Subscriber access provided by University of Florida | Smathers Libraries
Article
DOSY-NMR AS AN ALTERNATIVE TECHNIQUE TO IMPROVE ASPHALTENES CHARACTERIZATION Luiz Silvino Chinelatto Júnior, Sonia M. Cabral de Menezes, Hercílio de Angeli Honorato, Marcia Cristina Khalil de Oliveira, and Luiz Carlos do C. Marques Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b02973 • Publication Date (Web): 10 Jan 2018 Downloaded from http://pubs.acs.org on January 10, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
DOSY-NMR AS AN ALTERNATIVE TECHNIQUE TO IMPROVE ASPHALTENES CHARACTERIZATION Luiz Silvino Chinelatto Junior, Sonia M. Cabral de Menezes, Hercílio de Angeli Honorato, Marcia Cristina Khalil de Oliveira*, Luiz Carlos do C. Marques, *
[email protected] PETROBRAS/CENPES, Av. Horácio Macedo, 950, Cidade Universitária, ZIP21941-915, Rio de Janeiro, Brazil. KEYWORDS: asphaltenes, NMR, DOSY
ABSTRACT: An asphaltene sample from a Light Arabian crude oil was characterized using several analytical techniques such as TGA, NMR, FTIR and elemental analysis. The results showed that the sample has a high sulfur content, highly condensed aromatic compounds, and the average molecular structure is predominantly of continental/island type, but the presence of significant amount of methylene carbons in long alkyl chains is clear. A DOSY NMR experiment was carried out to detect the presence of asphaltene aggregates, evaluate their size and shed light on the presence of maltenic occluded species and entrained residual solvent. In the DOSY spectrum some sample signals were attributed to non-asphaltenic molecules showing that it is possible to achieve key features of the sample complexity in “one shot”.
1. Introduction Almost two centuries have passed since Boussingault (appud Nellesteyn)1 carried out the first study ever made study on asphaltenes reported in literature. In this study the term “asphaltenes” was used to name a massive (circa 1.25x103 kg.m-3), sticky, and dark residue accumulated at the bottom of a crude distillation column. ACS Paragon Plus Environment
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 22
Asphaltenes are non-crystalline, brittle and black solids. At high temperatures they can polymerize, and convert into coke or asphaltic residue. Since the beginning of the 20th century, crude oil has continuously gained importance among world´s energy resources, a fact that still prevails nowadays. Likewise, the oil industry has also been concerned with several problems that the complex asphaltenes molecules might cause in the production, transportation, and refining segments, as follows: •
upstream – flocculation and deposition in the reservoir, drilling tunnels, sand control screens, production pipelines, and top-side facilities; increasing of water-in-oil emulsion stability and crude viscosity;
•
midstream - flocculation and deposition in transfer pipelines, pump impellers, and sedimentation in stock tanks; increase of crude viscosity;
•
downstream – formation of sludge, and de-activation of FCC catalysts. Specific processing is necessary to remove asphaltenes from base lube oil loads.
Asphaltenes are an important component of crude oil because they can easily aggregate and affect its rheological properties of the fluid. Their aggregation, however, not only poses interesting scientific questions but also has great economic significance because asphaltene precipitation can clog reservoir formation and production pipelines, thence causing huge production losses. Using a new theoretical model, the porous asphaltene model, Zielinski et al., show that asphaltenes cluster sizes are no bigger than 2.2-4.4 nm in the native crude oils samples studied.2 There has been a world-wide effort concerning the application of several analytical chemistry techniques to characterize asphaltenes. How deep this characterization can be carried out depends on the availability of the lab to possess from low-cost, easy-to-handle analytical tools up to more sophisticated, complex and powerful techniques such as high-resolution nuclear magnetic resonance,3,4 small-angle neutron scattering,5 high resolution mass spectrometry,6 X-ray difractometry,7 tensiometry,8 atomic force microscopy,9 to name a few. Yen et al. after an extensive experimental work concluded that the asphaltenes consist of amphiphilic molecules of limited molecular mass susceptible to self-associate to form aggregates ACS Paragon Plus Environment
2
Page 3 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
containing around 5−6 molecules, which, in turn, would be capable of associating to form clusters, crystallites, or larger particles. Mullins using advanced experimental techniques corroborated Yen’s conclusions.10,11 Yarranton et al.12 have confirmed that asphaltenes effectively associate to form different types of aggregates, depending on their functional groups, concentration, and temperature.
The analytical quantification of asphaltenes in crude oils is straightforward. It just demands the use of a standardized method comprising a sequence of analytical steps, as follows: •
A selected flocculating agent: flocculant, precipitant or bad solvent: a light non-volatile nalkane: C3,…C7 with Hildebrand solubility parameter13 as different as possible from the asphaltenes one (δasph ~ 19.5 mPa0.5).
•
Selection of an optimum flocculating agent to crude volume ratio (vol/vol)
•
Optimization of experimental conditions: length of soaking time with flocculant, experimental temperature, filtration specifications, evaporating apparatus, among others issues.
Under laboratory conditions n-heptane (δc7 = 15.2 mPa0.5) is the precipitant that best meet these specifications, thereby being the standard fluid of IP-143(2000)14, ASTM D-6560,15 ASTM D8938016 and ASTM D2007-8017 methods. In fact, the modus operandus of these methods stems from the different solubility behavior of asphaltenes in a pair of selected solvents with different polarities and Hildebrand solubility parameters. For instance, under given laboratory conditions asphaltenes are insoluble in low molecular weight liquid alkanes, such as (pressure-liquefied) propane, n-heptane; and soluble in low molecular weight aromatic solvents such as toluene (δtol = 18.2 mPa0.5), xylene, and cumene. However, observation of trapped compounds in solvent-extracted fractions from standards asphaltene preparation protocols are abundant in the literature. This numerous protocols may lead to hurdles in the asphaltenes characterization process.18,19 The asphaltenes flocculated and precipitated - by adding to crude a standard volume (40 cm3/g of sample crude) of n-heptane-, show a chemical composition, which is different from asphaltenes precipitated by other precipitants such as propane, pentane or hexane. Hence, the composition of 3 ACS Paragon Plus Environment
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 22
precipitated asphaltenes is dependent on the flocculant composition or the depressurization process that crude is submitted to under actual reservoir conditions. It is important to bear in mind that the amount of asphaltenes in crudes do not hold a straight relationship with the amount of asphaltenes precipitated under actual field conditions nor with any asphaltenes molecular characteristics. For instance, Parra-Barraza et al.
20
carried out a study in
which crude asphaltenes samples were precipitated with the following n-heptane to crude (volumetric) ratios: 5:1, 15:1, and 40:1. According to this study, the percentage of precipitated asphaltenes is different from each other and there are no resins in the solids precipitated from crude when the 40:1 n-heptane to crude ratio is used. This explains why such a volumetric ratio has been standardized in different standard asphaltenes quantification methods. Furthermore, the composition of the asphaltenes obtained by depressurization process differs from the ones precipitated with flocculants in terms of the amount of aromatics which is generally higher in asphaltenes obtained by precipitation with flocculants. As for the asphaltenes, other groups of chemical components can be separated from crudes by equally applying a combination of a pair of selected solvents with different solubility parameters. To obtain a global view of potential characterization methods to describe such complex molecules, a sample of an asphaltene obtained from the same crude oil was separated for an interlaboratory test proposed by 2017 PetroPhase conference.
The asphaltene sample prepared by the group of
Marianny Y. Combariza at the Industrial University of Santander (Colombia) was divided in small portions and distributed among the participant laboratories. Each participant used their own expertise to characterize the sample and to compare the results at a PetroPhase conference in a dedicated session.
PETROBRAS Research & Development Center (CENPES) group used their in-house expertise in Nuclear Magnetic Resonance (NMR) to generate some average structural parameters trying to help in pre-viewing some features of the asphaltene average molecular structure. Diffusion-ordered NMR spectroscopy (DOSY) was applied to try to investigate the average sizes of the aggregates and ACS Paragon Plus Environment
4
Page 5 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
extend sample characterization.21 TGA, FTIR and elemental analysis were also carried out to complement characterization of the received asphaltene sample.
2. Materials and experimental procedures
2.1. Materials. Asphaltene sample was obtained from a Light Arabian crude oil by the Industrial University of Santander (Colombia). This group developed a very efficient method to separate and purify asphaltenes from crudes. The sample was obtained by using ASTM D6560 standard method and afterwards submitted to sonication to decrease up to 75 wt% of asphaltenes coprecipitants. In order to extract any other organic contaminants, successive hydrocarbon Soxhlet extractions were carried out.12
2.2. Elemental analysis. The asphaltene sample was analyzed for C, H, N, O, and S content. C, H, S and N were quantified by combustion followed by chromatography using a CHNS analyzer (Thermo Scientific Instruments Flash 2000 model). The oxygen content in the sample was determined by mass difference, therefore being subject to a higher error than that of the other elements. 2.3. Thermogravimetric analysis. TGA measures the change in the mass of a sample as a function of temperature in a controlled atmosphere. So, TGA measurements were carried out to ascertain the thermal stability of asphaltene sample, and, as well, to provide a preliminary estimate of the sample composition, as well. This technique is suitable to analyze materials that exhibit either mass loss or gain due to decomposition, oxidation or loss of volatiles (such as moisture) within the investigated range of temperature. The TGA analyses were carried out in a TGA analyzer (TA Instruments SDTQ 600 model) both in N2- (20ºC to 700ºC) and air- (700 ºC to 1000 ºC) atmospheres. A heating rate of 20 oC/min was used in both evaluations. 2.4. Fourier-Transform Infrared Spectroscopy (FTIR) was applied to characterize the chemical structure and functionalities of asphaltene sample. Arguably, the key- bands and peaks have been identified and assigned to different hydrocarbon structures and/or functional groups. The sample spectrum was recorded using an Avatar System 360 spectrometer over the range of 4000-650 cm-1 in ACS Paragon Plus Environment
5
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 22
the transmission mode with a resolution of 4 cm-1 with DTGS KBr Detector. Spectra were averaged from the accumulation of 32 scans. The asphaltene sample was analyzed using an attenuated total reflectance (ATR) accessory with one reflection diamond crystal.
2.5. Nuclear Magnetic Resonance (NMR). Solution state 13C NMR spectrum was recorded at 11.75 Tesla on a Bruker Avance III NMR spectrometer, operating at 125.74 MHz, using a 5 mm broad band probe. Sample was dissolved at 8 wt% in a 0.05 M chromium (III) acetylacetonate (as the relaxation agent) solution in deuterated (D) - Chloroform (D, 99.8%; CDCl3). The experiment was performed at 50 °C, in which temperature the sample appeared to be well-dissolved. The following experimental conditions were used: 29.8 KHz spectral width, 10.2 µs (90°) rf pulses, 0.6 s acquisition time, 30.0 s pulse delay and 13683 scans were accumulated. The nuclear Overhauser enhancement was suppressed by operating the spectrometer in the “inverse-gating” mode where the broad band proton-decoupling was turned on only during acquisition periods. Solution state 1H NMR spectrum was recorded at 11.75 Tesla on a Bruker Avance III NMR spectrometer, operating at 500.13 MHz, using a 5 mm broad band probe. Sample was dissolved at 4 wt% in methylene dichloride-D2 (D, 99.8%; CD2Cl2). The experiment was performed at 27 °C. The following experimental conditions were used: 10.0 KHz spectral width, 3.3 µs (30°) rf pulses, 2.5 s acquisition time, 10.0 s pulse delay and 256 scans were accumulated. 2.5.1. 1H Diffusion-Ordered NMR Spectroscopy. For the DOSY experiments, the samples were dissolved at 8 wt% in toluene-d8 (D, 99.5%). The experiment was carried out at 9.40 Tesla on an Agilent 400MR NMR spectrometer operating at 399.8 MHz, equipped with a gradient pulse amplifier (Performa I model), generating a 20 G cm-1 field strength, using a 5 mm direct detection and pulsed field gradient probe. The so-called “Oneshot pulse sequence”
22
was employed to
measure the self-diffusion of the aggregates, using 25 linear steps from 0 to 20 G cm-1, the gradient pulse duration (δ) was 3.5 ms and the diffusion delay (∆) 300.0 ms. The gradients were calibrated according to the manufacturer instructions, using the HOD/D2O (99%) standard solution at 25 ºC. The spectrum was acquired at 25ºC with a relaxation delay of 2.0 s, pulse duration of 14.0 µs (90°), 6 ACS Paragon Plus Environment
Page 7 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
and 256 scans. Data analysis was performed using Mestre Nova 12.0 software, applying the “Peak Heigths Fit” method, with non-linear Levenberg–Marquardt fitting using NUG-corrected decay model. 3. Results and discussion 3.1. Elemental analysis Table 1 presents the elemental analysis and the C/H ratio for the asphaltene sample. The C/H molar ratio reflects the sample aromaticity. The C/H ratio of 0.917 obtained suggests that the sample is moderately aromatic. The sample presents high levels of S (7.9wt%) and a moderate amount of N (1.4wt%). In general, most of S and N found in asphaltenes are present as part of the aromatic core of the average structure in thiophenic, pyrrolic and/or indolic rings.23,24 Although not evaluated in the present work, dissolved metals and metallic nanoparticles like vanadium, nickel, molybdenum and iron, are often found in asphaltene solutions, as part of soluble coordination complexes or forming nanoparticles.25 3.2. Thermogravimetric Analysis. The results of TGA analysis are presented in Figure 1, which depicts the weight loss curves (TG) and derivative thermogravimetric (DTG) evolution profiles respectively, as a function of reaction temperature, for the asphaltene sample. The thermal decomposition of the samples is responsible for the appearance of a major peak around 470 °C in the TGA spectra. This peak is attributed to the pyrolysis of the heaviest fraction of the solid. After the introduction of air at 700ºC, an exothermic weight loss is observed which can be attributed to the burning of coke (51%). This corresponds to the air-combustion of the high molecular weight compounds containing condensed rings. These results are in accordance with the ones usually reported in the literature for asphaltenes from different origins.21,26 The ~3% mass loss under 280ºC could be attributed to maltenic occluded species and/or entrained residual solvent (co-precipitant). 3.3. Fourier-Transform Infrared Spectroscopy. Infrared spectroscopy has been extensively used to characterize the chemical structure and functionalities of asphaltenes molecules and related model compounds.27,28,29 The FTIR spectrum between 4000 and 650 cm−1 for the received asphaltene ACS Paragon Plus Environment
7
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 22
sample is shown in Figure 2. The FTIR spectrum show intense C-H absorption bands arising from straight chain paraffins in the general region of 2840-3000 cm-1, particularly two distinct absorption peaks at 2849 cm-1 (R2CH2- symmetric methylene stretching), and 2918 cm-1 (R2CH2- asymmetric methylene stretching) and a shoulder at 2949 cm-1 (R2CH3 asymmetric methyl stretching). For instance, aliphatic CH groups can also be detected at 1452 and 1374 cm-1 because of asymmetric and symmetric bending. The lack of absorptions in the 1670-1800 cm-1 region is an indication of the absence of carboxylic and carbonyl groups. The (C=C)Ar stretching can be observed at 1695 cm-1. The 1029 cm-1 absorption band indicates the presence of sulfoxide or C-S linkage (S-O sulfoxide stretching). In the 950-700 cm-1 region, bands from out of plane bending out of plane for C-H aromatic bond were observed. The 855 cm-1 absorption band suggests the presence of aromatic rings with one isolated hydrogen, i.e., penta-substituted rings (CAr-H (1H); isolated hydrogen). The 807 cm1
absorption band may be attributed to systems containing two or three adjacent aromatic hydrogens,
i.e., tri- and tetra-substituted rings (CAr-H (2H or 3H)); two or three adjacent hydrogens). The 745 cm1
absorption band may be attributed to systems containing four adjacent aromatic hydrogens, i.e.,
ortho substitution of the aromatic rings (CAr-H (4H); four adjacent hydrogens). Finally, a sharp band responsible for rocking of chains with more than three contiguous methylene groups is detected at 722 cm-1 (R(CH2)n-R when n >3).
3.4. Nuclear Magnetic Resonance (NMR). The molar percentage of chemical compounds obtained by 1H and 13C NMR spectra (Figures 3 and 4) for the received sample are shown in Tables 2 and 3. The NMR spectra can be analyzed by observing carefully the aliphatic region and the aromatic regions according to literature.30,31 By 1H NMR (Table 2) the hydrogens attached to diaromatic rings or bigger (Hdar) can be distinguished from the ones attached to monoaromatic rings (Hmar). Table 2 clear shows that Hdar (6.6 molar%) is bigger than Hmar (4.2 molar%) as expected for an asphaltene molecule in which the core is composed by polyaromatic rings. In the aliphatic region of the spectrum, the %Hbeta-1 (32.5 molar%) is higher than Hbeta-2 (24.4 molar%).17 This an indication of the presence of several naphthenic hydrogens (possibly in naphthenic rings) (see Table 2). The presence
ACS Paragon Plus Environment
8
Page 9 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
of paraffinic hydrogens in alkyl chains attached to aromatic rings are also present, as we still have a significant amount of Hbeta-2 (Table 2). Table 3 shows that the total aromatic carbons (Cartot) are about 55% of total carbons in this asphaltene molecule. The bridged head (Car-b) are the majority of the aromatic carbons confirming the highly condensed aromatic system of this sample. Approximately, equal molar amounts of alkyl substituted (Car-alk) and non-substituted aromatic carbons (Car-H) can also be observed. In the aliphatic region it can be highlighted that branched methyls (Cmet-bran) are more than 3 times abundant than the methyls in terminal positions (Cmet-term). 3.5. 1H Diffusion-Ordered NMR Spectroscopy. DOSY provides a way to separate compounds in a mixture based on the difference translational diffusion coefficients and therefore the molecular size and aggregation states without the need for any physical separation. Diffusion coefficients are also function of physical properties of the surrounding environment such as viscosity, temperature, etc. of each chemical species in solution.
The measurement of diffusion is carried out by observing the attenuation of the NMR signals during a pulsed field gradient experiment. The degree of attenuation is a function of the magnetic gradient strength () and occurs at a rate proportional to the diffusion coefficient (D) of the molecule. The signal decay can be described by the Stejskal-Tanner equation:
= exp − ∆ − ⁄3 −
(1)
Where is the measured signal intensity, is the signal intensity in the absence of diffusion, is the gyromagnetic ratio, the gradient strength, the diffusion coefficient, the length of the gradient pulses, ∆ is the time between the two gradients in the pulse sequence (and hence defines the time for which diffusion is observed), and R is a constant that takes nuclear relaxation into account.
ACS Paragon Plus Environment
9
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 22
is a measure of the translational diffusion behavior of the molecules or aggregates and depends on the temperature, particle size and shape, and the intensity of the intermolecular forces. According to the Stokes-Einstein-Debye equation, is given by:
= η
(2)
Where is the Boltzmann constant, is the temperature, η is the viscosity of the solution, and RH is the hydrodynamic radius of the solute. The experiment is typically performed by varying the gradient strength for several values and collecting the Free Induction Decays, although or ∆ may also be varied to give signal attenuation. The former is the preferred method for mixture analysis because it easily allows for the pulse sequence timing to be constant throughout the experiment. A Fourier transformed data set contains families of resonances that belong to a pure component in the mixture will decay exponentially at the same rate with respect to the square of the gradient area . The resonances are then differentiated based upon this decay rate.32 In order to derive the diffusion constants, an inversion of the Laplace transform may be applied to extract the Gaussian decay constants. In DOSY, the results are displayed as a pseudo 2D spectrum in which signals are dispersed according to diffusion constant in one axis and chemical shift in the other. Therefore, the results remind the user a chromatographic separation, but in a NMR tube, and also provide NMR information that can be used for assignment of individual components. However, it does not require any particular sample preparation, and maintains the original chemical environment of the sample during analysis. It has been used not only for petroleum fractions characterization and to study asphaltene aggregation phenomena,14,33,34,35,36,37 but also for crude oils samples analysis.38,39,40
In order to try to discriminate aggregate sizes and explore occluded compounds in the received sample, 1H-DOSY spectrum has been recorded. Figure 5 shows the 1H DOSY NMR spectrum for the asphaltene sample at 8 wt%. Sample appeared to be well-dissolved in this concentration but in fact the solution was near saturation. This concentration was chosen in order to avoid the dilute concentration regime,41 in which the aggregates would stay non interactive and separated from each ACS Paragon Plus Environment
10
Page 11 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
other. In this way the strong solute-solvent contribution to the aggregation behavior has been mitigated. In fact, in more concentrated solutions, more attractive interactions exist between the clusters, and closer our model system is of the real environment that asphaltenes experiment in their original oil. While F2 dimension (x axis) gives 1H chemical shift information, F1 dimension (y axis) provides diffusion coefficient (D) of the species involved in the study.
The information encoded in DOSY F2 dimension is not exactly the same as reported in Table 2. On the one hand, the signals arising from 0.5% toluene-d7 mixture of isotopomers difficult the quantitative analysis specially for the aromatic signals of the sample. On the other hand, the use of an aromatic solvent may cause a dramatic simplification of the chemical shift analysis due to aromatic solvent-induced shift (ASIS).42 Although ASIS is a very old technique, it is by no means outdated. In Figure 5, chemical shift dimension (9.40 Tesla) shows a shoulder at 1.19 ppm that could not be seen even at 11.75 Tesla using CD2Cl2 as solvent (Figure 3).
In DOSY F1 dimension was observed a log-normal distribution for diffusion that is reasonable because D values are positive-defined and may vary for a large set of values, more than one order of magnitude. In Figure 5 it is possible to detect signals arising from toluene, with average D in the range of 1.31 x 10-5 to 1.75 x 10-5 cm2.s-1. This value is in agreement with the fact that toluene diffusion constants decrease with an increase of the sample concentration and the D value for pure toluene is about 1.96 x 10-5 cm2.s-1.34 Aromatic and aliphatic protons arising from sample molecules have average D value in the range of 5.80 x 10-7 to 2.66 x 10-6 cm2.s-1, enabling a huge separation between toluene solvent and the sample molecules. When D < 2.66 x 10-6 cm2.s-1 signals are related to asphaltene aggregates. In fact, literature data point out that asphaltenes may aggregate to form nano, micro and macro-aggregates in decreasing order of D value.14,35 For concentrated asphaltenes solutions (circa ≥ 8 wt%) structures presenting D ranging from 0.4 x 10-6 to 2.0 x 10-6 cm2.s-1 have been identified as macro-aggregates.14,36 Differences in concentration, structure, molecular weights, modes of aggregation and intermolecular interactions may result in slightly deviations of D values for same size aggregates. So, it is reasonable to believe the D signals depicted in Figure 5, ranging ACS Paragon Plus Environment
11
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 22
from 5.80 x 10-7 to 1,29 x 10-6 cm2.s-1, belong to aggregates in the same order of magnitude. Although these D values are within the theoretical range of macro-aggregates, an alternative (and probably more reliable) interpretation of Figure 5 spectrum is that all aggregates with D value ranging from 5.80 x 10-7 to 1.29 x 10-6 cm2.s-1, are in the nanoscale as toluene would have ripped off any micro or macro-aggregate materials. This explanation is in good agreement with “the so-called modified Yen model” which stipulates the dominant asphaltene molecular architecture, and its role in forming nanoaggregates and their respective clusters, not much bigger than the nanoaggregates, consisting of a rigid central bulk and a small fraction of mobile alkyl chains on the periphery.43 It can be pointed out that a characteristic property of the micelles formed in nonpolar solvents is their small aggregation number.44 Reverse micelles normally contain less than 20 molecules, which are very small as compared to the large aggregates found in aqueous solutions.
It should be noted that Stokes-Einstein-Debye equation was stablished for spherical particles that are much larger than the solvent molecules (RH > 5 RSolv). This is not exactly our case. In some systems, this equation fails for measurement and more sophisticated approach is needed.45 Additionally, non-uniform field gradients cause deviation of the apparent diffusion coefficient from its true value and must be calibrated accurately or its effects must be compensated.46,47,48,49 Finally, pulsed field gradient NMR diffusion data analysis has many processing procedures proposed in the literature.50,51,52 Therefore, equation 2 or some modification of this equation has not been used to calculate molecular mass and size of asphaltenes. We believe that the dominant aggregates with D value ranging from 5.80 x 10-7 to 1.29 x 10-6 cm2.s-1 in Figure 5 are in nanoscale as mentioned before.
In the literature the widely accepted molecular picture of asphaltenes molecules assume that most of aromatic rings are fused in one aromatic core with few aromatic protons. Looking closer to the aromatic region of the DOSY spectrum (Figure 5), one can notice that the corresponding aromatic signals in F2 dimension for the species with the lowest D values (5.80 x 10-7 cm2.s-1) were not observed. This structure unit description is consistent with the continental/island model that was first ACS Paragon Plus Environment
12
Page 13 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
proposed in the early 1960s.53 As D increases to 6.71 x 10-7 cm2.s-1, aromatic protons are detected due to decreasing in the number of fused aromatic rings of the species observed, and maybe more than one aromatic core are present. In this hypothesis, the long alkyl side chains present in the aggregates become entangled, thus contributing to a more compact archipelago54 structural motif. Although for higher D values (7.91 x 10-7 to 1.29 x 10-6 cm2.s-1) aromatic protons are no longer detected probably because of the lower concentration of those species and/or increasing of their alkyl chains. This interpretation assumes that both continental/island and archipelago structural motifs can coexist.
Moreover, although overlapped in the proton spectrum, it was possible to extract aliphatic signals with D of 2.66 x 10-6 to 1.52 x 10-5 cm2.s-1 in the 1H DOSY NMR spectrum. We believe those D values of the aliphatic part are not involved in aggregation formation and we speculate they are related to some paraffinic molecules and some residual solvent, respectively, co-eluted during the asphaltene isolation step using ASTM D6560-12 procedure under sonication. This observation is in agreement with the elemental analysis data, as the value of C/H = 0.917 obtained is slightly low for a highly condensed aromatic structure. This interpretation is also supported by the truly quantitative and fully relaxed
13
C NMR spectrum in Figure 4 that yields undeniable evidence of significant
amount of methylene carbons in long alkyl chains (resonance signal at ~30.4 ppm). The presence of paraffinic compounds not chemically bonded to an aromatic core is a possibility already mentioned that will certainly contribute to the intensity of this signal.
4. Conclusions 1
H and
13
C NMR results reveal that the received asphaltene sample is composed by a highly
condensed aromatic core structures with peripheral alkyl chains. This is consistent with continental/island model of asphaltenes. Elemental analysis shows that this asphaltene presents a high sulfur content. In fact, the 1029 cm-1 absorption band at the FTIR spectrum indicates the presence of sulfur-containing organic compounds. TGA shows a mass loss of 51% after air has been introduced, which is in agreement with the highly condensed aromatic system observed by 1H and ACS Paragon Plus Environment
13
Energy & Fuels
Page 14 of 22
13
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
C NMR. By its turn, DOSY suggests the presence of nano-sized aggregates and reveals some
residual solvent and some long alkyl chain compounds apparently not chemically bonded to the aromatic core. Those compounds stay occluded inside the asphaltene network even after the sonication step for decreasing co-precipitants. DOSY information is extremely important to extend sample characterization because the presence of non-asphaltenic molecules compromises analytical data and greatly affects the subsequent compositional and structural analysis. A closer analysis of DOSY spectrum suggests that continental/island and archipelago structural motifs should also coexist. DOSY data reinforces what already published in literature13,55 the asphaltene concept must be revisited, taking into account not only solubility profile, but molecular features such as C/H ratio, functional groups, degree of condensation, and molecular architecture. We intend to carry out further studies where improved processing strategies for non-uniform field gradients compensation and higher gradient power experiments will be used in order to obtain high quality DOSY data, avoiding misinterpretation of the results and allowing a more accurate determination of the size of asphaltene aggregates.
ACKNOWLEDGMENT: The authors thank the technical support of Alessandra Rangel Cassella (TGA and elemental analysis experiments), Luiz Alexandre Sacorague (FTIR analysis), and João Carlos Souza (accurate calibration of diffusion measurement), and PETROBRAS for permission to publish this work.
ACS Paragon Plus Environment
14
Page 15 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
Table 1. Elemental analysis of asphaltene sample Carbon (wt%)
82.5
Hydrogen (wt%)
7.5
Nitrogen (wt%)
1.4
Sulfur (wt%)
7.9
Oxygen (wt%)*
0.7
C/H
0.917
*Determined by mass difference.
Table 2. 1H Assignments for asphaltene sample
Chemical Shift (ppm)
Symbol
7.2 - 9.0
Hdar
6.0 -7.2
Hmar
6.0 - 9.0
Har
Total aromatic hydrogens
10.7
4.5 - 6.0
Holf
Olefinic hydrogens
-
3.4 – 4.5
Halfa-1
2.0 - 4.5
Halfa
1.5 – 2.0
Hbeta-1
1.0 – 1.5
Hbeta-2
1.0 - 2.0
Hbeta tot
Hbeta-1 + Hbeta-2
56.9
0.1 – 1.0
Hgamma
Paraffinic hydrogens in CH3s (isolated or terminal) and CH3 in position gamma or more to aromatic rings
12.0
0.1 - 4.5
Hsat
Total aliphatic hydrogens
89.3
Structure assignment Aromatic hydrogens linked to aromatic carbons in di or polyaromatic rings Aromatic hydrogens linked to monoaromatic rings
Hydrogens alfa to two aromatic rings simultaneously Paraffinic and naphthenic hydrogens (CH, CH2 and CH3) linked to aromatics in alfa position CH2 paraffinic and naphthenic beta to aromatic rings, CHs naphtenic beta to aromatic rings and some CHs in paraffinic chains Paraffinic hydrogens (CH2 and CHs) beta or further to aromatic rings, CH2s in long paraffinic chains and CH3s beta to aromatic rings
ACS Paragon Plus Environment
molar %
6.6 4.2
4.3 20.3 32.5
24.4
15
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 22
Table 3. 13C Assignments for asphaltene sample Chemical shift (ppm)
Symbol
137-160
Car-alk
Calculated from Har
Car-H
Cartot – (Car-H + Car-alk)
Car-b
100-160 0-70 14.1 10-12 and 19-20
Cartot Csat Cmet-term Cmet-bran
Structure Assignment Alkyl substituted aromatic carbons (except methyls) Aromatic protonated carbons Aromatic carbons at bridgedhead (Car-ar +Car-ar-ar + Car-Me) Total aromatic carbons Total aliphatic carbons Terminal methylic carbons Branched methylic carbons
molar % 14.1 15.6 25.2 54.9 45.1 1.3 4.5
Figura 1. TGA of asphaltene sample.
ACS Paragon Plus Environment
16
Page 17 of 22
Energy & Fuels 84 2017-003788-62 ASFALTENO P
3056,8
82
80
2918,4
68
3500
3000
2500 2000 Wavenumbers (cm-1)
1500
721,9
855,0 806,6
70
745,0
2849,0
72
1028,7
2949,1
74
1452,1 1374,5
1594,6
76
1309,1
1651,7
78
%T
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1000
Figura 2. FTIR spectra of asphaltene sample.
Figura 3. 1H NMR spectrum of asphaltene sample. Inset: aromatic region.
ACS Paragon Plus Environment
17
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figura 4.
13
Page 18 of 22
C NMR spectrum of asphaltene sample.
Figura 5. 1H DOSY NMR spectrum of asphaltene sample analyzed at 8 wt% in toluene-d8.
ACS Paragon Plus Environment
18
Page 19 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
5. REFERENCES
1
Bossingault, J. B.; apud Nellesteyn, F. J. The colloidal structure of bitumen, in: The Science of Petroleum;
Oxford University Press, London. 1938, vol. 4, 2760. 2
Zielinski, L.; Saha, I.; Freed, D. E.; Hürlimann, M. D. Langmuir. 2010, 26, 7, 5014-5021.
3
Majumdar, R. D.; Montina, T.; Mullins, O. C.; Gerken, M.; Hazendonk, P. Fuel. 2017, 193, 359-368.
4
Poveda, J. C.; Molina, D. R. J. Pet. Sci. Eng. 2012, 84–85, 1-7.
5
Eyssautier, J.; Levitz, P.; Espinat, D.; Jestin, J.; Gummel, J.; Grillo, I.; Barré, L. J. Phys. Chem. B. 2011, 115, 21, 6827–6837. 6
Chacón-Patiño, M. L.; Rowland, S. M.; Rodgers, R. P. Energy Fuels. 2017, 31, 12, 13509–13518.
7
Shayan, N. N.; Behruz Mirzayi, B. Energy Fuels. 2015, 29, 3, 1397–1406.
8
Morais, W. J. S.; Franceschi, E.; Dariva, C.; Borges, G. R.; Santos, A. F.; Santana, C. C. Energy Fuels. 2017, 31, 9, 10233–10244. 9
Schuler,B.; Meyer, G.; Peña, D.; Mullins, O. C.; Gross, L. J. Am. Chem. Soc. 2015, 137, 9870-9876.
10
Mullins, O. C., Fuel. 2007, 86, 309-312.
11
Mullins, O. C.; Sabbah, H.; Eyssautier, J.; Pomerantz, A. E.; Barré, L.; Andrews, A.B.; Ruiz-Morales,
Y.; Mostowfi, F.; McFarlane, R.; Goual, L.; Lepkowicz, R.; Cooper, T.; Orbulescu, J.; Leblanc, R. M.; Edwards, J.; Zare, R. N. Energy Fuels. 2012, 26, 7, 3986-4003. 12
Yarranton, H. W.; Barrera, D. M.; Ortiz, D. P., Energy Fuels. 2013, 27, 2474-2487.
13
Barton, A. F. M. Handbook of solubility parameters and other cohesion parameters. 1991, CRC Press,
Boca Raton, Florida, Second edition, ISBN 9780849301766. 14
IP-143 Standard Test. Determination of Asphaltenes (heptane insoluble) in Crude Petroleum and
Petroleum Products; 2004. Institute of petroleum, London, UK. 15
ASTM D6560-12. Standard Test Method for Determination of Asphaltenes (heptane insoluble) in Crude
Petroleum and Petroleum Products; 2012. 16
ASTM D893-14. Standard Test Method for Insoluble in Used Lubricating Oils; 2014. ACS Paragon Plus Environment
19
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
17
Page 20 of 22
ASTM D2007-11. Standard Test Method for Characteristic Groups in Rubber Extender and Processing
Oils and Other Petroleum-derived Oils by the Clay-gel Absorption Chromatographic Method. 2016. 18
Juyal, P.; McKenna, A. M.; Fan, T.; Cao, T.; Rueda-Velásquez, R. I.; Fitzsimmons, J. E.; Yen, A.; Rodgers,
R. P.; Wang, J.; Buckley, J. S.; Gray, M. R.; Allenson, S. J.; Creek, J. Energy Fuels. 2013, 30(4), 1899-1908. 19
Chacón-Patiño, M. L.; Vesga-Martínez, S. J.; Blanco-Tirado, C.; Orrego-Ruiz, J. A.; Gómez-Escudero, A.;
Combariza, M. Y. Energy Fuels. 2016, 30, 4550-4561. 20
Parra-Barraza, H.; Hernandez-Montiel, D.; Lizardi-Mendoza, J.; Valdez, M. Fuel. 2003, 82 (8), 869-874.
21
Oliveira, E.C.S.; Neto, A. C.; Junior, V. L.; Castro, E. V. R.; Menezes, S. M. C. Fuel. 2014, 117, Part A,
146–151. 22
Pelta, M. D.; Morris, G. A.; Stchedroff, M. J.; Hammond, S. J. Magn. Reson. Chem. 2002, 40, S147-S152.
23
Sheremata, J. M.; Gray, M. R.; Dettman, H.; McCaffrey, W. C. Energy Fuels. 2004, 18, 1377-1384.
24
Sabbah, H.; Morrow, A.; Pomerantz, A. E. ; Zare, R. N. Energy Fuels. 2011, 25, 1597-1604.
25
Nelson, J.; Yamanaka, M.; Lopez-Linares, F.; Poirier, L.; Rogel, E. Energy Fuels. 2017, 31, 11, 11971-
11976. 26
Trejo, F.; Rana, M.S.; Ancheyta, J. Catalysis Today. 2010, 150, 3-4, 272-278.
27
Ghosh, A. K.; Chaudhuri, P.; Kumar, B.; Panja, S. S. Fuel 2016, 185, 541-554.
28
Orrego-Ruiz, J. A.; Guzmán, A.; Molina, D.; Mejía-Ospino, E. Energy Fuels. 2011, 25, 3678-3686.
29
Coelho, R. R.; Hovell, I.; Monte, M. B. de M.; Middea, A.; de Souza, A. L. Fuel Processing Techonolgy.
2006, 87, 325-333. 30
Hasan, M. U.; Ali, M. F.; Bukhari, A. Fuel. 1983, 62, 518-523.
31
Jennings, P.W.; Pribanic, J.A. Binder Characterization and Evaluation by Nuclear Magnetic Resonance
Spectroscopy; Strategic Highway Research Program (SHRP-A-335). 1993. 32
Antalek, B. Concepts Magn. Reson. 2002, 14, 4, 225-258. ACS Paragon Plus Environment
20
Page 21 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
33
Kapur, G. S.; Findeisen, S. B. Fuel. 2000, 79, 1347-1351.
34
Durand, E.; Clemancey, M.; Quoineaud, A-A.; Verstraete, J.; Espinat, D.; Lancelin, J. M. Energy Fuels
2008, 22, 2604-2610. 35
Durand, E.; Clemancey, M.; Lancelin, J. M.; Verstraete, J.; Espinat, D.; Quoineaud, A. A. Energy Fuels
2010, 24, 1051-1062. 36
Durand, E.; Clemancey, M.; Lancelin, J. M.; Verstraete, J.; Espinat, D.; Quoineaud, A. A. J. Phys. Chem.
C. 2009, 113, 16266–16276. 37
Lisitza, N. V.; Freed, D. E.; Sen, P. N.; Song, Y-Q. Energy Fuels. 2009, 23, 1189-1193.
38
de Oliveira, M. C. K.; Lopes, H.; Teixeira, C. da Silva.; Chinelatto Jr, L. S.; Gonzalez, G.; Altoé, R. Energy
Fuels. 2017, 31(12), 13198-13214. 39
Louis-Joseph, A.; Benamsili, L.; Korb, J.-P. Microporous Mesoporous. 2013, 178, 23-26.
40
Vuković, J. P.; Hrenar, T.; Novak, P.; Friedrich, M.; Plavec, J. Energy Fuels. 2017, 31, 8, 8095-8101.
41
Biswas, R.; Bhattacharyya, S.; Bagchi, B. J. Phys. Chem. B. 1998, 102, 3252-3256.
42
Nikki, K. Magn. Reson. Chem. 1990, 28, 385-388.
43
Mullins, O. C. Energy Fuels 2010, 24, 4, 2179-2207.
44
Schechter, R. S.; Bourrel, M. Microemulsions and Related Systems: Formulation, Solvency, and Physical
Properties; Marcel Dekker: New York, 1988. 45
Turton, D. A.; Wynne, K. J. Phys. Chem. B 2014, 118, 17, 4600–4604.
46
Connell, M. A.; Bowyer, P.J.; Bone, P. A.; Davis, A. L.; Swanson, A. G.; Nilsson, M.; Morris, G. A. J.
Magn. Reson. 2009, 198, 121–131. 47
Damberg, P.; Jarvet, J.; Gräslund, A. J. Magn. Reson. 2001, 148, 343–348.
48
Price, W. S. Concepts Magn. Reson. 1998, 10, 4, 197-237.
ACS Paragon Plus Environment
21
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
49
Page 22 of 22
Morris, G. A. Diffusion-Ordered Spectroscopy (DOSY), in: D.M. Grant, R.K. Harris (Eds.), Encyclopedia
of Nuclear Magnetic Resonance. John Wiley & Sons Ltd., Chichester. 2002, 35–44. 50
Afef Cherni, Emilie Chouzenoux, M-A Delsuc. PALMA, an improved algorithm for DOSY signal
processing. [Research Report] LIGM - Laboratoire d'Informatique Gaspard-Monge. 2016. 51
Nilsson, M. J. Magn. Reson. 2009, 200, 296-302.
52
Huo, R.; Wehrensm R.; van Duynhoven, J.; Buydens, L. M. C. Anal. Chim. Acta. 2003, 490, 1-2, 231-251.
53
Yen, T. F.; Erdman, J. G.; Pollack, S. S. Anal. Chem. 1961, 33, 1587-1594.
54
Alvarez-Ramírez, F.; Ruiz-Moralez, Y. Energy Fuels. 2006, 27, 1971-1808.
55
Strausz, O. P.; Torres, M.; Lown, E. M.; Safarik, I.; Murgich, J. Energy Fuels 2006, 20, 2013-2021.
ACS Paragon Plus Environment
22
