Energy & Fuels 1999, 13, 287-296
287
Structure and Reactivity of Petroleum-Derived Asphaltene† Levent Artok, Yan Su, Yoshihisa Hirose, Masahiro Hosokawa, Satoru Murata, and Masakatsu Nomura* Department of Applied Chemistry, Faculty of Engineering, Osaka University, Suita, Osaka 565-0871, Japan Received October 9, 1998. Revised Manuscript Received December 4, 1998
The structural characteristics of a pentane-insoluble asphaltene isolated from the vacuum residue of an Arabian crude mixture have been investigated by pyrolysis gas chromatography/ mass spectrometry (py/GC/MS), 1H/13C NMR, gel permeation chromatography (GPC), and matrixassisted laser desorption/ionization time-of-flight (MALDI TOF) mass spectrometry. Assignments of NMR signals of the asphaltene have been discussed briefly on the basis of the information from the literature and compared with those of an aliphatic fraction isolated from the ruthenium ion catalyzed oxidation products of the asphaltene. The comparison data indicated that aliphatic substitution 〉C1 on aromatics are little; however, most of the chain methylene groups are located within a polymeric-saturated fraction of the asphaltene. The average size of aromatic fused ring systems has been determined to be 4-5 for the sample. Pyrolysis tests implied that the asphaltene sample is constructed with relatively large polycyclic units connected by relatively strong bonds. Our results also support a view that asphaltene is the mixture of complex polydispersed molecules with large variation of molecular sizes.
Introduction In the petroleum industry, further utilization of distillation end points (i.e., residua) is of high interest because petroleum refineries will have to deal with much heavier crudes in the future decades. Petroleum asphaltenes, which are operationally defined as pentane- or heptane-insoluble toluene-soluble organic material of crude oil or the bottoms from a vacuum still, are the heaviest fraction of the crude oil, and their amounts and structures are known to be source dependent. In upgrading processes of residua, asphaltenes are responsible for sludge formation due to their flocculation, which reduces the flow and plugs down stream separators, exchangers, and towers. They are also bad actors in poisoning and reducing the activity of hydrocracking catalysts with their high heteroatom content, trace metals, and high tendency to coke formation.1-3 The coke material, which may form in thermal upgrading processes of residua, such as visbreaking and delayed coking, is a much less valuable byproduct that limits efficient conversion to distillable products. Wiehe postulated a mechanism of coke formation which proposes that asphaltenes progressively † The paper was presented at the Symposium on Advances in the Chemistry of Asphaltene and Related Substances, 5th North American Chemical Congress, Cancun, Mexico, November 11-15, 1997, and part of the work has been submitted to Sekiyu Gakkaishi. (1) Speight, J. G. Fuel Science and Technology Handbook; Marcel Dekker: New York, 1990; p 143. (2) Speight, J. G. The Desulfurization of Heavy Oils and Residua; Marcel Dekker: New York, 1981; pp 145-170. (3) Bartholomew, C. H. In Catalytic Hydroprocessing of Petroleum and Distillates; Oballa, M. C., Shih, S. S., Eds.; Marcel Dekker: New York, 1994; pp 1-32.
aggregate and polymerize to large units called “asphaltene cores”, and then, asphaltene cores undergo the liquid-liquid phase separation and lead to polymerization reactions.4 Storm et al. suggested that the small micelles flocculate and grow to the larger particles, and the resulting particles polymerize into macroscopic coke particles.5 In order to develop more efficient conversion processes and overcome their problematic issues, the role of asphaltenic materials in these upgrading processes should also be interpreted at a molecular level. In these circumstances, a better comprehension of asphaltene structure is essential. Although enormous amount of effort has been spent on the structural elucidation of asphaltenes for several decades, their precise molecular description does not exist yet. On the basis of detailed NMR work along with complementary information from various analytical techniques employed, many researchers have concluded that asphaltenes are the mixture of polydispersedcondensed polyaromatic units, with heteroatoms contents, bearing alicyclic sites, and substituted and connected with each other via aliphatic chains. In their researches asphaltenes were precipitated from either the crude sample or residue. The latter type asphaltene structurally may be different from the former type because at distillation temperatures, in general 300500 °C, some extent of cracking and condensation reactions may take place simultaneously. (4) Wiehe, I. A. Ind. Eng. Chem. Res. 1993, 32, 2447-2454. (5) Storm, D. A.; Barresi, R. J.; Sheu, E. Y.; Bhattacharya, A. K.; DeRosa, T. F. Energy Fuels 1998, 12, 120-128.
10.1021/ef980216a CCC: $18.00 © 1999 American Chemical Society Published on Web 02/02/1999
288 Energy & Fuels, Vol. 13, No. 2, 1999
There are a number of studies which have postulated chemical models for asphaltenes, the most recent ones being based on the 1H/13C NMR data and elemental composition. The models, in general, consist of one or two units of polyaromatic units of varying condensation degree combined with alicyclic sites and connected by aliphatic chains, most of the aliphatic chains being attached to the aromatic carbons.6-12 Heteroatoms are incorporated to the structure as either internal constituent of the aromatic rings or functionalities. Although NMR technique would provide invaluable parameters, such as aromaticity, the degree of aromatic condensation and substitution, and aliphatic carbon distribution, some other methods are needed to perceive the type and distribution of the individual molecular segments constituting asphaltene molecules. Some thermal, reductive, or oxidative degradation methods present more exact information regarding the covalently bonded molecular units, aliphatic substituents, and bridge structures. Some researchers have used degradative methods such as pyrolysis and oxidation methods to gain more precise insight into the molecular characteristics of asphaltenes. The former method involves formation of smaller fragments and accompanies their identification, the identified components being considered as covalently bonded moieties of asphaltene molecules.13 Strausz et al. were the first group that applied the ruthenium ions catalyzed oxidation (RICO) reaction to asphaltenes to recognize aliphatic types.14 They processed the invaluable information from the RICO reaction along with those from NMR and pyrolysis studies to comprehend the structure of Alberta asphaltenes and consequently proposed a very different model structure: instead of a single condensed aromatic system with a large number of rings, a set of smaller aromatic units, heteroaromatics and naphthenic units with aliphatic substituents linked by aliphatic bridges comprised the structure. Particularly, the presence of relatively polymeric naphthenic and aliphatic sites in this molecule is a striking feature. Another group also postulated an open 3-D representation for Boscan asphaltene.15 The structure of the heavy fraction of crude oil and its conversion to valuable products have also been of our interest. Recently, we applied the ruthenium ions catalyzed oxidation (RICO) reaction on the asphaltene.16 That study has presented detailed information regarding the aliphatic type and their relative abundance in (6) Hirsch, E.; Altgelt, K. H. Anal. Chem. 1970, 42, 1330-1339. (7) Takegami, Y.; Watanabe, Y.; Suzuki, T.; Mitsudo, T.-aki; Itoh, M. Fuel 1980, 59, 253-259. (8) Suzuki, T.; Itoh, M.; Takegami, Y.; Watanabe, Y. Fuel 1982, 61, 402-410. (9) Cristopher, J.; Sarpal, A. S.; Kapur, G. S.; Krishna, A.; Tyagi, B. R.; Jain, M. C.; Jain, S. K.; Bhatnagar, A. K. Fuel 1996, 75, 9991008. (10) Storm, D. A.; Edwards, J. C.; DeCanio, S. J.; Sheu, E. Y. Energy Fuels 1994, 8, 561-566. (11) Ali, L. H.; Al-Ghanman, K. A.; Al-Rawi, J. M. Fuel 1990, 69, 519-521. (12) Rafenomanantsoa, A.; Nicole, D.; Rubini, P.; Lauer, J.-C. Energy Fuels 1994, 8, 618-628. (13) Ritchie, R. G. S.; Roche, R. S.; Steedman, W. Fuel 1979, 58, 523-530. (14) Strausz, O. P.; Mojelsky, T. W.; Lown, E. M. Fuel 1992, 71, 1355-1363, and references therein. (15) Kowalewski, I.; Vandenbroucke, M.; Huc, A. Y.; Taylor, M. J.; Faulon, J. L. Energy Fuels 1996, 10, 97-107. (16) Su, Y.; Artok, L.; Murata, S.; Nomura, M. Energy Fuels 1998, 12, 1265-1271.
Artok et al.
the sample. Moreover, the oxidation reaction has yielded some aromatic polycarboxylic acids which have implied the presence of various types of aromatic units. In this paper, we have processed the information from the NMR work of the asphaltene sample together with data from the RICO reaction of the same asphaltene to elucidate the distribution of the aliphatic carbons more precisely. The detailed analytical information over this sample is summarized within a model structure. Experimental Section According to a procedure described elsewhere,16 the pentaneinsoluble asphaltene sample used in this work was prepared from the propane-insoluble fraction of the vacuum residue of Arabian crude mixture (80% light and 20% medium). The elemental composition of the asphaltene sample is 83.7% C, 7.5% H, 0.84% N, 6.8% S, 0.012% Ni, 0.038% V, on a dry basis and has a H/C atomic ratio of about 1.08. NMR analyses were conducted by a JEOL JNM-GSX-400 spectrometer operating at 400 MHz for 1H NMR and 100 MHz for 13C NMR measurements. The NMR samples were prepared by mixing approximately 100 mg of the sample with 1 mL of CDCl3; tetramethylsilane (TMS) was used as an internal standard. The quantitative 13C NMR measurements were acquired by adding a relaxation agent, chromium trisacetylacetonate (Cr(acac)3, 0.2 M) in inverse-gated decoupling system with a pulse delay of 5 s, acquisition time of 1.088 s and pulse width of 3.3 µs. The distortionless enhancement by polarization transfer (DEPT) spectra were collected for flip angles of 45°, 90°, and 135°. The acquisition time was the same as those for the quantitative carbon runs. A pulse delay of 2 s and a carbonproton coupling constant of 125 Hz were used. The carbon 90° pulse was 10 µs, while proton 90° pulse was 26.3 µs. The GPC tests of the THF or CHCl3 solutions of the asphaltene (0.5 mg/mL and 1.4 mg/mL, respectively) were performed by a Shimadzu system with 1 mL/min flow rate of THF or CHCl3 carrier solvents, respectively, at a UV wavelength of 270 nm. The columns used in these tests were Shodex KF-802 and Shodex AC-802 for THF and CHCl3 carrier solvents, respectively. Standard polystyrene samples were used for the calibration of relationships between molecular weight and retention time. The details of py/GC and py/GC/MS procedure can be found elsewhere.17 Briefly, about 1 mg of sample was placed on a pyrofoil followed by sealing and folding the pyrofoil. The pyrolysis was done at 670 °C for 3 s. The amounts of tar and coke fractions were determined from the weight of the remains on the pyrocell and pyrofoil, respectively. The weight of the volatile fraction was estimated from the difference. The reproducibilities of tar and coke fractions were within 10 and 5%, respectively. MALDI-TOF spectra were obtained by a Voyager RP mass spectrometer of Perspective Biosystems Co. The linear TOF mode was used with an accelerating voltage of 30 kV in positive ion. One microliter of THF solution of the sample with 2.5 µg/mL concentration was applied to target and let it evaporate at atmospheric condition. The experimental details of XPS study of the asphaltene sample were given elsewhere.18
Results and Discussion Molecular Weight Distribution. GPC, VPO, and mass spectrometric techniques are the most commonly (17) Murata, S.; Mori, T.; Murakami, A.; Nomura, M.; Nakamura, K. Energy Fuels 1997, 11, 1188-1193. (18) Nomura, M.; Artok, L.; Murata, S.; Yamamoto, A.; Hama, H.; Gao, H.; Kidena, K. Energy Fuels 1998, 12, 512-523.
Petroleum-Derived Asphaltene
Figure 1. GPC analyses of the asphaltene.
Figure 2. MALDI TOF mass spectrum of the asphaltene.
used ones among the various methods which have been applied to determine molecular weight of asphaltenes and various other fossil fuels. However, none of them seems to provide reliable information because of the various disparities of the methods. In the case of GPC and VPO techniques, association of molecular units, adsorption on column material, and potential lack of appropriate standards are the major impediments, so that values obtained are highly dependent on temperature, dilution, and solvent type,19-22 while incomplete ionization due to complex and polydispersed nature of the material precludes accurate molecular weight distribution by mass spectrometric methods.23 Figure 1 compares the GPC test results of the asphaltene which were obtained with THF and CHCl3 carrier solvents. The GPC charts show a range starting well below 200 Da extending over 100 000 Da. MALDITOF indicates ions higher than 200 Da (Figure 2), this discrepancy confirming the existence of adsorption phenomenon in the case of GPC procedure. With the exclusion of the range below 200 Da, the numberaveraged molecular weight was estimated as 1753 Da with CHCl3, giving a maximum at 2615 Da. In the case of THF carrier solvent, these values shifted to 801 and 657, respectively, indicating less favored association with THF solvent; nevertheless the observation of (19) Moschopedia, S. E.; Fryer, J. F.; Speight, J. G. Fuel 1976, 55, 227-232. (20) Woods, J. R.; Kotlyar, L. S.; Montgomery, D. S.; Sparks, B. D.; Ripmeester, J. A. Fuel Sci. Technol. Int. 1990, 8, 149-171. (21) Chung, K. E.; Anderson, L. L.; Wiser, W. H. Fuel 1979, 58, 847852. (22) Ternan, M.; Rahimi, P. M.; Clugston, D. M.; Dettman, H. D. Energy Fuels 1994, 8, 518-530. (23) Burlingame, A. L.; Boyd, R. K.; Gaskell, S. J. Anal. Chem. 1998, 70, 647R-716R, and references therein.
Energy & Fuels, Vol. 13, No. 2, 1999 289
elution times corresponding to molecular weights
