Transformation of Petroleum Asphaltenes in Supercritical Alcohols


Transformation of Petroleum Asphaltenes in Supercritical Alcohols...

17 downloads 75 Views 1MB Size

Article pubs.acs.org/EF

Transformation of Petroleum Asphaltenes in Supercritical Alcohols Studied via FTIR and NMR Techniques Andrey M. Chibiryaev,*,†,‡ Ivan V. Kozhevnikov,† Anton S. Shalygin,† and Oleg N. Martyanov†,‡ †

Boreskov Institute of Catalysis, Siberian Branch of the Russian Academy of Sciences, Academician Lavrentiev Avenue 5, 630090 Novosibirsk, Russia ‡ Novosibirsk State University, Pirogov Street 1, 630090 Novosibirsk, Russia S Supporting Information *

ABSTRACT: The aliphatic alcohols (methanol, ethanol, and 1- and 2-propanols) were used for the first time as a reaction media for the upgrading of crude oil asphaltenes. The process was realized in a batch reactor under supercritical conditions (at 350 °C). The three main fractions of the products (hexane- and benzene-soluble fractions, HSF and BSF, and insoluble residue, IR) were analyzed using attenuated total reflection Fourier tranform infrared (ATR-FTIR) and 1H and 13C nuclear magnetic resonance (NMR) spectroscopy to characterize structural changes of the initial asphaltenes (IA). According to NMR data, the aliphatics are the main part of the hexane-soluble fraction (HSF) and benzene-soluble fraction (BSF). The alcohols were appeared to influence the content of both aliphatics and aromatics in the products. The content of aliphatics in the HSF increases in the line from “lighter” to “heavier” alcohols used but reduces in the BSF. However, the content of aromatics in the HSF increases from “heavier” to “lighter” alcohols, while this order is reversed for the BSF. According to the ATR-FTIR spectroscopy data, the aromatics-to-aliphatics ratios observed for the insoluble residues are 2−3 times higher as compared with the initial asphaltenes but 2 times lower for the HSF. The BSF are composed of less-condensed aromatics than those of the IA. It is shown that the alcohols used as a reaction media are incorporated in the product molecules as alkoxy substituents in aromatic ethers Ar−OAlk. According to NMR and ATR-FTIR data obtained, the alkylation−dealkylation and alkoxylation reactions make a crucial contribution to the chemical transformations of the asphaltenes.

1. INTRODUCTION Nowadays, oil refining industry tends to convert the low-value heavy feedstocks (atmospheric and vacuum residues, bitumen, asphaltenes, etc.) into valuable products (fuels, petroleumderived lubricants, motor fuel or gasoline additives, etc.). The upgrading of asphaltenes naturally enriched with sulfur-, nitrogen-, and metal-containing compounds is one of the main challenges to develop the efficient heavy oil processing technologies. In general, being present in large quantity in various heavy crudes,1 asphaltenes cause a lot of problems during the production, refining, and upgrading of oils. The significant increase of the viscosity reducing the mass transfer in pipelines, the undesired aggregation and precipitation of heavy fractions leading to the fouling and the formation of deposits, the tendency to form coke, which deactivates and poisons the catalysts is an incomplete list of the problems generated by the asphaltenes.2−4 Different analytical techniques were used to characterize isolated asphaltenes and their aggregates.5−7 The vaporpressure osmometry (VPO),8,9 size-exclusion chromatography (SEC),10 13C and 1H nuclear magnetic resonance (NMR) and magnetic resonance tomography (MRI),11 small-angle neutron (SANS)12 and X-ray (SAXS)13 scattering, time-resolved fluorescence depolarization (TRFD),14 Fourier transform infrared spectroscopy (FTIR) in attenuated total reflection mode,15,16 Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS),6 ESR,17,18 etc., provide the data on the characteristic size and mean molecular structure of asphaltenes, in particular the number of fused rings, the length © 2017 American Chemical Society

of aliphatic chains, the common functional groups, and the molecular weight. Despite the wide possibilities of these analytical tools, each of them has some limitations and should be used in combination with others to obtain the reliable results and complete picture.19 To develop an efficient technology for the processing of heavy oils, it is necessary not just to characterize the asphaltenes or their aggregates but also to analyze their behaviors and chemical transformation on a molecular scale under real operating conditions of particular refining process. It is a crucial step to find a way for qualified processing of the asphaltene containing crudes. Some efforts were spent to understand the mechanism of asphaltenes transformation during hydroprocessing of heavy oils.10,20−25 The characterization and analysis of hydroprocessed asphaltenes and the impact of the reaction conditions on their structure were reported earlier.26,27 During the last two decades, a lot of efforts have been made to apply supercritical fluids (SCF) as reaction media for the upgrading of the asphaltenes, bitumen, and other heavy fractions.28−41 The heavy oil upgrading technology with “alternative” hydrogen is in high demand. The supercritical medium containing alkyl aromatics or saturated hydrocarbons can provide the “hydrogen-rich” conditions. The noticeable benefits of supercritical fluids that can be used even without a Received: June 7, 2017 Revised: October 19, 2017 Published: October 24, 2017 2117

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article

Energy & Fuels

The reaction occurred for 3 h with stirring rate of 800 rpm in all runs at a constant temperature of 350 ± 1 °C and a density of reaction mixture of approximately 0.33 g/cm3. The reaction pressure depended on the alcohol used and rose slowly during the reaction: for methanol, from 210 up to 242 atm; for ethanol, from 179 up to 189 atm; for 1propanol, from 112 up to 138 atm; and for isopropanol, from 118 up to 160 atm, respectively. After cooling of the reactor to room temperature (30 min), the residual pressure after the reaction was 3−6 atm. 2.3. Separation and Partition the Products into the Main Fractions. Because all reactions were carried out in alcohols, the final reaction mixture was divided into two portions: alcoholic solution of the products and insoluble solids (“Solids 1” in Figure 1). The solvent

heterogeneous catalyst have been supported by many studies28−32,36 and summarized in a recent review.42 Recently the capabilities of supercritical water (SCW) to upgrade the asphaltenes were investigated at 380 °C using a batch reactor.43 It was shown that more than half of the asphaltenes (IA) were successfully transferred into gaseous and low-molecular-weight products soluble in hexane or benzene. The highest content of saturated aliphatic hydrocarbons was found in the hexanesoluble fraction (HSF), whereas the highest content of aromatics moved into the fraction insoluble in neither hot hexane nor benzene. Lower-alcohol media is another kind of promising “hydrogen-rich” SCF that can be applied for the upgrading of heavy oil fractions. Supercritical (sc) ethanol and methanol were used earlier for deasphalting of vacuum residue and crude oil by their extraction with supercritical alcohols as a solvent or polar component of a complex solvent.44−47 Unfortunately, so far, the studies of the asphaltenes transformation in sc lower alcohols were not performed. Here, we report for the first time the experimental data obtained by FTIR and NMR techniques concerning the structural changes of asphaltenes that occur in sc lower alcohols. FTIR spectroscopy is a versatile and powerful technique that is widely used to monitor the transformation of petroleum asphaltenes structure under different conditions.48,49 The 1H and 13C NMR method is widely used to examine the distribution of carbon and hydrogen atoms in different structural groups found in aliphatic and aromatic parts of asphaltenes50 to estimate average molecular weights and structure of the “average asphaltene molecule”.51,52 Both NMR and FTIR data can successfully complement each other.53 The objective of this work was to analyze the chemical transformation of asphaltenes provided by supercritical alcohols using FTIR and NMR methods. The comprehensive effects of different supercritical alcohols upon asphaltene conversion and chemical composition of the products (distribution of some functional groups) are discussed. The data obtained in this work are discussed as compared with those achieved using supercritical water at similar temperature.

Figure 1. Scheme of fractions isolation. was distilled off from the alcoholic solution at atmospheric pressure to give a solid residue (minor fraction, “Solids 2” in Figure 1), which was combined with the main portion of solid products. The last was a major fraction. All the solid products were redistributed by consequent extraction with hot n-hexane and benzene to give a hexane-soluble fraction (HSF), a benzene-soluble fraction (BSF), and a residue insoluble neither in n-hexane nor in benzene (IR). To do so, the combined solids were replaced in a Soxhlet apparatus charged with ∼250 mL of n-hexane that was refluxed until the solvent dripping back down into the distillation flask became colorless (4−8 h). After extraction, the solvent was removed from the hexane solution of the extractives by means of a rotary evaporator with a heating bath at 85 °C under atmospheric pressure and a slow continuous flow of gaseous CO2 to concentrate the extract to 15−20 mL. After cooling to room temperature, the extract was filtered out, and the residual solvent was removed from the prepared concentrate to give the HSF. The extraction of the residual solids in a Soxhlet extractor was repeated with benzene in the same manner to give the BSF and an IR. The insoluble residue was dried a vacuo (under reduced pressure of 12−15 Torr) at 100 °C under argon. The scheme of the total procedure of the products separation and partition is shown in Figure 1. 2.4. NMR and FTIR Techniques. The NMR spectra were recorded using Bruker DRX 500 instrument (500.13 MHz for 1H and 125.75 MHz for 13C) under conditions recommended by ASTM standard no. D5292-9955 for chloroform-D solutions using the signal of the solvent (CDCl3) as an internal standard: δH, 7.24 and δC, 76.90 ppm. The total molar content of aromatic atoms (hydrogen or carbon) was calculated using the formula:

2. EXPERIMENTAL SECTION 2.1. Reagents and Solvents. All solvents were used without additional purification: methanol (≥99.8% with a water content ≤0.05%, J.T. Baker), ethanol (≥99.8%, Sigma-Aldrich), 1-propanol (≥99.5%, Sigma-Aldrich), 2-propanol (≥99.5%, Sigma-Aldrich), nhexane (95%, Sigma-Aldrich), and benzene (≥99%, Sigma-Aldrich). Critical parameters of used alcohols are as follows. For methanol: Tcr = 240 °C, Pcr = 80 atm, ρcr = 0.27 g/cm3; ethanol: Tcr = 244 °C, Pcr = 63 atm, ρcr = 0.28 g/cm3; 1-propanol: Tcr = 264 °C, Pcr = 51 atm, ρcr = 0.28 g/cm3; isopropanol (2-propanol): Tcr = 235 °C, Pcr = 47 atm, ρcr = 0.27 g/cm3. The Tatar heavy crude oil (from the oil field of the Republic of Tatarstan, Russia) of 4.5% sulfur content was used as a source of asphaltenes. The asphaltenes were precipitated and purified following the modified ASTM method D6560-1254 by replacing heptane and toluene as solvents for n-hexane and benzene, respectively. The amount of asphaltenes isolated by the method was up to 7.0 wt % of the crude oil. 2.2. General Experimental Procedure. The batch reactor (see the Experimental Setup section in the Supporting Information) was charged with ∼120 ± 1 mL of alcohol (methanol, ethanol, 1-propanol, or isopropanol) and 3.00 g of ground asphaltenes. Before heating, the bolted closure reactor was purged with argon. The heating time was 47−50 min to reach the reaction temperature of 350 °C starting from 25 °C with heating rate of ∼7 °C/min. 2118

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article

Energy & Fuels

SCW at 380 °C leads to considerably higher conversion and lower amount of the insoluble residue (49 wt %, Table 1).43 The content of carbon in the product fractions decreases in the line IR > HSF ≥ IA > BSF with slight deviation for methanol (Table 2). The content of hydrogen follows different order (HSF > BSF > IA > IR), which actually repeats the content of oxygen in the fractions (see the Supporting Information). The total content of hydrogen in the products obtained after the processing of asphaltenes in sc propanol was 102.5% as compared to the content of hydrogen in initial asphaltenes (IA), while the total amount of the solid products recovered (HSF + BSF + IR) was 97.7 wt % of the IA (Table 2; see also the Supporting Information). The data obtained points to the intensive alkoxylation of the IA in lower alcohols, especially in sc methanol. 3.2. 1H and 13C NMR Spectral Data. The asphaltenes as well as all fractions of the products were analyzed using NMR technique. The data obtained have accuracy better than 0.05%.57,58 1H and 13C NMR methods are a powerful tool with which to determine the content of aromatic hydrogen and carbon in multicomponent hydrocarbon systems including fuels, lubricant,59 and asphaltenes.55 Unfortunately, the registration of NMR spectra of insoluble residue formed as a product of the reactions in lower alcohols was rather difficult due to insufficient solubility of IR in regular solvents usually used for NMR spectroscopy. So, the NMR measurements were made only for the HSF, BSF, and IA. 3.2.1. Analysis of the C and H Contents and Carom-to-Caliph and Harom-to-Haliph Ratios. The relative values of the C and H contents (Harom, Haliph, Carom, and Caliph) as well as the Carom/ Caliph and Harom/Haliph ratios can be obtained directly from the analysis of NMR spectra. According to NMR data, the content of Harom and Haliph in soluble fractions for different alcohols used varies from 8.2 to 17.6% and from 82.4 to 91.8%, respectively, while the Harom/Haliph ratio in these products varies from 0.09 to 0.21. In particular, the Harom/Haliph ratio varies within wider range for BSF (0.12−0.21) than for the HSF (0.09−0.10) obtained for different alcohols. A smaller variation is observed for the content of Carom and Caliph in soluble fractions for different alcohols used that ranges from 30.7 up to 56.1% and from 43.9 up to 69.4%, respectively. One can see the similar behavior of Carom/Caliph and Harom/Haliph ratios for the same type of fractions obtained in different alcohols. These values vary within wider range for the BSF as compared to the HSF:

Harom = S(Haromatic)/[S(Haromatic) + S(Haliphatic)] × 100% or

Carom = S(Caromatic)/[S(Caromatic) + S(Caliphatic)] × 100% where S (Haromatic) and S (Haliphatic) are the total integrated signals of aromatic and aliphatic hydrogen atoms, respectively. Similar characters are used for the Carom formula. The FTIR spectroscopic analysis was done using a Bruker Vertex 70v spectrometer equipped with a diamond ATR accessory (Specac Ltd., UK) and a mercury-cadmium-telluride detector. A total of 100 scans were taken for each sample recorded from 4000 to 370 cm−1 at a resolution of 4 cm−1, and the spectra were transformed by the ATR correction function of the OPUS software using a refractive index of sample (n = 1.5). The used peak intensity measurement method was described earlier.56 The intensities of intrinsic bands were calculated with baseline correction. The FTIR spectra deconvolution was performed by a peak fitting to the symmetrical Gaussian peak shapes using the Origin package software.

3. RESULTS AND DISCUSSION Recently, the chemical transformation of the same petroleum asphaltenes was studied in water under supercritical conditions (SCW) at 380 °C and 226 atm.43 The reaction temperature in sc alcohols was 350 °C to avoid significant thermal decomposition of the alcohols used. In this study, the asphaltenes-to-alcohol ratio (w/w ≈ 0.032) was close to the asphaltenes-to-water ratio (w/w ≈ 0.038) used earlier for the SCW process. 3.1. Initial Chemical Information about the Fractions. The reaction products were collected, separated and analyzed the same way for all alcohols used. The solid product (fraction “All solids” in Figure 1) remaining after solvent removal was separated into three main components: a hexane-soluble fraction, a benzene-soluble fraction, and an insoluble residue, which can be dissolved neither in hot n-hexane nor in benzene. These fractions were studied in details for all alcohols used via different methods, including elemental C,H,N,S,O-analysis, NMR and FTIR spectroscopies. The volatile products formed in the reaction as well as some sulfur-containing compounds were analyzed using the gas chromatography−mass spectrometry (GC−MS) technique. All experimental details of the separation procedure and characterization of the products are given in the Supporting Information. The weight of all fractions is shown in Table 1. The insoluble residue was always the main fraction of the products for all

Harom:

Table 1. Main Fractions of Products Obtained processing of asphaltenes by supercritical fluids (wt %) fraction hexane-soluble fraction (HSF) benzene-soluble fraction (BSF) insoluble residue (IR) total Σ of IAb a

MeOH EtOH 9.7 2.8 80.7 93.2

18.3 5.9 68.1 92.3

PrOH

iPrOH

H2Oa

21.0 4.4 72.3 97.7

15.1 5.9 75.8 96.8

30.0 16.3 48.6 94.9

HSFPrOH < HSFi − PrOH < HSFEtOH < HSFMeOH < IA

IA < BSFMeOH < BSFEtOH < BSFi − PrOH < BSFPrOH

Haliph: IA < HSFMeOH < HSFEtOH < HSFi − PrOH < HSFPrOH BSFPrOH < BSFi − PrOH < BSFEtOH < BSFMeOH < IA

Data from ref 43. bAmount of initial asphaltenes (IA) is 100 wt %.

Harom/Haliph:

alcohols used. The highest amount of the IR was observed for methanol: IREtOH (68 wt %) < IRPrOH (72 wt %) < IRi‑PrOH (76 wt %) < IRMeOH (81 wt %) (Table 1). The lowest amount of the IR as well as the highest conversion of the IA was observed for sc ethanol. For comparison, the processing of asphaltenes in

HSFPrOH ≈ HSFi − PrOH ≤ HSFEtOH ≈ HSFMeOH < IA

BSFMeOH ≈ IA < BSFEtOH < BSFi − PrOH < BSFPrOH Carom: 2119

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article

Energy & Fuels Table 2. C and H Analysis of Initial Asphaltenes and Main Fractions Obtained fraction initial asphaltenes sc MeOH

sc EtOH

sc PrOH

sc i-PrOH

sc H2Oc

HSF BSF IR total Σb HSF BSF IR total Σb HSF BSF IR total Σb HSF BSF IR total Σb HSF BSF IR total Σb

C, wt %

H, wt %

C/Ha

empirical formula

81.07 80.16 80.91 82.30 94.3 81.05 80.81 81.34 92.5 81.42 80.53 83.06 99.5 81.01 80.33 82.16 97.0 82.20 81.34 68.90 87.2

7.07 10.34 8.61 6.30 89.5 10.02 8.59 6.29 93.7 10.68 8.60 6.40 102.5 10.57 8.59 6.44 97.8 11.70 6.32 3.50 85.5

0.964 0.652 0.791 1.099 1.016 0.680 0.791 1.087 0.952 0.641 0.787 1.092 0.936 0.644 0.786 1.073 0.956 0.591 1.082 1.656 0.962

C100H103.7N2.4S3.6O1.9 C100H153.4N1.5S2.2O4.1 C100H126.5N1.9S2.7O3.4 C100H91.0N1.8S3.1O1.6 C100H98.4N1.7S3.0O1.9 C100H147.0N1.7S2.3O2.9 C100H126.4N1.4S2.9O2.8 C100H92.0N1.9S3.2O1.7 C100H105.0N1.8S3.0O2.1 C100H156.0N1.5S2.1O2.6 C100H127.0N1.2S2.9O3.0 C100H91.6N1.8S3.1O1.5 C100H106.8N1.7S2.9O1.8 C100H155.2N1.2S2.2O2.9 C100H127.2N1.1S2.8O2.9 C100H93.2N1.8S3.2O1.5 C100H104.6N1.6S3.0O1.8 C100H169.3N0.3S2.4 C100H92.4N2.4S3.3 C100H60.4N3.2S4.3 C100H104.0N2.1S3.4

a

C/H is a molar ratio. bThe percentage is calculated relatively the content of the elements in initial asphaltenes (IA), which is taken as 100%. cData from ref 43.

quaternary carbon units. One can find that the value of Caliph/ Haliph > 0.5 is observed for the BSFMeOH only (0.55). It is known that the Carom-to-Harom ratio characterizes the degree of substitution and/or condensation of aromatic rings: the larger the value, the higher the degree of substitution. The Carom-to-Harom ratio in the products obtained after processing of asphaltenes in lower alcohols varies from 2.29 to 2.75 for all soluble fractions that is two times less than the corresponding value for initial asphaltenes. It implies that molecular structure of the soluble products includes the less condensed and less substituted aromatic rings as compared to the IA. One can find that the values of the Carom-to-Harom and Caliph-to-Haliph ratios increases in the following way (Table 3):

HSFPrOH < HSFi − PrOH < HSFEtOH < HSFMeOH < IA

BSFMeOH < BSFEtOH < BSFi − PrOH < BSFPrOH < IA

Caliph: IA < HSFMeOH < HSFEtOH < HSFi − PrOH < HSFPrOH IA < BSFPrOH < BSFi − PrOH < BSFEtOH < BSFMeOH

Carom/Caliph: HSFPrOH ≈ HSFi − PrOH < HSFEtOH < HSFMeOH < IA

Carom/Harom:

BSFMeOH < BSFEtOH < BSFi − PrOH ≈ BSFPrOH < IA

HSFi − PrOH < HSFPrOH < HSFEtOH ≈ HSFMeOH < IA

Surprisingly, the content of aromatics (Harom and Carom) in HSF increases from “heavier” to “lighter” alcohols, while the order is reversed for the BSF. The opposite situation is observed for the content of aliphatics (Haliph and Caliph), which increases from “lighter” to “heavier” alcohols in the HSF and decreases in the BSF. Thus, the symbatic changes are observed for Harom-to-Haliph and Carom-to-Caliph ratios and the content of aromatics in HSF and BSF obtained for different alcohols. 3.2.2. Carom-to-Harom and Caliph-to-Haliph Ratios. The useful information about the transformation of asphaltenes in supercritical alcohols can be extracted via analysis of the Carom-to-Harom and Caliph-to-Haliph ratios based on NMR and elemental analysis data (Tables 2 and 3). The Caliph-to-Haliph ratio for HSF and BSF obtained in all alcohols changes within the range 0.46−0.50 that is typical for saturated aliphatics. It means that the aliphatic compounds found in the products preferentially composed of straight aliphatic carbon chains, (CH2)n, (0.50). The smaller Caliph-to-Haliph ratio (0.5) points to branched carbon skeleton of the aliphatics with tertiary C−H (1.00) or

BSFPrOH < BSFi − PrOH < BSFEtOH < BSFMeOH < IA

Caliph /Haliph: HSFMeOH < IA < HSFPrOH ≈ HSFi − PrOH < HSFEtOH

BSFi − PrOH < BSFPrOH < IA < BSFEtOH < BSFMeOH

The values of the Caliph-to-Haliph ratio behave in a similar way to the content of aliphatics (Caliph and Haliph) and increase from “lighter” to “heavier” alcohol in the HSF but decrease in the BSF. The trend of the changes of Carom-to-Harom ratio differs from the Caliph-to-Haliph ratio and the Caliph and Haliph values discussed above. In contrast to all previous cases, the Carom-toHarom ratio increases from “heavier” to “lighter” alcohols for all soluble products (HSF and BSF). This is the only case than a parameter behaves in a similar way for both hexane- and benzene-soluble fractions. To visualize the NMR data obtained, the content of the aromatic and aliphatic C and H atoms in the soluble fractions 2120

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article

Energy & Fuels Table 3. Hydrogen and Carbon Contents According to 1H and 13C NMR Spectral Data fraction

Carom, %

Caliph, %

Carom/ Caliph

Harom, %

Haliph, %

Harom/ Haliph

Caliph/ Halipha

Carom/ Haroma

C/Hb

ratio of aromatic and aliphatic C and H atoms

initial asphaltenes sc MeOH HSF BSF sc EtOH HSF BSF sc PrOH HSF BSF sc i-PrOH HSF BSF

56.12 35.53 38.36 33.73 47.51 30.65 51.10 30.94 50.79

43.88 64.47 61.64 66.27 52.49 69.35 48.90 69.06 49.21

1.28 0.55 0.62 0.51 0.91 0.44 1.04 0.45 1.03

10.95 9.09 11.03 9.01 14.56 8.17 17.56 8.51 16.04

89.05 90.91 88.97 90.99 85.44 91.83 82.44 91.49 83.96

0.12 0.10 0.12 0.10 0.17 0.09 0.21 0.09 0.19

0.475 0.462 0.548 0.495 0.486 0.484 0.467 0.486 0.461

4.942 2.548 2.749 2.547 2.582 2.405 2.291 2.343 2.489

0.964 0.652 0.791 0.680 0.791 0.641 0.787 0.644 0.786

arom aliph aliph Carom 56.1 H11.4 C43.9 H92.3 arom arom aliph aliph C35.5 H13.9 C64.5 H139.6 arom aliph aliph Carom 38.4 H14.0 C61.6 H112.5 arom aliph aliph Carom H 33.7 13.2 C66.3 H133.8 arom arom aliph aliph C47.5 H18.4 C52.5 H108.3 arom aliph aliph Carom 30.7 H12.7 C69.3 H143.3 arom aliph aliph Carom H 51.1 22.3 C48.9 H104.7 arom aliph aliph Carom H 30.9 13.2 C69.1 H142.0 arom arom aliph aliph C50.8 H20.4 C49.2 H106.8

Taking into account the empirical formula CAHBNCSDOE of the fraction (see Table 2), Caliph/Haliph = (Caliph × A)/(Haliph × B). The similar formula is used for the ratio of Carom/Harom. bC/H is a molar ratio according to elemental data of CHNSO analysis. a

are observed for IA. The different solubility of the products in hot hexane or benzene and the C and H content distribution (Table 3) indicates that the BSF products contain heavier aromatics with shorter aliphatic side chains (as compared to the HSF). The Caliph-to-Haliph ratios and the content of aliphatics in the HSF increase in the order of MeOH−EtOH−i-PrOH− PrOH, while the Carom-to-Harom ratios and the content of aromatics (Carom and Harom) decrease. Thus, the data observed demonstrate the influence of alcohols used on the distribution of aliphatic and aromatics in certain soluble products. 3.3. FTIR Study of the Products. All solid products (HSF, BSF, and IR) obtained after the processing of asphaltenes in supercritical alcohols were studied in details using FTIR spectroscopy in comparison with initial asphaltenes and the products obtained earlier after the processing of the same asphaltenes in supercritical water.43 The spectra obtained can be formally divided into three regions: (a) 600−900, (b) 900− 1800, and (c) 2600−3100 cm−1. The region a identifies the main aromatic C−H stretching or out-of-plane bending modes as well as rocking modes of −(CH2)n− alkyl chains. The bands at 860−900 and 810−860 cm−1 can be attributed to the out-ofplane bending modes of the Carom−H bond of aromatics with one and two adjacent H, respectively, whereas bands at 750− 800 and 735−750 cm−1 are assigned to the stretching modes of the Carom−H bond of aromatics with three and four adjacent H, correspondingly.60 The modes having higher wavenumbers are related to aromatics with larger numbers of substituents. The stretching modes of the C−H bonds of aromatics with five adjacent H can be observed at 690−710 and 730−770 cm−1,60

can be plotted (Figure 2). The horizontal blue bars stand for the content of carbon atoms (both aromatic and aliphatic

Figure 2. C and H atom distribution between aliphatic and aromatic structural parts of the products of HSF and BSF according to NMR data.

ones), and the horizontal green bars are for the content of hydrogen atoms (both aromatic and aliphatic ones). The vertical red line in the middle is a borderline between the aromatic and aliphatic components. The diagram demonstrates clearly the difference in composition of the main soluble fractions (HSF and BSF) and the initial asphaltenes. It indicates that the soluble fractions consist of low-molecular-weight organic compounds preferentially composed of aliphatics with simple aromatic fragments (low-condensation and lowsubstitution aromatics) in contrast to initial asphaltenes. Indeed, the highest Carom-to-Caliph and Carom-to-Harom ratios

Figure 3. FTIR spectra of IA, IRMeOH, IREtOH, IRPrOH, IRi‑PrOH, and IRSCW in the (a) 600−900, (b) 900−1800, and (c) 2600−3100 cm−1 regions. 2121

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article a k = 1.243 is the linear correlation coefficient of the experimental plot n (CH2)/n (CH3) vs I (2927 cm−l)/I (2957 cm−l).50 bThese stretching modes are overlapped by the solvent bands of CHCl3, in which the spectrum was recorded.43 cThere are no detectable bands of carbonyl or carboxyl CO groups.

IR BSF

2.09 2.59 0.17 2.36 −c −c 1.52 1.89 0.03 −b 0.08 0.71

HSF IR

2.06 2.56 0.17 2.31 0.06 0.27 1.82 2.26 0.12 2.44 0.08 0.42

BSF HSF

1.62 2.02 0.03 1.71 0.03 0.52 2.05 2.55 0.15 3.16 0.06 0.29

IR BSF

1.77 2.19 0.10 2.77 0.08 0.43 1.87 2.38 0.03 1.45 0.05 0.65

HSF IR

2.37 2.94 0.18 2.58 0.06 0.25 1.90 2.36 0.13 2.90 0.11 0.46

BSF HSF

1.79 2.23 0.04 1.52 0.05 0.56 2.46 3.05 0.21 2.66 0.07 0.27

IR BSF IA

2.28 2.83 0.07 2.47 0.03 0.31

characteristic ratios

CH2 (ν2922)/CH3 (ν2950) n(CH2)/m(CH3) = I(ν2927)/I(ν2957) × ka CC (ν1600)/(CH2 (ν2920) + CH3 (ν2950)) S [(1H + 2H)/(3H + 4H)] = I (ν858+ν807)/I (ν750) CO (ν1650−1770)/(CH2 (ν2920) + CH3 (ν2950)) CO (ν1650−1770)/(CO (ν1650−1770) + CC (ν1600))

HSF

1.82 2.26 0.05 2.99 0.09 0.63

SCW i-PrOH PrOH EtOH MeOH

Table 4. Selected Band-Intensity Ratios of Some Functional Groups in the Spectra of The IA and All Other Fractions 2122

1.78 2.21 0.03 1.75 0.05 0.65

which are usually difficult to identify or deconvolute due to the overlapping of the bands. The bands at 675 and 725 cm−1 can be assigned to rocking modes of −(CH2)n− alkyl chains with n ≥ 4.61−63 Peaks in the region b can be related to the different functional groups: CO, C−O, CC, CN, and C−H of CH2 or CH3, and some others. Finally, the region c contains the peaks related to the corresponding stretching C−H modes of aliphatic methyl and methylene groups or aromatic bonds. 3.3.1. Study of the Insoluble Residues. The FTIR spectra of the IA and all the IR are shown in Figure 3a−c. FTIR spectrum of the IA has some characteristic bands at 858, 807, 745, 725, and 675 cm−1, which should be assigned to the Carom−H bonds of aromatics (Figure 3a). In the region b (Figure 3b), the IA has bands at 1030, 1312, 1374, 1437, 1453 (the highest intensity), 1600, 1646, 1664, and 1725 cm−1. The weak bands centered at 1664 and 1725 cm−1 are attributed to the stretching mode of CO functional group (amides, ketones, or esters). The band at 1646 cm−1 is attributed to the stretching mode of the CC aliphatic double bond in alkenes, whereas the strong broad band centered at 1600 cm−1 is assigned to the stretching mode of the CC aromatic bond. The high-intensity bands at 1437−1453 and 1374 cm−1 should be assigned to the deformation vibration of CH2 and CH3 groups, while the broad band at ∼1312 cm−1 can be attributed to CC stretching modes of aromatics.64 Additionally, some weak bands located at 1000−1300 cm−1 can be assigned to C−O stretching modes of aromatic ethers. The FTIR spectra of the insoluble residue in the region b are similar to the spectra registered for the initial asphaltenes but have some small differences. The bands of the spectrum of IA at 1600 and 1453 cm−1 are shifted in a corresponding manner to 1580 and 1437−1439 cm−1 for the spectra of IR. Also an additional weak band is detected in the IR spectra at 1212 cm−1 (CAlk−O bond of ethers). This is direct evidence that the alkoxylation reaction of the asphaltenes (or their destruction with simultaneous alkoxylation) takes place in supercritical alcohols. The FTIR spectra of IR obtained after the processing of asphaltenes in supercritical alcohols differ noticeably from the spectra of the insoluble residue obtained after SCW processing (IRSCW). A band at 1600 cm−1 (the stretching mode of the C C aromatic bonds) has higher intensity as compared with the bands at 1375 and 1437 cm−1 (CH3 and CH2 groups, respectively); that means that the IRSCW are composed of more polyaromatic components, which have no ether groups (CAlk−O). The bands at 2827−2957 cm−1 are typical for the asphaltenes and the products usually obtained after their chemical processing.65 The high-intensity bands at 2850−2920 and 2866−2952 cm−1 are attributed to asymmetric and symmetric stretching frequencies of CH2 and CH3 groups, respectively (Figure 3c). The intensity of the similar peaks in the spectra of the IR is weaker, and the ratio (I2920/I2952) of the intensity of these bands is different (Table 4). 3.3.2. Comparative Study of HSF and BSF. The FTIR spectra of soluble fractions (HSF and BSF) at 650−1850 cm−1 are shown in Figures 4a and 5a. The spectra are quite similar to each other. The peculiarities are observed for the spectrum of HSFSCW, especially at 1122, 1300, 1648, and 1718 cm−1 (CO stretching bond). The bands related to the stretching modes of CH2 and CH3 groups are more intensive for hexane-soluble fraction (HSF) as compared to the BSF, whereas the bands

2.10 2.61 1.48 2.43 0.87 0.37

Energy & Fuels

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article

Energy & Fuels

Figure 4. FTIR spectra of HSF in the (a) 650−1850 cm−1 and (b) 2820−3030 cm−1 regions.

Figure 5. FTIR spectra of BSF in the (a) 650−1850 cm−1 and (b) 2820−3030 cm−1 regions.

assigned to aromatic CC and C−H bonds are more intensive for the BSF. The band at ∼1260 cm−1 is typical for all the benzenesoluble fractions obtained after processing of asphaltenes in supercritical alcohol and attributed of the Carom−O bond (aromatic ethers Ar−OAlk). The other bands at 1700−1740 cm−1, which can be related to the stretching modes of CO group, are typical for all soluble products (HSF and BSF). The CO group observed should be assigned to the esters but not to ketones and aldehydes because the supercritical alcohols reduce these compounds easily at high temperature without any catalyst.66−71 The FTIR spectra of all fractions (HSF, BSF, and IR) are similar to each other at 2810−3010 cm−1 and display the same stretching modes at ∼2853, 2869, 2924, and 2954 cm−1 (Figures 3c−5c) that confirms a similarity of aliphatics of all fractions obtained after processing of asphaltenes in lower alcohols and SCW. 3.3.3. Analysis of Characteristic Intensity Ratios of the Fractions. The relative intensity of particular bands was used to evaluate the molar content of some functional groups. The molar ratio of nCH2-to-mCH3 usually correlates with the intensity ratio of the bands at 2927 and 2957 cm−1 (or at 2922 and 2950 cm−1).56,72 To evaluate relative content of aromatics in initial asphaltene and products, the aromatics-to-aliphatics ratio [ICC (ν1600)/I(CH2 (ν2920) + CH3 (ν2950)] was used taking into account the degree of aromatics substitution and condensation. The latter parameter was calculated as a ratio of aromatics with one adjacent proton to those having four

adjacent protons, which is known as so-called S (1H/4H) or P (1H/4H) index.50,73 Other useful indices are the carbonyls-toaliphatics ratio [ICO (ν1650−1770)/I(CH2(ν2920)+CH3(ν2950)] and an index of carbonyl abundance [I CO (ν 1650−1770 )/ ICO (ν1650−1770)+CC (ν1600)]. These parameters were calculated for initial asphaltenes and all fractions obtained after asphaltenes processing in supercritical alcohols and SCW (Table 4). As expected, the parameters CH2 (ν2922)/CH3 (ν2950) and n(CH2)/m(CH3) = I(ν2927)/I(ν2957)×k (lines 1 and 2 in Table 4) behave the similar way. The n(CH2)/m(CH3) molar ratio increases in the order of HSF−BSF−IR (line 2 in Table 4) for all alcohols used with some deviation for PrOH and vary within the range 1.89−3.05 that means that two or three CH2 groups are presented in the asphaltenes or the products per one CH3 fragment. These values corroborates the data obtained for other types of asphaltenes.50,72,73 One can see that the type of alcohol used is not crucial for the changes of the CH2-to-CH3 molar ratio observed in the products. The decrease of the n(CH2)-tom(CH3) molar ratio in soluble products (HSF and BSF) relative to the initial asphaltenes is the result of the alkylation and alkoxylation reaction of asphaltenes and alcohols. The aromatics-to-aliphatics ratio increases in the order of HSF−BSF−IR (line 3 in Table 4) and varies within the ranges of 0.03−0.04, 0.05−0.13 and 0.15−0.21 for HSF, BSF, and IR products, respectively. One can see that this parameter is two times smaller for HSF products as compared to that for the initial asphaltenes (0.07) and 2−3 times higher for the IR 2123

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article

Energy & Fuels

aliphatics and CO functional groups in the products should behave in a similar way. On the contrary, the amount of aromatics and CO groups have opposite trends. Thus, the index of carbonyl abundance (line 6, Figure 4) and the aromatics-to-aliphatics ratios (line 3, Figure 4) behave an opposite way. However, the correlation between the amounts of the aliphatics and CO functional groups results in small variation of the carbonyl-to-aliphatics ratio for the products obtained. Finally, it should be noted that each of BSFs is not the same as the initial asphaltenes. The quasisimilarity between the benzene-soluble fraction (BSF) and the IA is only in their good solubility in benzene, nothing else. According to our data, BSF and IA are quite different compounds, which have different both C-to-H ratios and C and H contents (Tables 2−4). Also their N, O, and S contents are different enough. Moreover, BSF and IA compounds consist of different functional groups (Figures 3 and 5; the wavenumber range from 700 to 1800 cm−1). Thus, from the chemical standpoint, the BSF cannot be considered as unreacted (unconverted) asphaltenes but as products of asphaltenes upgrading (transformation).

products. It means that the amount of aromatic compounds in benzene-soluble fraction is much higher than in the hexanesoluble fraction that is in agreement with the NMR data obtained. The dealkylation reaction leads to the decrease of S (1H/ 4H) index, whereas the alkylation (alkoxylation) reaction results in its increase. The intensity of the bands at 860 and 750 cm−1 attributed to out-of-plane C−H deformation modes of aromatics with one adjacent proton and with four adjacent protons, respectively, can be used to estimate this index. Unfortunately, these bands and some other stretching modes of the FTIR spectra of products are overlapped at 660−920 cm−1 (Figures 3a−5a here and Figures S2 and S3). At the same time, the intense bands can be registered at 807 cm−1 (the out-ofplane C−H deformation vibrations of aromatics with two adjacent protons) and 725 cm−1 (the rocking vibrations of −(CH2)n− alkyl chains with n ≥ 4). Thus, to estimate the degree of aromatics substitution and condensation, we used a similar ratio S [(1H + 2H)/(3H + 4H)] instead of S (1H/4H) index. The new relationship is determined as a ratio of the sum of integral intensities of the bands at 858 and 807 cm−1 (out-ofplane Carom−H deformation modes of aromatics with one and two adjacent protons) to the bands at ∼750 cm−1 (the out-ofplane Carom−H deformation modes of aromatics with three and four adjacent protons). The relative integral intensities of the bands have been determined by a well-known procedure via deconvolution of each FTIR spectrum (Figures S2 and S3). Adjusted R-squared values were 0.995−0.997. The S [(1H + 2H)/(3H + 4H)] index is shown in Table 4 (line 4) for all products obtained for different alcohols and SCW. The values vary within the ranges of 1.5−1.8, 2.4−3.0, and 2.3−3.2 for HSF, BSF, and IR, respectively, but exhibit no common trend. The same index of the initial asphaltenes is equal to 2.47. This index for benzene-soluble fraction is higher than corresponding value for HSF and also IA. The latter fact is worth attention. It means that the BSF are composed of moresubstituted (as compared to the HSF) or less-condensed (as compared with the IA) aromatics. These results corroborate the NMR data, mainly the Carom-to-Harom ratios (see Table 3). Empirical indices of the carbonyl-to-aliphatics ratio and of carbonyl abundance were evaluated after preliminary deconvolution of the spectra (deconvoluted spectrum of the IA, as an example, is shown in Figures S2 and S3). All of these molecular parameters are shown in Table 4 (lines 5 and 6, respectively). The carbonyl-to-aliphatics ratio for HSF, BSF, and IR (line 5, Table 4) varies within the range of 0.03−0.05, 0.08−0.11, and 0.06−0.07, respectively. The negligibility and small range of the parameter variation observed for different fractions attracted our attention. On the contrary, the processing of the same asphaltenes by SCW leads to a 30-fold increase of this ratio. The latter phenomenon observed is probably connected with well-known oxidation reaction of aliphatics by supercritical water (so-called supercritical water oxidation, SCWO).74 Another parameter, the index of carbonyl abundance, has essential values that vary within the wider ranges of 0.52−0.65, 0.42−0.63, and 0.25−0.29 for HSF, BSF, and IR (line 6 in Table 4), respectively. The same index for initial asphaltenes (IA) is equal to 0.31. In our opinion, the aromatics-to-aliphatics ratios and the index of carbonyl abundance (lines 3 and 6, Figure 4) interrelate with each other. Indeed the CO groups preferentially are located in aliphatic chains of the product molecules. Consequently the more aliphatics in the products, the larger amount of CO groups. Thus, the amount of

4. CONCLUSIONS The transformation of heavy oil asphaltenes in supercritical lower alcohols (methanol, ethanol, and 1- and 2-propanol) was studied in details for the first time using FTIR and NMR techniques. The ATR-FTIR and NMR (1H and 13C) data obtained about redistribution of some functional groups (C O, C−O, Carom−H, Carom=Carom, CH2, CH3, etc.) in the products provided the insights into the chemical transformation of asphaltenes in sc alcohols. The quantitative evaluation of the ratios of Carom/Caliph, Harom/Haliph, Caliph/Haliph and Carom/Harom for the products made by NMR in comparison with the information about molar ratios of nCH2 to mCH3, aromatics to aliphatics, carbonyls to aliphatics, and carbonyl abundance in the products obtained via ATR-FTIR spectra analysis allowed us to establish the structural difference between the initial asphaltenes and the products for each alcohol tested in comparison to the transformation observed in supercritical water. The main routes of the transformations of asphaltenes in sc alcohols were identified and characterized. It was shown that the aliphatics are the main component of the soluble fractions (HSF and BSF) obtained after processing of asphaltenes, while the insoluble residues are composed mainly of polyaromatic compounds. The aliphatics of all fractions obtained after processing of asphaltenes in lower alcohols and SCW are similar to each other. The experimental data unambiguously point to the intensive alkoxylation of the initial asphaltenes by lower alcohols at 350 °C, especially by the supercritical methanol; the alcohols used are incorporated in the product molecules as alkoxy substituents in aromatic ethers Ar−OAlk. Thus, the alkylation and dealkylation and alkoxylation reactions make a crucial contribution to the chemical transformations of the asphaltenes in supercritical alcohols. Because the insoluble residue is the main fraction, which is composed of polycyclic highly condensed aromatic hydrocarbons having a relatively small number of alkyl substituents (the aliphatics), it seems necessary to provide a hydrogenation process of the asphaltenes during the conversion. As to the viability of the asphaltene processing using sc alcohols, the promising way to decrease the IR content has simultaneous use as a catalyst and appropriate source of the hydrogen (conventional and nonconventional ones). In any case, the 2124

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article

Energy & Fuels investigation of the impact of alcohol molecules on different chemical or molecular fragments of the initial asphaltenes at elevated temperatures and pressures is crucial to developing new and efficient processes of asphaltene upgrading.





REFERENCES

(1) Schabron, J. F.; Pauli, A. T.; Rovani, J. F.; Miknis, F. P. Predicting coke formation tendencies. Fuel 2001, 80, 1435−1446. (2) Speight, J. G. Fuel science and technology handbook; Marcel Dekker: New York, 1990. (3) Speight, J. G. The desulfurization of heavy oils and residua; Marcel Dekker: New York, 1981. (4) Catalytic hydroprocessing of petroleum and distillates; Oballa, M. C., Shih, S. S., Eds.; Marcel Dekker: New York, 1994. (5) Herod, A. A.; Bartle, K. D.; Kandiyoti, R. Characterization of heavy hydrocarbons by chromatographic and mass spectrometric methods: an overview. Energy Fuels 2007, 21, 2176−2203. (6) Klein, G. C.; Kim, S.; Rodgers, R. P.; Marshall, A. G.; Yen, A. Mass spectral analysis of asphaltenes. II. Detailed compositional comparison of asphaltenes deposit to its crude oil counterpart for two geographically different crude oils by ESI FT−ICR MS. Energy Fuels 2006, 20, 1973−1979. (7) Speight, J. G. Application of spectroscopic techniques to the structural analysis of petroleum. Appl. Spectrosc. Rev. 1994, 29, 269− 307. (8) Acevedo, S.; Guzman, K.; Ocanto, O. Determination of the number average molecular mass of asphaltenes (Mn) using their soluble A2 fraction and the vapor pressure osmometry (VPO) technique. Energy Fuels 2010, 24, 1809−1812. (9) Acevedo, S.; Gutierrez, L. B.; Negrin, G.; Pereira, J. C.; Mendez, B.; Delolme, F.; Dessalces, G.; Broseta, D. Molecular weight of petroleum asphaltenes: a comparison between mass spectrometry and vapor pressure osmometry. Energy Fuels 2005, 19, 1548−1560. (10) Trejo, F.; Ancheyta, J.; Morgan, T. J.; Herod, A. A.; Kandiyoti, R. Characterization of asphaltenes from hydrotreated products by SEC, LDMS, LAMDI, NMR, and XRD. Energy Fuels 2007, 21, 2121− 2128. (11) Morozov, E. V.; Martyanov, O. N. Probing flocculant-induced asphaltene precipitation via NMR imaging: from model toluene− asphaltene systems to natural crude oils. Appl. Magn. Reson. 2016, 47, 223−235. (12) Headen, T. F.; Boek, E. S.; Stellbrink, J.; Scheven, U. M. Small angle neutron scattering (SANS and V-SANS) study of asphaltene aggregates in crude oil. Langmuir 2009, 25, 422−428. (13) Larichev, Y. V.; Nartova, A. V.; Martyanov, O. N. The influence of different organic solvents on the size and shape of asphaltene aggregates studied via small-angle X-ray scattering and scanning tunneling microscopy. Adsorpt. Sci. Technol. 2016, 34, 244−257. (14) Groenzin, H.; Mullins, O. C. Chapter 2. Asphaltene molecular size and weight by Time-resolved fluorescence depolarization. In Asphaltenes, Heavy Oils, and Petroleomics; Mullins, O. C., Sheu, E. Y., Hammami, A., Marshall, A. G., Eds.; Springer: New York, 2007; pp 17−62. (15) Gabrienko, A. A.; Subramani, V.; Martyanov, O. N.; Kazarian, S. G. Correlation between asphaltene stability in n-heptane and crude oil composition revealed with in situ chemical imaging. Adsorpt. Sci. Technol. 2014, 32, 243−256. (16) Gabrienko, A. A.; Martyanov, O. N.; Kazarian, S. G. Effect of temperature and composition on the stability of crude oil blends studied with chemical imaging in situ. Energy Fuels 2015, 29, 7114− 7123. (17) Trukhan, S. N.; Kazarian, S. G.; Martyanov, O. N. Electron Spin Resonance of slowly rotating vanadyls − effective tool to quantify the sizes of asphaltenes in situ. Energy Fuels 2017, 31, 387−394. (18) Trukhan, S. N.; Yudanov, V. F.; Gabrienko, A. A.; Subramani, V.; Kazarian, S. G.; Martyanov, O. N. In situ Electron Spin Resonance study of molecular dynamics of asphaltenes at elevated temperature and pressure. Energy Fuels 2014, 28, 6315−6321. (19) Morozov, E.; Trukhan, S.; Larichev, Y.; Subramani, V.; Gabrienko, A.; Kazarian, S.; Martyanov, O. In-situ studies of crude oil stability and direct visualization of asphaltenes aggregation

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.energyfuels.7b01630. A file containing the scheme and description of experimental setup, the detailed procedures of separation and partition of the products into the main fractions, their elemental analysis (C, H, N, S, and O), and NMR spectra and tables showing the data of elemental analysis and obtained experimental range in element contents in the fractions. (ZIP)



SCW = supercritical water SCWO = supercritical water oxidation

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +7-383-326-97-30. Fax: +7-383-330-97-52. ORCID

Andrey M. Chibiryaev: 0000-0002-4566-579X Oleg N. Martyanov: 0000-0001-9999-8680 Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This research was performed within project no. 15-19-00119 of the Russian Science Foundation. NOMENCLATURE Caliph = total molar content of aliphatic carbon atoms (wt %) Carom = total molar content of aromatic carbon atoms (wt %) Haliph = total molar content of aliphatic hydrogen atoms (wt %) Harom = total molar content of aromatic hydrogen atoms (wt %) Pcr = critical pressure (atm) S[(1H + 2H)/(3H + 4H)] index = degree of aromatic substitution and condensation is defined a ratio of integrated intensities (peak areas) of (ν858 + ν807) bands of aromatics with one and two adjacent protons and of ν750 bands of aromatics with three and four adjacent protons, respectively Tcr = critical temperature (°C) Torr = unit of pressure, ∼1.316 × 10−3 atm

Greek symbols

δC−H = C−H deformation vibration ρcr = critical density (g·cm−3)

Abbreviations

adjusted R2 = adjusted R-squared value I = intensity sc = supercritical Acronyms

ASF = alcohol-soluble fraction ATR = attenuated total reflection BSF = benzene-soluble fraction HSF = hexane-soluble fraction IA = initial asphaltenes IR = insoluble residue 2125

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article

Energy & Fuels processes via some spectroscopy techniques. In Abstracts of Papers of the 248th National Meeting of the American Chemical Society; American Chemical Society: San Francisco, CA, 2014. (20) Trejo, F.; Ancheyta, J.; Rana, M. S. Structural characterization of asphaltenes obtained from hydroprocessed crude oils by SEM and TEM. Energy Fuels 2009, 23, 429−439. (21) Ancheyta, J.; Centeno, G.; Trejo, F.; Marroquín, G. Changes in asphaltene properties during hydrotreating of heavy crudes. Energy Fuels 2003, 17, 1233−1238. (22) Ancheyta, J.; Centeno, G.; Trejo, F. Effects of catalyst properties on asphaltenes composition during hydrotreating of heavy oils. Pet. Sci. Technol. 2004, 22, 219−225. (23) Bartholdy, J.; Andersen, S. I. Changes in asphaltene stability during hydrotreating. Energy Fuels 2000, 14, 52−55. (24) Bartholdy, J.; Lauridsen, R.; Mejlholm, M.; Andersen, S. I. Effect of hydrotreatment on product sludge stability. Energy Fuels 2001, 15, 1059−1062. (25) Gawel, I.; Bociarska, D.; Biskupski, P. Effect of asphaltenes on hydroprocessing of heavy oils and residua. Appl. Catal., A 2005, 295, 89−94. (26) Ancheyta, J.; Trejo, F.; Rana, M. S. Asphaltenes: chemical transformation during hydroprocessing of heavy oils; CRC Press (Taylor & Francis Group): New York, 2009. (27) Ancheyta, J.; Centeno, G.; Trejo, F.; Speight, J. G. Asphaltene characterization as function of time on-stream during hydroprocessing of Maya crude. Catal. Today 2005, 109, 162−166. (28) Hosseinpour, M.; Fatemi, S.; Ahmadi, S. J. Catalytic cracking of petroleum vacuum residue in supercritical water media: impact of αFe2O3 in the form of free nanoparticles and silica-supported granules. Fuel 2015, 159, 538−549. (29) Nhieu, P.; Liu, Q.; Gray, M. R. Role of water and fine solids in onset of coke formation during bitumen cracking. Fuel 2016, 166, 152−156. (30) Sato, T. Upgrading of heavy oil by hydrogenation through partial oxidation and water-gas shift reaction in supercritical water. J. Jpn. Pet. Inst. 2014, 57, 1−10. (31) Vilcáez, J.; Watanabe, M.; Watanabe, N.; Kishita, A.; Adschiri, T. Hydrothermal extractive upgrading of bitumen without coke formation. Fuel 2012, 102, 379−385. (32) Sato, T.; Trung, P. H.; Tomita, T.; Itoh, N. Effect of water density and air pressure on partial oxidation of bitumen in supercritical water. Fuel 2012, 95, 347−351. (33) Kim, D. W.; Koriakin, A.; Lee, C.-H. A parameter study for coprocessing of petroleum vacuum residue and oil palm empty fruit bunch fiber using supercritical tetralin and decalin. Fuel 2016, 181, 895−904. (34) Xu, C. C.; Su, H.; Ghosh, M. Hydro-treating of asphaltenes in supercritical toluene with MgO-supported Fe, Ni, NiMo, and CoMo catalysts. Energy Fuels 2009, 23, 3645−3651. (35) Gai, X.-K.; Arano, H.; Lu, P.; Mao, J.-W.; Yoneyama, Y.; Lu, C.X.; Yang, R.-Q.; Tsubaki, N. Catalytic bitumen cracking in sub- and supercritical water. Fuel Process. Technol. 2016, 142, 315−318. (36) Morimoto, M.; Sugimoto, Y.; Sato, S.; Takanohashi, T. Effect of supercritical water on desulfurization behavior of oil sand bitumen. J. Jpn. Pet. Inst. 2012, 55, 261−266. (37) Ates, A.; Azimi, G.; Choi, K. H.; Green, W. H.; Timko, M. T. The role of catalyst in supercritical water desulfurization. Appl. Catal., B 2014, 147, 144−155. (38) Fedyaeva, O. N.; Vostrikov, A. A.; Sokol, M. Y.; Fedorova, N. I. Hydrogenation of bitumen in supercritical water flow and the effect of zinc addition. Russ. J. Phys. Chem. B 2013, 7, 820−828. (39) Mandal, P. C.; Goto, M.; Sasaki, M. Removal of nickel and vanadium from heavy oils using supercritical water. J. Jpn. Pet. Inst. 2014, 57, 18−28. (40) Scott, D. S.; Radlein, D.; Piskorz, J.; Majerski, P.; DeBruijn, T. J. W. Upgrading of bitumen in supercritical fluids. Fuel 2001, 80, 1087− 1099.

(41) Viet, T. T.; Lee, J.-H.; Ryu, J. W.; Ahn, I.-S.; Lee, C.-H. Hydrocracking of vacuum residue with activated carbon in supercritical hydrocarbon solvents. Fuel 2012, 94, 556−562. (42) Yan, T.; Xu, J.; Wang, L.; Liu, Y.; Yang, C.; Fang, T. A review of upgrading heavy oils with supercritical fluids. RSC Adv. 2015, 5, 75129−75140. (43) Kozhevnikov, I. V.; Nuzhdin, A. L.; Martyanov, O. N. Transformation of petroleum asphaltenes in supercritical water. J. Supercrit. Fluids 2010, 55, 217−222. (44) Kim, D.-W.; Ma, F.; Koriakin, A.; Jeong, S.-Y.; Lee, C.-H. Parametric study for upgrading petroleum vacuum residue using supercritical m-xylene and n-dodecane solvents. Energy Fuels 2015, 29, 2319−2328. (45) Torrente, M. C.; Galán, M. A. Extraction of kerogen from oil shale (Puertollano, Spain) with supercritical toluene and methanol mixtures. Ind. Eng. Chem. Res. 2011, 50, 1730−1738. (46) Lodi, L.; Cárdenas Concha, V. O.; Medina, L. C.; Filho, R. M.; Wolf Maciel, M. R. Experimental study of a pilot plant deasphalting process in ethanol sub- and supercritical phase. Pet. Sci. Technol. 2015, 33, 550−555. (47) Demirbas, A. Deasphalting of crude oils using supercritical fluids. Pet. Sci. Technol. 2016, 34, 665−670. (48) Gabrienko, A. A.; Martyanov, O. N.; Kazarian, S. G. Behavior of asphaltenes in crude oil at high-pressure CO2 conditions: in situ Attenuated Total Reflection−Fourier Transform infrared spectroscopic imaging study. Energy Fuels 2016, 30, 4750−4757. (49) Gabrienko, A. A.; Morozov, E. V.; Subramani, V.; Martyanov, O. N.; Kazarian, S. G. Chemical visualization of asphaltenes aggregation processes studied in situ with ATR-FTIR spectroscopic imaging and NMR imaging. J. Phys. Chem. C 2015, 119, 2646−2660. (50) Calemma, V.; Iwanski, P.; Nali, M.; Scotti, R.; Montanari, L. Structural characterization of asphaltenes of different origins. Energy Fuels 1995, 9, 225−230. (51) Dong, X.-G.; Lei, Q.-F.; Fang, W.-J.; Yu, Q.-S. Thermogravimetric analysis of petroleum asphaltenes along with estimation of average chemical structure by nuclear magnetic resonance spectroscopy. Thermochim. Acta 2005, 427, 149−153. (52) Michon, L.; Martin, D.; Planche, J.-P.; Hanquet, B. Estimation of average structural parameters of bitumens by 13C nuclear magnetic resonance spectroscopy. Fuel 1997, 76, 9−15. (53) Liang, W.; Que, G.; Chen, Y.; Liu, C. Chapter 10. Chemical composition and characteristics of residues of Chinese crude oils. In Asphaltenes and Asphalts. 2. Developments in Petroleum Science; Yen, T. F., Chilingarian, G. V., Eds.; Elsevier Science: Amsterdam, The Netherlands, 2000; Vol. 40B, pp 281−304. (54) ASTM. ASTM D 6560-12 (revised 2012). Standard test method for determination of asphaltenes (heptane insolubles) in crude petroleum and petroleum products; ASTM: West Conshohocken, PA, 2012. (55) ASTM. ASTM D 5292-99 (reapproved 2004). Standard test method for aromatic carbon contents of hydrocarbon oils by high resolution nuclear magnetic resonance spectroscopy; ASTM: West Conshohocken, PA, 2004. (56) Coelho, R. R.; Hovell, I.; De Mello Monte, M. B.; Middea, A.; De Souza, A. L. Characterisation of aliphatic chains in vacuum residues (VRs) of asphaltenes and resins using molecular modelling and FTIR techniques. Fuel Process. Technol. 2006, 87, 325−333. (57) Bouquet, M. Determination of hydrogen content of petroleum products using a low resolution pulsed n.m.r. spectrometer. Fuel 1985, 64, 226−228. (58) Gautier, S.; Quignard, A. Low resolution 1H nuclear magnetic resonance. The ultimate tool for accurate determination of hydrogen content in petroleum products. Rev. Inst. Fr. Pet. 1995, 50, 249−282 (in French). (59) Fitch, J. C.; Barnes, M. Hydrocarbon analysis. In Fuels and lubricants handbook: technology, properties, performance, and testing; Totten, G. E., Westbrook, S. R., Shah, R. J., Eds.; ASTM International: West Conshohocken, PA, 2003; pp 649−674. (60) Castro, L. V.; Vazquez, F. Fractionation and characterization of Mexican crude oils. Energy Fuels 2009, 23, 1603−1609. 2126

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127

Article

Energy & Fuels (61) Pretsch, E.; Bü h lmann, P.; Badertscher, M. Structure determination of organic compounds: Tables of spectral data, 4th ed.; Springer-Verlag: Berlin−Heidelberg, Germany, 2009. (62) Structures and dynamics of asphaltenes; Mullins, O. C., Sheu, E. Y., Eds.; Springer US: Boston, MA, 1998. (63) Andersen, S. I. Separation of asphaltenes by polarity using liquid−liquid extarction. Pet. Sci. Technol. 1997, 15, 185−198. (64) Allamandola, L. J.; Tielens, A. G. G. M.; Barker, J. R. Interstellar polycyclic aromatic hydrocarbons − The infrared emission bands, the excitation/emission mechanism, and the astrophysical implications. Astrophys. J., Suppl. Ser. 1989, 71, 733−775. (65) Carbognani, L.; Espidel, J.; Izquierdo, A. Chapter 13. Characterization of asphaltenic deposits from oil production and transportation operations. In Asphaltenes and Asphalts. 2. Developments in Petroleum Science; Yen, T. F., Chilingarian, G. V., Eds.; Elsevier Science: Amsterdam, The Netherlands, 2000, Vol. 40B, pp 335−362. (66) Sominsky, L.; Rozental, E.; Gottlieb, H.; Gedanken, A.; Hoz, S. Uncatalyzed Meerwein−Ponndorf−Oppenauer−Verley reduction of aldehydes and ketones under supercritical conditions. J. Org. Chem. 2004, 69, 1492−1496. (67) Daimon, A.; Kamitanaka, T.; Kishida, N.; Matsuda, T.; Harada, T. Selective reduction of unsaturated aldehydes to unsaturated alcohols using supercritical 2-propanol. J. Supercrit. Fluids 2006, 37, 215−219. (68) Kargin, Y. F.; Ivicheva, S. N.; Buslaeva, E. Y.; Yurkov, G. Y.; Volodin, V. D. Reduction of various metal salts in opal matrices with supercritical isopropanol. Inorg. Mater. 2006, 42, 966−970. (69) Lermontov, S. A.; Shkavrov, S. V.; Kuryleva, N. V. Uncatalyzed Meerwein−Ponndorf−Verley reduction of trifluoromethyl carbonyl compounds by high-temperature secondary alcohols. J. Fluorine Chem. 2003, 121, 223−225. (70) Kamitanaka, T.; Matsuda, T.; Harada, T. Reduction of acetophenone using supercritical 2-propanol: the substituent effect and the deuterium kinetic isotope effect. Tetrahedron Lett. 2003, 44, 4551−4553. (71) Kamitanaka, T.; Matsuda, T.; Harada, T. Mechanism for the reduction of ketones to the corresponding alcohols using supercritical 2-propanol. Tetrahedron 2007, 63, 1429−1434. (72) Fossen, M.; Kallevik, H.; Knudsen, K. D.; Sjöblom, J. Asphaltenes precipitated by a two-step precipitation procedure. 2. Physical and chemical characteristics. Energy Fuels 2011, 25, 3552− 3567. (73) Ibrahim, H. H.; Idem, R. O. Correlations of characteristics of Saskatchewan crude oils/asphaltenes with their asphaltenes precipitation behavior and inhibition mechanisms: Differences between CO2and n-heptane-induced asphaltene precipitation. Energy Fuels 2004, 18, 1354−1369. (74) Watanabe, M.; Mochiduki, M.; Sawamoto, S.; Adschiri, T.; Arai, K. Partial oxidation of n-hexadecane and polyethylene in supercritical water. J. Supercrit. Fluids 2001, 20, 257−266.

2127

DOI: 10.1021/acs.energyfuels.7b01630 Energy Fuels 2018, 32, 2117−2127