2 °C min−1 to 335 °C. 5071


Energy & Fuels

Article

Figure 5. Plots of the n-alkane, alkylcyclopentane, and alkylcyclohexane distributions as a function of the carbon number versus distillation temperature for the seven GO cuts. The secondary oven temperature program had a temperature of 20 °C higher than that of the primary oven temperature program. Helium was used as the carrier gas at a flow rate of 1.5 mL min−1. The modulation period was 6 s with 1.5 s hot pulse duration and a 35 °C modulator temperature offset versus the primary oven temperature. The MS transfer line was held at 280 °C, and TOFMS was operated in the 70 eV electron ionization (EI) mode with a collected mass range of m/z 50−700. The ion source temperature was kept at 230 °C; the detector was operated at 1700 V; and the acquisition rate was 100 spectra s−1.

Aromatic fractions were analyzed as follows: the primary oven temperature program was 70 °C for 1.00 min, ramp at 20 °C min−1 to 170 °C, and then ramp at 2 °C min−1 to 340 °C. The secondary oven temperature program had a temperature of 20 °C higher than that of the primary oven temperature program. Helium was used as the carrier gas at a flow rate of 1.5 mL min−1. The modulation period was 10 s with 2.5 s hot pulse duration and a 30 °C modulator temperature offset versus the primary oven temperature. The MS transfer line was held at 310 °C, and TOFMS was operated in the 70 eV EI mode with 5072


Energy & Fuels

Article

Figure 6. Plots of the alkylbenzothiophene, naphthalene, fluorene, phenanthrene, pyrene, and benzo(g,h,i)perylene distributions as a function of the carbon number versus distillation temperature for the seven GO cuts. basis of a 10:1 signal/noise ratio (S/N). Compound identification was performed by examination and comparison to literature mass spectra, retention time, elution order,25−35 authentic standards, and National Institute of Standards and Technology (NIST) Mass Spectral Database (version 2005). Semi-quantification of identified saturated hydrocarbons and aromatic compounds was achieved using the relation between their peak areas and the concentration of n-tetracosane-d50 and pyrene-d10 as internal standards for the saturated and aromatic fractions, respectively. The standard solutions were added to each fraction in a final volume of 400 μL. The concentration was corrected to the initial GO mass (in milligrams) using the formula: (analyte area/standard area) × (standard mass in nanograms/gas oil mass in milligrams), where standard mass is the standard concentration (nanograms/microliter) × standard volume (in microliters). ESI FT-ICR MS. Samples (approximately 2 mg) were dissolved in 2 mL of toluene, and then 0.5 mL of this solution were transfer to a 1 mL vial and diluted with 0.5 mL of methanol, containing 1% formic acid for ESI(+) or 1% ammonium hydroxide for ESI(−). Solvents and additives were of high-performance liquid chromatography (HPLC) grade, purchased from Sigma-Aldrich, and used as received. General ESI conditions were as follows: capillary voltage of 3.10 kV, flow rate of 5 μL min−1, and tube lens of 148 V for ESI(+) and −100 V for ESI(−). Ultrahigh-resolution MS was performed with a Thermo Scientific 7.2 T electrospray ionization Fourier transform ion cyclotron resonance mass spectrometer (ThermoScientific, Bremen, Germany). A scan range of m/z 200−1000 was used, and 100 microscans were collected in each run. The average resolving power (Rp) was 400 000 at m/z 400, where

Table 2. Classes of Polar Compounds Assigned for the Molecular Formula Given by High-Accuracy ESI FT-ICR MS Mass Measurements substances

carbon number

Positive-Ion Mode C11- to C61-alkylquinolines C20−C70 C0- to C54-alkylbenzoacridines C17−C71 C10- to C35-alkylthiophenequinolines C21−C46 C8- to C41-alkylthiopheneacridines C23−C56 C11- to C41-alkylfuranquinolines C22−C53 C4- to C43-alkylfuranacridines C19−C58 C8- to C47-alkylphenanthrolines C20−C59 C5- to C47-alkyldibenzophenanthrolines C25−C67 Negative-Ion Mode C15- to C40-alkylindoles C23−C48 C6- to C48-alkylcarbazoles C18−C60 C0- to C43-alkyldibenzocarbazoles C20−C63 C0- to C40-alkyltribenzocarbazoles C24−C64 C1- to C46-alkyl-2-phenylindoles C15−C60

MM range (Da) 283−983 229−985 325−675 347−809 323−757 275−821 292−838 350−938 327−677 251−839 267−869 317−877 207−837

a collected mass range of m/z 50−700. The ion source temperature was 230 °C; the detector was operated at 1600 V; and the acquisition rate was 100 spectra s−1. After data acquisition, samples were submitted to a data-processing method, where the individual peaks were automatically detected on the 5073


Energy & Fuels

Article

Figure 7. Representative ESI FT-ICR MS for three GO cuts.

Figure 8. Plots of the relative abundances of representative compound classes obtained by ESI FT-ICR MS analysis for the seven GO cuts. 5074


Energy & Fuels

Article

Figure 9. Plots of relative abundances of the N class as a function of the carbon number detected by ESI(+) for the seven GO cuts. Rp was calculated as M/ΔM50%,36 that is, by the m/z value divided by the peak width at 50% peak height. Time-domain data (ICR signal or transient signal) were acquired for 700 ms. Microscans were co-added using Xcalibur 2.0 software (ThermoScientific, Bremen, Germany). The MM distribution for each sample was first verified by linear quadrupole ion trap (LTQ) analysis to ensure the validity of the molecular-weight distribution based on FT-ICR MS. In addition to external calibration, an internal recalibration was applied to the peak list (using PetroMS software)19,36 prior to final peak assignment. A set of theoretical homologues series for a specific heteroatom class (most abundance class for each ion mode) was selected as internal calibrants because of their presence in all samples, low errors, and high average peak intensities.

Formula Assignment. For each spectrum, automated analysis was used to assign formula to peaks with a S/N > 3. Formula assignment was performed using the compound identification algorithm program developed by Corilo et al.19,36 Allowed elements were 12C, 1H, 16O, 14 N, 32S, and 13C. The maximum allowed formula error was 1 ppm, and the mass limit for empirically assigning elemental formula was 500 Da. Formula above 500 Da were assigned through the detection of homologous series. If no chemical formula matched a m/z value within the allowed error, the peak was not included in the list of elemental formula. For each elemental composition, CcHhNnOoSs, the heteroatom class, type [double-bond equivalents (DBE) = number of rings plus double bonds involving carbon], and carbon number, c, were tabulated for subsequent generation of heteroatom class relative 5075


Energy & Fuels

Article

Figure 10. Normalized distribution of the N2 class with DBE = 12 as a function of the carbon number detected by ESI(+) for three GO cuts. abundance distributions and graphical DBE versus carbon number images.

alkylbenzenes and alkylhydronaphthalenes of high MM and sulfur compound classes besides alkyldibenzonaphtothiophenes were also identified. Chromatographic Separations. The high resolution as well as sensitivity of GC × GC separation and ultrafast TOFMS detection allowed for compound separation, which is not feasible using unidimensional GC with a quadrupole (q) mass analyzer,9 therefore solving many coelutions occurring for these complex mixtures. Thus, GC × GC−TOFMS seems essential for a comprehensive analysis and important separations in GO samples. As a representative example, Figure 1 shows a surface plot GC × GC extracted ion chromatogram (m/z 83 is a common fragment for these three classes but not the base peak) for the C662 sample, which illustrates the proper separation of n-alkanes, alkylcyclopentanes, and alkylcyclohexanes. Figures 2 and 3 also illustrate the efficient separation of alkyldecalins, secohopanes, and tri-, tetra-, and pentacyclic terpanes. Note that all homologous series were wellresolved with better than baseline resolution, therefore allowing individual semi-quantification to these saturated hydrocarbons. Previously, only alkylbenzonaphthothiophenes were detected in Brazilian EHGO samples,10 but as Figure 4 shows, the GC × GC−TOFMS analysis was able to separate and identify classes of alkylated benzothiophenes and dibenzothiophenes. This result increases the knowledge of sulfur compounds present in Brazilian GO. Some sulfur compound classes have been identified in Brazilian oil samples by GC/MS,37 and ESI−MS has been used for its identification.38 GC × GC with detection via

■

RESULTS AND DISCUSSION GC × GC−TOFMS. Table 1 lists selected compound classes identified in saturated and aromatic fractions, showing the carbon number and molecular mass (MM) range achieved. Note that many classes and individual compounds were properly identified in the GO samples by GC × GC−TOFMS. For instance, polycyclic aromatic hydrocarbons (PAHs) ranging from one to six benzenic rings could be detected. The saturated and aromatic fractions showed a molecular composition mainly based on n-alkanes, alkylcyclohexanes, alkylcyclopentanes, alkylated methylcyclohexanes, and alkyldecalins; tri-, tetra-, and pentacyclic terpanes, steranes, and secohopanes; alkylbenzenes, alkylnaphthalenes, and alkylhydronaphthalenes; several PAHs exemplified by fluorene, phenanthrene, pyrene, and benzo[g,h,i]perylene, as well as alkylated PAHs; triaromatic steroids; sulfur compounds, represented by alkylbenzothiophenes, alkyldibenzothiophenes, and alkylbenzonaphthothiophenes; and some oxygenated compounds, such as alkylphenols. For the range of carbon atoms, the saturated fractions were found to be quite similar to EHGO samples analyzed previously,9,10 but a most significant difference is observed by the detection of secohopanes and naphthenic compounds (such as alkylcyclopentanes, alkylcyclohexanes, alkylated methylcyclohexanes, and alkyldecalins). For the composition of aromatics, 5076


Energy & Fuels

Article

Figure 11. Normalized distribution of the O and NO classes with DBE = 5 and 12 as a function of carbon detected by ESI(−) for three GO cuts.

sulfur chemiluminescence and GC × GC−TOFMS have also been used to determine and quantify sulfur compounds in diesel oils, heavy petroleum, and coal tar.6,39,40 The same classes of sulfur compound identified in coal tar extracts were also identified here in the Brazilian GO samples, excepted for the thiophenes and thioxanthenes. Concentration versus Distillation Temperature. The very detailed chemical composition information provided by GC × GC−TOFMS and the comparison between the seven samples allowed for the observation of some important trends concerning the compound concentration as a function of the distillation temperature for the molecular distillation process. These detailed trends can be visualized by each class as illustrated in Figures 5 and 6 for the saturated and aromatic fractions. Alkylbenzenes, alkylnaphthalenes, alkylhydronaphthalenes, n-alkanes, alkylcyclopentanes, and alkylcyclohexanes (Figure 5) showed similar and expected changes in concentration profiles in all seven GO samples. Note the classical Gaussian profiles with substantial shifts of these profiles to higher MM with the increase of the final distillation temperature. These trends therefore provide detailed monitoring at the molecular level of the whole distillation process. For instance, for the PAH classes, the concentration regularly decreases with the distillation temperature, as Figure 6 illustrates for alkylbenzothiophenes, naphthalene, fluorene, phenanthrene, pyrene, and benzo[g,h,i]perylene. For the sequence of non-alkylated PAHs, with a regular increase of aromatic rings, an increase in the concentration from benzene to pyrene was observed (Figure 6).

For C3-alkyl-PAH distribution, higher concentrations were observed to phenanthrenes, pyrenes, and chrysenes in each distillation temperature. ESI FT-ICR MS. Detailed overviews of the composition of the polar components of the seven GO cuts were obtained from ESI(±) FT-ICR MS (Table 2). These assignments were based on the formulas revealed by accurate mass measurements and their calculated unsaturation levels as measured by DBE. Figure 7 shows the broadband ESI(±) FT-ICR MS from three GO cuts: C490, C549, and C662. As expected, spectra comparison shows that the MM distribution along the mass spectra shifts toward a higher MM (m/z) of the polar components as a function of the increasing distillation temperatures in both modes of ionization. The bimodal distributions observed on ESI(−) spectra are due to the presence of benzo- and dibenzocarbazole series (DBE 12 and 15; m/z 200−360), which have huge greater ionization efficiency by ESI compared to other compound classes, especially in this case where, because of the low acid concentration, they are not suffering suppression. The second Gaussian mass distribution is due to highweight compounds from other classes, mainly carboxylic acid and other nitrogen compounds (m/z > 300). Figure 8 shows detailed changes in relative abundances of a representative member of a specific class of polar constituents for all seven samples analyzed, that is, as a function of final temperature cuts. For ESI(+), a specific and representative member of the N, N2, NS, and NO heteroaromatic classes (pyridines and analogues) was monitored. For ESI(−) also, a 5077


Energy & Fuels

Article

(5) Dutriez, T.; Borras, J.; Courtiade, M.; Thiébaut, D.; Dulot, H.; Bertoncini, F.; Hennion, M. C. J. Chromatogr., A 2011, 1218, 3190− 3199. (6) Mahé, L.; Dutriez, T.; Courtiade, M.; Thiébaut, D.; Dulot, H. J. Chromatogr., A 2011, 1218, 534−544. (7) Batistella, C. B.; Sbaite, P.; Wolf Maciel, M. R.; Maciel Filho, R.; Winter, A.; Gomes, A.; Medina, L.; Kunert, R. Proceedings of the 2nd Mercosul Congress on Chemical Engineering and 4th Mercosur Congress on Process Systems Engineering; Costa Verde, Rio de Janeiro (RJ), Brazil, 2005. (8) Winter, A.; Batistella, C. B.; Wolf Maciel, M. R.; Maciel Filho, R.; Medina, L. C. Proceedings of the European Congress of Chemical Engineering (ECCE-6); Copenhagen, Denmark, 2007. (9) Á vila, B. M. F.; Aguiar, A.; Gomes, A. O.; Azevedo, D. A. Org. Geochem. 2010, 41, 863−866. (10) Á vila, B. M. F.; Pereira, R.; Gomes, A. O.; Azevedo, D. A. J. Chromatogr., A 2011, 1218, 3208−3216. (11) Speight, J. G. The Chemistry and Technology of Petroleum, 4th ed.; CRC Press: Boca Raton, FL, 2006; p 945. (12) Marshall, A. G.; Rodgers, R. P. Acc. Chem. Res. 2004, 37, 53−59. (13) Rodgers, R. P.; Schaub, T. M.; Marshall, A. G. Anal. Chem. 2005, 77, 20A−27A. (14) Hughey, C. A.; Rodgers, R. P.; Marshall, A. G.; Walters, C. C.; Qian, K.; Mankiewicz, P. Org. Geochem. 2004, 35, 863−880. (15) Qian, K.; Rodgers, R. P.; Hendrickson, C. L.; Emmett, M. R.; Marshall, A. G. Energy Fuels 2001, 15, 492−498. (16) Qian, K.; Robbins, W. K.; Hughey, C. A.; Cooper, H. J.; Rodgers, R. P.; Marshall, A. G. Energy Fuels 2001, 15, 1505−1511. (17) Hughey, C. A.; Hendrickson, C. L.; Rodgers, R. P.; Marshall, A. G. Energy Fuels 2001, 15, 1186−1193. (18) Hughey, C. A.; Rodgers, R. P.; Marshall, A. G.; Walters, C. C.; Qian, K.; Robbins, W. K. Org. Geochem. 2002, 33, 743−759. (19) Corilo, Y. E.; Vaz, B. G.; Simas, R. C.; Nascimento, H. D. L.; Klitzke, C. F.; Pereira, R. C. L.; Bastos, W. L.; Santos Neto, E. V.; Rodgers, R. P.; Eberlin, M. N. Anal. Chem. 2010, 82, 3990−3996. (20) Hughey, C. A.; Minardi, C. S.; Galasso-Roth, S. A.; Paspalof, G. B.; Mapolelo, M. M.; Rodgers, R. P.; Marshall, A. G.; Ruderman, D. L. Rapid Commun. Mass Spectrom. 2008, 22, 3968−3976. (21) Kim, S.; Stanford, L. A.; Rodgers, R. P.; Marshall, A. G.; Walters, C. C.; Qian, K.; Wenger, L. M.; Mankiewicz, P. Org. Geochem. 2006, 36, 1117−1134. (22) Hughey, C. A.; Galasso, S. A.; Zumberge, J. E. Fuel 2007, 86, 758− 768. (23) Pomerantz, A. E.; Ventura, G. T.; McKenna, A. M.; Cañas, J. A.; Auman, J.; Koerner, K.; Curry, D.; Nelson, R. K.; Reddy, C. M.; Rodgers, R. P.; Marshall, A. G.; Peters, K. E.; Mullins, O. C. Org. Geochem. 2010, 41, 812−821. (24) Juyal, P.; McKenna, A. M.; Yen, A.; Rodgers, R. P.; Reddy, C. M.; Nelson, R. K.; Andrews, A. B.; Atolia, E.; Allenson, S. J.; Mullins, O. C.; Marshall, A. G. Energy Fuels 2011, 25, 172−182. (25) Philp, R. P. Fossil Fuel Biomarkers: Methods in Geochemistry and Geophisics; Elsevier Science Publishers B.V.: Amsterdam, The Netherlands, 1985; Vol. 23, p 294. (26) De Grande, S. M. B.; Aquino Neto, F. R.; Mello, M. R. Org. Geochem. 1993, 20, 1039−1047. (27) Farrimond, P.; Telnaes, N. Org. Geochem. 1996, 25, 165−177. (28) Nytoft, H. P.; Bojesen-Koefoed, J. A.; Christiansen, F. G. Org. Geochem. 2000, 31, 25−39. (29) Nytoft, H. P.; Bojesen-Koefoed, J. A.; Christiansen, F. G.; Fowler, M. G. Org. Geochem. 2002, 33, 1225−1240. (30) Nytoft, H. P.; Lutnaes, B. F.; Johansen, J. E. Org. Geochem. 2006, 37, 772−786. (31) Peters, K. E.; Moldowan, J. M.; McCaffrey, M. A.; Fago, F. J. Org. Geochem. 1996, 24, 765−783. (32) Wang, Z.; Stout, S. A.; Fingas, M. Environ. Forensics 2006, 7, 105− 146. (33) Azevedo, D. A.; Tamanqueira, J. B.; Dias, J. C. M.; Carmo, A. P. B.; Landau, L.; Gonçalves, F. T. T. Fuel 2008, 87, 2122−2130.

representative member of the N (imidazole), O (phenols), O2 (carboxylic acids), and NO heteroaromatic classes was monitored. Note the detailed molecular level view provided by ESI FT-ICR MS on trends for such concentrations with linear increases, linear decreases, or mass distributions. From these trends, it is clear that higher boiling point cuts will contain increased diversity of multi-heteroatomic polar components, as previously reported for crude oil. The changes in profile of selective classes can also be monitored via contour plots as a function of the carbon number versus DBE, as Figure 9 illustrates for the N class in the ESI(+) mode. A pronounced increase in complexity is evident by the spread in both the carbon number and DBE with the increase in final temperature cuts. Detailed compositional changes for a specific class and specific DBE can also be monitored, as illustrated in Figure 10 for the DBE 12 series of the N2 and NS classes, as well as in Figure 11 for the DBE 5 and 12 series of the O and NO classes. Again, the increase in DBE and carbon number is observed as the molecular distillation temperature increases.

■

CONCLUSION The combination of GC × GC−TOFMS and ESI(±) FT-ICR MS techniques provided comprehensive characterization of GO cuts comprising medium, heavy and extra heavy gas oil samples from the same atmospheric petroleum residuum obtained by molecular distillation. Many nonpolar and polar compound classes were identified, such as pentacyclic terpanes, steranes, secohopanes, several PAHs, sulfur compounds, e.g., alkylbenzothiophenes, alkyldibenzothiophenes, and alkylbenzonaphthothiophenes, and alkylphenols, as well as quinolines, benzoacridines, carbazoles, benzocarbazoles, thiophenequinolines, and also furanacridines, and detailed profiles for the changes in composition in terms of class, carbon number, and DBE could be obtained as a function of distillation temperatures. A much more detailed petroleomic characterization was possible using these two powerful and complementary techniques.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (B.M.F.A.); bonigontijo@ yahoo.com.br (B.G.V.); [email protected] (D.A.A.); eberlin@ iqm.unicamp.br (M.N.E.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank CAPES and CNPq (Brazilian research councils) for fellowships and FAPESP, FAPERJ, FUJB, CAPES, CNPq, ANP, and Petrobras for financial support. The authors also acknowledge the editor and reviewers for the comments and suggestions that served to improve the quality of the manuscript.

■

REFERENCES

(1) Blomberg, J.; Schoenmakers, P. J.; Beens, J.; Tijssen, R. J. High Resolut. Chromatogr. 1997, 20, 539−544. (2) Vendeuvre, C.; Bertoncini, F.; Espinat, D.; Thiébaut, D.; Hennion, M. C. J. Chromatogr., A 2005, 1090, 116−125. (3) Dutriez, T.; Courtiade, M.; Thiébaut, D.; Dulot, H.; Bertoncini, F.; Vial, J.; Hennion, M. C. J. Chromatogr., A 2009, 1216, 2905−2912. (4) Dutriez, T.; Courtiade, M.; Thiébaut, D.; Dulot, H.; Hennion, M. C. Fuel 2010, 89, 2338−2345. 5078


Energy & Fuels

Article

(34) Meredith, W.; Snape, C. E.; Carr, A. D.; Nytoft, H. P.; Love, G. D. Org. Geochem. 2008, 39, 1243−1248. (35) Silva, T. F.; Azevedo, D. A.; Rangel, M. D.; Fontes, R. A.; Aquino Neto, F. R. Org. Geochem. 2008, 39, 1249−1257. (36) Corilo, Y. E.; Vaz, B. G.; Simas, R. C.; Nascimento, H. D. L.; Klitzke, C. F.; Pereira, R. C. L.; Bastos, W. L.; Santos Neto, E. V.; Eberlin, M. N. Proceedings of the 58th ASMS on Mass Spectrometry and Applied Topics; Salt Lake City, UT, 2010. (37) Afonso, J. C.; Cardoso, J. N.; Schmal, M. Fuel 1992, 71, 409−415. (38) Rudzinski, W. E.; Zhang, Y.; Luo, X. J. Mass Spectrom. 2003, 38, 167−173. (39) Hua, R.; Li, Y.; Liu, W.; Zheng, J.; Wei, H.; Wang, J.; Lu, X.; Kong, H.; Xu, G. J. Chromatogr., A 2003, 1019, 101−109. (40) Machado, M. E.; Caramão, E. B.; Zini, C. A. J. Chromatogr., A 2011, 1218, 3200−3207.

5079


Comprehensive Chemical Composition of Gas Oil Cuts Using Two

Recommend Documents