E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33 23

Deasphalting of Olefin Pyrolysis Fuel Oil by Combination of Chemical–Physical Methods

E. S. Moosavi1, M. Gharibi*2, R. Karimzadeh*3, B. Asbaghi2, 2 3 3 S. S. Shahabi , S. Aalizad and M. Zare

1. Chemical and Materials Engineering Department, Buein Zahra Technical University, Buein Zahra, Qazvin, Iran. 2. National Petrochemical Company, Petrochemical Research and Technology Company, P.O. Box 1435884711, Tehran, Iran. 3. Chemical Engineering Faculty, Tarbiat Modares University, P.O. Box 14155-4838, Tehran, Iran.

(Received 14 August 2017, Accepted 26 February 2018) [DOI: 10.22059/jchpe.2018.239677.1207]

Abstract Keywords

Fuel Oil (PFO) thermal cracking of two PFO samples on liquid, solid This work investigated the effect of different conditions of Pyrolysis Asphaltene; Naphthalene; and gas product yields and asphaltene removal efficiency in a newly- Product yield; designed experimental setup. There is no need to use a catalyst, simple Pyrolysis fuel oil (PFO); operating system and experiment conditions. The ability to use wa Thermal cracking. ter as a cheap carrier gas and high asphaltene extraction efficiency- paredwithout in thevarious use of operating solvents conditionsare outstanding and the benefits optimum of this experimental method in upgrading PFO. The yields of liquid, solid and gas products were com- ing condition of PFO in terms of liquid yield and asphaltene removal in conditions were obtained. The results revealed the best thermal crack this setup for samples. The optimum conditions were 390 and 380°C rate.for the In reactorthese circumstances temperature aboutof PFO-1 70 and 53PFO-2, wt% respectively; of the liquid 150°Cprod- uct,for the 25 temperaturewt% of the solid of carrier products, gas and and 100 5 wt% ml/min of the for gascarrier product gas flow are generated and the asphaltene separation was reached about 95 and 96.5%.

1. Introduction cts so that their upgrading processes have gained worldwide interest which requires the use of he residues of production lines in petro- units to prevent high yields of coke and more chemical industry, such as pyrolysis fuel oil beneficial yields of distillate products in order to T in olefin plant, are the least valuable produ- convert oil residue into more desirable products. 1 Introduction A mixture of high boiling aromatic hydrocarbons, produced as a by-product during pyrolysis of oil * Corresponding Authors. fractions is a dark brown to black colored viscous liquid with pour point approximately 15 °C and a | Email: [email protected] (M. Gharibi) Tel.: +98 21 44787517; fax: +98 21 44787501 distinctive mothball-like odor classified as a haz- | Email: [email protected] (R. Karimzadeh) ardous substance with carcinogenic components Tel.: +98 21 82883315; fax: +98 21 88006544 24 E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33 according to EC directives and used as a raw ma- produce deasphalted liquid product, as a light terial for manufacturing carbon black or burned feedstock. as industrial fuels. The pyrolysis fuel oil (PFO) Solvent deasphalting (SDA), as another separa- feedstock contains about 30 to 35 (w/w %) of tion process in which the residue is separated by asphaltenes in matrix solution which consists the molecular weight (density) instead of boiling highest molecular weight and aromatic hetero point, would require a high volume of solvent compounds with aliphatic substitutions, the ma- which would allow us to deduce the fact that this jority of impurities and hetero elements, and low process cannot be regarded as a beneficial one, hydrogen to carbon ratios. Furthermore, it is the from an economic point of view [16]. Asphaltenes most complicated fraction which contains more are not crystallized and cannot be divided into than thousands different molecules [1-7]. individual components or narrow fractions. In the last few years, characterization of asphal- Thermal cracking of heavy oils, especially the re- tene precipitation has been investigated [8-10]. sidual feedstock, and investigation of the yields While asphaltenes act as coke precursors which and properties of the product of this process in would lead to catalyst deactivation, the compli- different conditions and severities have been the cated macromolecular structure of asphaltene point of focus and interest in most of the previ- strongly affects upgrading and refining opera- ously conducted studies [9, 10, 17-20]. Neverthe- tions and determines the quality and processing less, due to the high content of asphaltenes and effect of PFO to a certain degree. Problems asso- aromatic compounds in the range of naphthalene ciated with asphaltene precipitation and deposi- in PFO, it is not possible to upgrade this new ma- tion have been widely reported in the petroleum terial (i.e. PFO) in a single step by the currently production and processing industry [11-14]. En- available methods (such as Eruka process). hanced oil recovery techniques have also led to Therefore, the primary process should include some asphaltene problems. These these com- the elimination of these troublesome compounds. pounds could be prone to condensation and cok- However, the use of a new system is necessary in ing during processing and would deactivate cata- order to resolve the problem of naphthalenic lysts. compounds which could eclipse and stop the op- In recent years, several technologies have been eration because of their deposition along the developed for residual oil upgrading and several path. researchers have been trying to improve the pro- The main concern and focal point of this research cessing technologies of heavy fuel oils by reduc- would be the study of PFO upgrading, olefin plant ing asphaltene because it has been proved that by-product, and using a newly designed setup in one of the most important problems related to order to remove asphaltene from two PFO sam- governing processes of these technologies would ples and, in the meantime, naphthalenes removal be high content of resins and asphaltenes. from one of the PFO samples to the next upgrad- Thermal cracking processes which could be used ing processes and obtain valuable compounds. in the refining industry, including delayed coking, The innovative aspect of the newly designed ex- fluid coking, flexi-coking, visbreaking, and Eureka perimental setup is the use of a special condenser process, not only do facilitate the operation to for the separation of aromatic compounds such as have a condition of low pressure and high to naphthalene because these materials immediately moderate working temperatures with no demand result in condenser congestion during exiting the for expensive catalysts, but also provide us with a reactor. This section has not been predicted in simple robust control operation [15]. previous studies. Therefore, this setup can also be used to refine the compounds containing naph- It could be possible to remove the asphaltene thalene. fraction, as the precursors of coke formation, from heavy and residual feedstock by Eureka While generating the maximum amount of liquid process which is a commercially proven thermal product containing the minimum amount of as- cracking process that produces deasphalted liq- phaltene has been another objective of this study, uid product, as a light feedstock and aromatic pe- the yields of liquid, solid and gas products have troleum pitch, from vacuum residues in order to been compared in various operating conditions and an optimum experimental condition is ob- E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33 25 according to EC directives and used as a raw ma- produce deasphalted liquid product, as a light tained under different experimental conditions, specified inside temperature in a heating mantle terial for manufacturing carbon black or burned feedstock. asphaltenes are precipitated in a rejected fuel oil equipped with a temperature controller and insu- as industrial fuels. The pyrolysis fuel oil (PFO) with light alkane precipitants (i.e., n-pentane, and lated with glass wool jacket, was used as a ther- Solvent deasphalting (SDA), as another separa- feedstock contains about 30 to 35 (w/w %) of n-heptane) and characterized using some com- mal cracking reactor. In each test, the reactor was tion process in which the residue is separated by asphaltenes in matrix solution which consists the mon analytical techniques. loaded with 100 g of PFO. Pre-heated carrier gas molecular weight (density) instead of boiling highest molecular weight and aromatic hetero (N or Steam) stream was injected at a constant point, would require a high volume of solvent Due to the variety of feed in petrochemical plants 2 compounds with aliphatic substitutions, the ma- flow rate into the reactor to evacuate the prod- which would allow us to deduce the fact that this such as naphtha, gaseous feeds: propane, ethane, jority of impurities and hetero elements, and low ucts immediately after formation and to mix the process cannot be regarded as a beneficial one, etc., two olefin plants were selected from each hydrogen to carbon ratios. Furthermore, it is the feed. Two design approaches were applied for from an economic point of view [16]. Asphaltenes category. Having prepared the two samples of most complicated fraction which contains more two different PFO samples upgrading process. In are not crystallized and cannot be divided into PFOs from different Iranian petrochemical Com- than thousands different molecules [1-7]. the first setup, three condensing steps were uti- individual components or narrow fractions. panies, the produced gas, liquid (deasphalted lized. Because PFO-1 contains compounds, which In the last few years, characterization of asphal- maltenes) and solid fractions characterized by Thermal cracking of heavy oils, especially the re- caused condenser congestion, and for the separa- tene precipitation has been investigated [8-10]. variety of analytical techniques while the ob- sidual feedstock, and investigation of the yields tion of these compounds it is necessary to use a While asphaltenes act as coke precursors which tained naphthalene from PFO-1 is characterized and properties of the product of this process in balloon condenser placed before the tube con- would lead to catalyst deactivation, the compli- by Gas Chromatography-Mass Spectrometry (GC- different conditions and severities have been the denser. In the first condenser (WB), which was cated macromolecular structure of asphaltene MS) solid product containing asphaltene, gas point of focus and interest in most of the previ- immersed in a water bath, a two-neck flask was strongly affects upgrading and refining opera- product, and the organic components are ana- ously conducted studies [9, 10, 17-20]. Neverthe- utilized to remove the high-temperature conden- tions and determines the quality and processing lyzed using X-Ray Fluorescence (XRF), gas chro- less, due to the high content of asphaltenes and sable product portion of PFO-1 to prevent plug- effect of PFO to a certain degree. Problems asso- matography and Detailed Hydrocarbon Analysis aromatic compounds in the range of naphthalene ging of the coil condenser. The remained non- ciated with asphaltene precipitation and deposi- (DHA), respectively. Physical properties of the in PFO, it is not possible to upgrade this new ma- condensed vapor was passed through the con- tion have been widely reported in the petroleum liquid product include n-heptane insoluble frac- terial (i.e. PFO) in a single step by the currently denser (C1) and ice-water bath (G1, G2) to recover production and processing industry [11-14]. En- tion, refractive index, density, and viscosity were available methods (such as Eruka process). light fractions of the oil product. The non- hanced oil recovery techniques have also led to investigated. Therefore, the primary process should include condensable gas products and inert carrier gas some asphaltene problems. These these com- the elimination of these troublesome compounds. (N2 gas) were purged to flare. In the second setup, pounds could be prone to condensation and cok- However, the use of a new system is necessary in only the first condenser (WB), was eliminated ing during processing and would deactivate cata- order to resolve the problem of naphthalenic 2. Experimental Methods from the upgrading. This setup was utilized for lysts. compounds which could eclipse and stop the op- upgrading of PFO-2. Each experimental run pro- 2.1. Materials In recent years, several technologies have been eration because of their deposition along the duced three products: liquid, gas and solid oil. developed for residual oil upgrading and several path. The proposed PFO samples (sample #1: PFO-1 The reactor was purged with N , for leakage test- researchers have been trying to improve the pro- and sample #2: PFO-2) were waste materials 2 The main concern and focal point of this research ing, and then it was subjected to a pre-specified cessing technologies of heavy fuel oils by reduc- which were obtained from two different Iranian would be the study of PFO upgrading, olefin plant inside temperature. The effect of four parameters ing asphaltene because it has been proved that petrochemical olefin plants. PFO-1 and PFO-2 by-product, and using a newly designed setup in including reactor temperature (350-450 °C), N2 one of the most important problems related to were the by-products of olefin plants with naph- order to remove asphaltene from two PFO sam- flow rate (50,100, 150 ml/min), carrier gas nozzle governing processes of these technologies would tha and gas mixture feedstock, respectively. Giv- ples and, in the meantime, naphthalenes removal position, and carrier gas type (N2 or steam) on be high content of resins and asphaltenes. en that the feeds of the petrochemicals producing asphaltene removal rate from PFO, time of ther- from one of the PFO samples to the next upgrad- these two PFO samples are different, their charac- mal cracking, and product yields were studied. Thermal cracking processes which could be used ing processes and obtain valuable compounds. teristics, including the asphaltene content, are in the refining industry, including delayed coking, The innovative aspect of the newly designed ex- also different. All chemical solvents such as pen- fluid coking, flexi-coking, visbreaking, and Eureka perimental setup is the use of a special condenser tane, heptane, acetone (with 99% purity), used in process, not only do facilitate the operation to for the separation of aromatic compounds such as the present investigation, were purchased from 2.3. Characterization methods have a condition of low pressure and high to naphthalene because these materials immediately Merck. Steam was produced from deionized wa- moderate working temperatures with no demand result in condenser congestion during exiting the The n-heptane insoluble fraction of PFO and liq- ter. The purification of N2 cylinder was 99.9%. for expensive catalysts, but also provide us with a reactor. This section has not been predicted in uid product, as the asphaltene fraction, density, simple robust control operation [15]. previous studies. Therefore, this setup can also be refractive index (ATAGO NAR-1L LIQUID append- used to refine the compounds containing naph- ing ASTM D1218) and viscosity (Normal ab serial It could be possible to remove the asphaltene thalene. 2.2. Experimental setup and procedure 75 in a bath of wisdom model wvb-30 appending fraction, as the precursors of coke formation, ASTM D 446) of liquid products were determined The PFO thermal cracking experiments were per- from heavy and residual feedstock by Eureka While generating the maximum amount of liquid according to ASTM standards D3279, D1481, formed using a bench scale experimental setup, a process which is a commercially proven thermal product containing the minimum amount of as- D1218 and D445, respectively. The viscosity of detailed schematic diagram of which is shown in cracking process that produces deasphalted liq- phaltene has been another objective of this study, liquid products was measured three times and an Fig. 1. Three-neck Pyrex flask, with an internal uid product, as a light feedstock and aromatic pe- the yields of liquid, solid and gas products have average value of the measured viscosity data was capacity of 500 ml, which was subjected to a pre- troleum pitch, from vacuum residues in order to been compared in various operating conditions noted. and an optimum experimental condition is ob- 26 E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33

Figure 1. Schematic diagram of thermal cracking experimental setup: (V-1) N2 Regulator, (V-2, and V-3) Valve; (SG) Steam Generator; (F-1) Gas Flow meter: (H-1) Carrier gas pre-heater; (R-1) Thermal Cracking Reactor; (T-1) Reactor Temperature Controller; (WB) Water Bath; (C-1) Condenser; (G-1, G-2) collect bottles for liquid product; (F-2) Flare.

PFO-1 PFO-2 CO, and CO2; and TCD-2 for detection of H2 and Density,Viscosity,Refractive He. All characterization methods which were Index,Molecular Weight,GC- DHA,Asphaltene Content,Sulfur used in this study are shown in Fig. 2. Content

Thermal Cracking 3. Results and Discussion (Steam, N2) 3.1. Reactor tests Due to the difference in composition of the two Outlet Liquid Solid PFO samples, different results were observed Gas Product fraction during their upgrading tests. PFO-2 contained some water which should be removed during the first stage of upgrading. After adjusting the tem- XRF, GC GC-DHA analysis, GC Mass, Asphaltene Analysis Asphaltene Content perature of the mantle at 200 °C, reactor temper- Yeild ature continuously increased to reach 100 °C, Figure 2. Schematic diagram of characterization methods which is the temperature of PFO's water separa- tion. At the temperature range of 70-80 °C, a trace of vapor was observed in the condenser. A con- The PFO samples and liquid hydrocarbon prod- tinuous weak flame which was related to the ucts composition were determined by gas chro- burned compounds was also detected. Increasing matography (GC-DHA; Varian CP-3800, capillary temperature to about 90 °C resulted in observa- column SilPona CB) equipped with flame ioniza- tion of a yellow liquid in the condenser, in addi- tion detector (FID), along with GC-mass spectros- tion to water vapor which resulted in reduced copy (GC-MS; Varian CP-3800, Column CP-Sil 8) flame volume. During water separation stage, the for quantitative and qualitative analyses of com- temperature remained constant for about 10 min. pounds in the fraction, which was separated as After removal of all water, the liquid product was naphthalene at 215-225 °C. In order to prevent collected at a temperature about 115 °C. A brown GC contamination, a filter was installed before it liquid has been produced due to upgrading reac- so as to collect a small amount of unconverted tions which flows at a temperature range from liquid hydrocarbons. The concentrations of gas 120 °C to 200 °C, in the condenser. hydrocarbon products were determined by gas chromatography (GC; Agilent-7890A, column HP- A decreasing temperature gradient was observed Al/S), equipped with three detectors: FID for de- in the aforementioned reactions which could be tection of CH4, C2H4, C2H6, C3H6, etc.; thermal con- proved to be a result of some endothermic reac- ductivity detector TCD-1 for detection of N2, Ar, tions in some temperature ranges. At the end of PFO-2 test, the water in the liquid product was E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33 27

isolated and the rates of dry products were re- putting down nitrogen in the feed could result in ported. The deasphalting test of PFO-1 was dif- a significant increase in the rate of liquid product, ferent from PFO-2 test may be because PFO-1 had (as shown in Fig. 4). Since PFO samples are highly been dried. viscous, putting down carrier gas nozzle in the feed helps uniforming and homogenizing the In these experiments, unlike PFO-2 tests, at a sample during cracking which has resulted in bet- temperature of about 220 °C, flames could be ter cracking operations due to inert gas flow and seen. Tests duration of the two samples were al- bubble formation within the sample. most similar.

Generally, increasing the temperature and the carrier gas flow rate led to the reduction of test 3.1.3. Carrier gas flow rate effect duration. In carrier gas flow rate of 100 ml/min, The carrier gas flow rate is one of the effective Figure 1. Schematic diagram of thermal cracking experimental setup: (V-1) N2 Regulator, (V-2, and V-3) Valve; (SG) after 100 min of the test, flame was observed. The parameters which influences products yield. Yield Steam Generator; (F-1) Gas Flow meter: (H-1) Carrier gas pre-heater; (R-1) Thermal Cracking Reactor; (T-1) Reactor results indicated that, although carrier gas did not Temperature Controller; (WB) Water Bath; (C-1) Condenser; (G-1, G-2) collect bottles for liquid product; (F-2) Flare. have any role in reactions, it acted like a mixer of liquid, gas and solid products, as a function of N flow rate, are shown in Table 1. By increasing and increased the penetration rate of molecules 2 and test rate. the N2 flow rate from 50 to 150 ml/min, the liquid yield increased at first and then it slightly de- PFO-1 PFO-2 CO, and CO2; and TCD-2 for detection of H2 and Density,Viscosity,Refractive He. All characterization methods which were creased. Solid and gas products yield decreased Index,Molecular Weight,GC- used in this study are shown in Fig. 2. with increasing N2 flow rate. The best experi- DHA,Asphaltene Content,Sulfur 3.1.1. Reactor temperature effect Content mental condition is observed when the liquid The liquid, gas and solid yields (wt %), obtained yield is at its maximum level and the asphaltene at the different temperatures, are presented in quantity is at the lowest level. In this work, the Thermal Cracking 3. Results and Discussion (Steam, N2) Fig. 3. It can be seen that the liquid yield in- high quantity and quality of the liquid product 3.1. Reactor tests creased by temperature while the solid yield de- were achieved at 390 °C and N2 flowrate of 100 creased with an increase in the temperature. With ml/min. Due to the difference in composition of the two increasing temperature, physically breaking the N was replaced by steam in some experiments Outlet Liquid Solid PFO samples, different results were observed large molecules bonds into smaller ones would be 2 Gas Product fraction with optimum condition (390 °C & 100 ml during their upgrading tests. PFO-2 contained easier. This factor facilitates the production of N /min). Given that the inert gas has not taken some water which should be removed during the light liquid samples. 2 first stage of upgrading. After adjusting the tem- part in cracking reaction. The results showed that XRF, GC GC-DHA analysis, GC Mass, Asphaltene The gas yield remained partly constant by tem- the steam did not highly change the products Analysis Asphaltene Content perature of the mantle at 200 °C, reactor temper- Yeild ature continuously increased to reach 100 °C, perature; however, the minimum portion of yields. For example, the asphaltene quantity of products devoted to gas and the lowest gas was the liquid product was higher when steam was Figure 2. Schematic diagram of characterization methods which is the temperature of PFO's water separa- tion. At the temperature range of 70-80 °C, a trace obtained at 380 °C and 390 °C. These observa- used as carrier gas. Since the PFO samples are of vapor was observed in the condenser. A con- tions showed that temperature is one of the most highly viscous, increasing the carrier gas flow rate The PFO samples and liquid hydrocarbon prod- tinuous weak flame which was related to the important factors affecting thermal cracking effi- helps uniforming and homogenizing the sample ucts composition were determined by gas chro- burned compounds was also detected. Increasing ciency. This also corresponds to Sawarkar et al. during cracking which has resulted in better matography (GC-DHA; Varian CP-3800, capillary temperature to about 90 °C resulted in observa- results [21]. Based on the obtained results, 390 °C cracking operations due to inert gas flow and column SilPona CB) equipped with flame ioniza- tion of a yellow liquid in the condenser, in addi- and 380 °C were chosen as operating tempera- bubble formation within the sample. tures for upgrading of PFO-1 and PFO-2, respec- tion detector (FID), along with GC-mass spectros- tion to water vapor which resulted in reduced copy (GC-MS; Varian CP-3800, Column CP-Sil 8) flame volume. During water separation stage, the tively temperature remained constant for about 10 min. 3.2. Characterization results for quantitative and qualitative analyses of com- pounds in the fraction, which was separated as After removal of all water, the liquid product was 3.2.1. Feedstock characterizations 3.1.2. The effect of carrier gas nozzle loca- naphthalene at 215-225 °C. In order to prevent collected at a temperature about 115 °C. A brown tion The asphaltene contents of original samples were GC contamination, a filter was installed before it liquid has been produced due to upgrading reac- measured using ASTM standard D 3279-01 and so as to collect a small amount of unconverted tions which flows at a temperature range from Two different positions are considered for carrier filter papers (Whatman No. 2, England) with a liquid hydrocarbons. The concentrations of gas 120 °C to 200 °C, in the condenser. gas nozzle in this study. It is an important process hydrocarbon products were determined by gas parameter affecting the products' yield. chromatography (GC; Agilent-7890A, column HP- A decreasing temperature gradient was observed was mixed with 40 volumes of liquid precipitant Al/S), equipped with three detectors: FID for de- in the aforementioned reactions which could be After fixing the other factors, the nozzle location (eitherpore size n- pentaneof 2 μm or[22]. n-heptane). First, one It volume is worthwhile of PFO tection of CH4, C2H4, C2H6, C3H6, etc.; thermal con- proved to be a result of some endothermic reac- parameter was placed out of reaction feed for the to note that different liquid precipitant-to-oil vol- ductivity detector TCD-1 for detection of N2, Ar, tions in some temperature ranges. At the end of first experiment series, but it was located into the ume ratios were tested to study the volume-ratio PFO-2 test, the water in the liquid product was feed in the other test series. It is observed that effect on the asphaltene yield. The purity of both 28 E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33 n-pentane and n-heptane, used as respective pre- the filter, lost its color. The precipitated asphal- cipitants, were 99.9%. It is found that precipitant tenes were slowly dried at 100 °C, on the heated has a strong influence on the yield and physico- magnetic stirrer until their weight stop changing chemical properties of the precipitated asphal- followed by reading of an electric balance (Met- tenes and these new techniques provide us with a tler Toledo, Canada). On the other hand, the fil- promising opportunity in order for the rejected trate was distilled at the boiling-point tempera- fuel oils to be transformed into the feed of petro- ture of either n-pentane or n-heptane, under the chemical plants in a beneficial economic prespec- atmospheric pressure for at least 2 h. The re- tive. mainder was the deasphalted heavy oil (i.e., mal- tenes).

Table 1. Yield of liquid, gas and solid products (wt%) as a function of N2 flow rate (ml/min).

N2 Flow 50 100 150 PFO-1 70 73 72 Liquid Yield PFO-2 41 53 51 PFO-1 24 22 22 Solid Yield PFO-2 45 35 35 PFO-1 6 5 5 Gas Yield PFO-2 14 12 15

For GC-DHA analysis of the feedstock, each sam- Figure 3. Dependence of gas, liquid, and solid products ple was mixed with acetone for 24 h. After filtra- yield on reactor temperature. tion, dissolved portions of all samples were used for analysis. The composition of PFO fraction was

very complex and it was a mixture of paraffins, 55 naphthenes, aromatics, and olefins. Table 2 shows the most common components along with the 50 physicochemical properties in the two proposed feedstocks. According to Table 2, sulfur content of 45 PFO-1 was more than PFO-2 and aromatics con- Liquid Product(wt%) Liquid 40 tent of PFO-1 was much more than PFO-2. How- ever, PFO-2 had significant water content where- 35 as the other feedstock did not have any water content. Due to this reason, for n-heptane insolu- 30 ble fraction content of PFO-2 two values, wet and 370 375 380 385 390 dry, were reported. The n-heptane insoluble frac- Temperature(oC) tion can be considered the same as the asphaltene In out content. Hence, it is possible to use asphaltene content as a comparison index of the two feed- Figure 4. Changes in the liquid product yields of PFO at stocks. This fraction contains asphaltene and different temperatures by changing the nozzle position. compounds that would be capable of being con- verted to asphaltene in the future. It was ob- served that PFO-2 had a high volume of olefin and The PFO–n-pentane or n-heptane mixture was iso-paraffin compounds. It should be noted that, agitated using a magnetic stirrer (SP46925, Barn- in case of PFO-1, the quantity of olefines was neg- stead/ Thermolyne Corporation, USA) for 12 h ligible and there was no iso-paraffins. Although the compositions of these feedstocks are so dif- (Whatman, England). The filter cake, mainly ferent, their dry asphaltene contents are almost composedand then filtered of precipitated with 2 μm asphaltenes, pore size filter was papersrinsed the same. with a liquid precipitant. The rinsing process con- tinued until the precipitant, which comes out of E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33 29 n-pentane and n-heptane, used as respective pre- the filter, lost its color. The precipitated asphal- cipitants, were 99.9%. It is found that precipitant tenes were slowly dried at 100 °C, on the heated Table 2. PFO samples properties. has a strong influence on the yield and physico- magnetic stirrer until their weight stop changing 3.2.2. Liquid product characterization chemical properties of the precipitated asphal- followed by reading of an electric balance (Met- Properties PFO-1 PFO-2 Property analysis tenes and these new techniques provide us with a tler Toledo, Canada). On the other hand, the fil- S(ppm) 1470 1038 promising opportunity in order for the rejected trate was distilled at the boiling-point tempera- Aromatics (%wt) 86.492 27.841 The results of bulk analyses performed on the fuel oils to be transformed into the feed of petro- ture of either n-pentane or n-heptane, under the Olefines (%wt) 0.344 20.371 liquid product, obtained from the thermal crack- Paraffins (%wt) 1.77 8.873 chemical plants in a beneficial economic prespec- atmospheric pressure for at least 2 h. The re- ing in different conditions, can be found in Table Naphthenes - 1.038 3. As it can be seen, by increasing the tempera- tive. mainder was the deasphalted heavy oil (i.e., mal- Iso-paraffins - 35.766 ture, in various N2 flow rates, density, viscosity, tenes). a Asphaltene 31.56 19 refractive index and molecular weight of liquid (%wt, in C7) 31.7b

a in wet PFO, b in dry PFO product did not changed significantly while the optimum asphaltene removal rates from PFO-1 Table 1. Yield of liquid, gas and solid products (wt%) as a function of N2 flow rate (ml/min). and PFO-2 (i.e. 98% and 96%) were obtained at 390 °C and 380 °C and 100 ml/min of N2 flow N2 Flow 50 100 150 rate. PFO-1 70 73 72 Liquid Yield PFO-2 41 53 51 In addition, viscosity range of a typical diesel fuel PFO-1 24 22 22 Solid Yield is approximately between 1-3.8 mPa.s, which is PFO-2 45 35 35 similar to the liquid product of this study. Sulfur PFO-1 6 5 5 Gas Yield content of liquid products of PFO-1 and PFO-2 are PFO-2 14 12 15 about 1170 and 606 ppm, respectively. The vis-

cosity, density, average molecular weight and re- For GC-DHA analysis of the feedstock, each sam- fracting index of liquid product in the best crack- 3 Figure 3. Dependence of gas, liquid, and solid products ple was mixed with acetone for 24 h. After filtra- ing conditions, are 1.79-2.82 cSt, 1 g/cm , 201- yield on reactor temperature. tion, dissolved portions of all samples were used 205, and 1.6, respectively. for analysis. The composition of PFO fraction was very complex and it was a mixture of paraffins, 55 naphthenes, aromatics, and olefins. Table 2 shows DHA analysis the most common components along with the 50 physicochemical properties in the two proposed Figs. 6 and 7 illustrate the results of GC-DHA feedstocks. According to Table 2, sulfur content of analysis for the liquid products obtained from 45 PFO-1 was more than PFO-2 and aromatics con- thermal cracking in the presence of N2 and steam,

Liquid Product(wt%) Liquid as two types of carrier gases. Figs. 6 and 7 are 40 tent of PFO-1 was much more than PFO-2. How- ever, PFO-2 had significant water content where- related to PFO-1 and PFO-2, respectively. After cracking, aromatic amounts of liquid products, 35 as the other feedstock did not have any water content. Due to this reason, for n-heptane insolu- obtained from PFO-1, decreased to about 67% wt. 30 ble fraction content of PFO-2 two values, wet and In the presence of steam, the production of aro- matic, olefin, paraffin and iso-paraffin increased 370 375 380 385 390 dry, were reported. The n-heptane insoluble frac- Figure 5. Distribution of hydrocarbon compositions of Temperature(oC) tion can be considered the same as the asphaltene feedstock obtained from DHA analysis: (a) PFO-1 and (b) while the amount of unknown hydrocarbons de- content. Hence, it is possible to use asphaltene PFO-2. creased, as shown in Fig. 6. According to Fig. 7, In out after thermal cracking of PFO-2 in both states, content as a comparison index of the two feed- stocks. This fraction contains asphaltene and aromatics content increased and, at the same Figure 4. Changes in the liquid product yields of PFO at Detailed hydrocarbons information of the two time, the rate of iso-paraffins decreased signifi- different temperatures by changing the nozzle position. compounds that would be capable of being con- verted to asphaltene in the future. It was ob- feedstocks, obtained from GC-DHA analysis, are cantly. Also, in the presence of steam, the produc- served that PFO-2 had a high volume of olefin and illustrated in Fig. 5. As it is shown in Fig. 5(a) tion of aromatic and unknown hydrocarbons in- most of the organic compounds in PFO-1 are C creased while the amount of olefin, paraffin and, The PFO–n-pentane or n-heptane mixture was iso-paraffin compounds. It should be noted that, 10 in case of PFO-1, the quantity of olefines was neg- and C11 aromatics with high content of naphtha- iso-paraffin decreased. agitated using a magnetic stirrer (SP46925, Barn- lene and methylnaphtalene, whereas Fig. 5(b) ligible and there was no iso-paraffins. Although An important point to keep in mind is that the stead/ Thermolyne Corporation, USA) for 12 h indicates that most hydrocarbons in PFO-2 are the compositions of these feedstocks are so dif- represented GC-DHA data for the feedstocks were C10 hydrocarbons composed of aromatics, ole- ferent, their dry asphaltene contents are almost related to their portion which dissolved in ace- (Whatman, England). The filter cake, mainly fins, paraffins and iso-paraffins and larger and then filtered with 2 μm pore size filter papers the same. tone and injected to the GC analyzer. On the other composed of precipitated asphaltenes, was rinsed amounts of iso-paraffins are available in C and C 8 9 hand, the obtained liquid products were injected with a liquid precipitant. The rinsing process con- groups. tinued until the precipitant, which comes out of directly to the instrument and their results about total compounds which were produced after Mass (Fig. 8) these compounds can be naphtha- E. S. Moosavi et al. / Journal of ChemicalMass and(Fig. Petroleum 8) these Engineering, compounds 52 (1), can June be 2018 naphth / 23-33a- 30totalcracking. compounds Thus, the proposedwhich were heavy produced residues wereafter lene, methylnaphthalene, ethyl naphthalene, and thermallycracking. Thus, cracked the successfully. proposed heavy DHA residues analysis were data azulene.lene, methylnaphthalene, . The amount of ethyl Naphthalene naphthalene, is 31.33 and azulene. . The amount of Naphthalene is 31.33 ofthermally proposed cracked feedstocks successfully. and their DHA liquid analysis products data, wt% and 29.93 wt% of total aromatic with N and total compounds which were produced after Mass (Fig. 8) these compounds can be naphth2 a- whichof proposed obtained feedstocks in both andstates, their are liquid listed products in Table, steamwt% and as 29.93 carrier wt% gas, of totalrespectively. aromatic withGiven N 2 thatand cracking. Thus, the proposed heavy residues were lene, methylnaphthalene, ethyl naphthalene, and 4.which obtained in both states, are listed in Table naphthalenesteam as carrier is colorless, gas, respectively. azulene is aGiven dark bluethat thermally cracked successfully. DHA analysis data azulene. . The amount of Naphthalene is 31.33 4. isomernaphthalene of naphthalene is colorless, and azulene ethyl naphthalene is a dark blue has of proposed feedstocks and their liquid products, wt% and 29.93 wt% of total aromatic with N2 and lightisomer brown of naphthalene color; thus, and the ethyl separated naphthalene naphth hasa- which obtained in both states, are listed in Table steam as carrier gas, respectively. Given that lenelight fractionbrown color;was light thus, green. the separated naphtha- 4.3.2.3. GC-Mass analysis of Naphthalene lenenaphthalene fraction wasis colorless, light green. azulene is a dark blue 3.2.3.Table GC -3.Mass Cracked analysis oil products of Naphthalenecharacterization results Naphthalene is the most abundant single constit- isomer of naphthalene and ethyl naphthalene has Naphthalene is the most abundant single constit- uentExperiment of coal condition tar, a volatile product from destruc- light brown color; thus, the separated naphtha- uent of coal tar, a volatileµ productRI from destruc- 3.2.3.tive distillation GC-Mass analysisof coal, andofo Naphthalene it is alsoo formedρ by lene fractionwt% was lightPFO-2 green. (Wt %) o Gas flow @167 F @100 F wt% Ttive C distillation Gasof coal, and it is also g/cm3formed AR% by 60 PFO-2 (Wt %) modern(ml/min) processes for(cSt) the high(nd) -temperature 60 modernNaphthalene processes is the most for abundantthe high single-temperature constit- Cracked oil-2 (Wt % in N2) cracking (breaking up of large molecules) of pe- 50 Cracked oil-2 (Wt % in N2) uent of coal tar, a volatilePFO-1 results product from destruc- 50 crackingtroleum. It(breaking is commercially up of large produced molecules) by crystall of pe-i- tive distillation of coal, and it is also formed by 40 wt% PFO-2Cracked (Wt oil-2 %) (Wt % in Steam) 370troleum 100. It is commerciallyN2 1.01 produced1.59 by crystall1 92i- Cracked oil-2 (Wt % in Steam) zation from the intermediate fraction of con- 4060 wt% PFO-2 (Wt %) 380modern 100processes N2 for1.05 the high1.59 -temperature1 92 densedzation fromcoal thetar intermediateand the heavier fraction fraction of co n-of 3060 390 100 N2 1.03 1.59 1 98 CrackedCracked oil-2 oil-2 (Wt (Wt % % in inN2) N2) densedcracking coal(breaking tar and up ofthe large heavier molecules) fraction of pofe- 305050 395cracked 100petroleum. N2 It 1.05was found1.59 out 1that the95 crackedtroleum . petroleum.It is commercially It was produced found out by crystallthat thei- 2040 CrackedCracked oil-2 oil-2 (Wt (Wt % % in inSteam) Steam) 400high - boiling100 residueN2 of1.07 oil contains1.59 naphthalene1 94 2040 highzation- boiling from residuethe intermediate of oil contains fraction naphthalene of con- 30 410in admixture 100 withN2 small1.07 quantit 1.59 1- methyl88 10 indensed admixture coal withtar andsmall the quantit heavier fraction- methyl of 103020 390 150 - Nmethyl2 1.09 naphthalene 1.59 and1 1,6- 95di 390cracked 50petroleum. N2 It 1.02was found1.58 out 1that the93 0 methyl naphthalene- methyl [23]. naphthalene ies of and α 1,6- di 20100 390high - boiling100 residueSteam of1.04 oil contains1.59 naphthalene1 90 naphthalene,methyl naphthalene β [23]. ies of α 0 innaphthalene, admixture βwith PFOsmall-2 resultsquantit - methyl 10 380 100 - Nmethyl2 1.13 naphthalene 1.60 and1 1,6- 91di 380 100 N2 1.19 1.60 1 95 0 methyl naphthalene [23]. ies of α 380 wt%100 N2 1.26 1.60 1 93 100 wt% PFO-1 (Wt %) Figure 7. Distribution of hydrocarbons by Group Type (DHA analysis) in PFO-2 and the liquid 380naphthalene, 100 β N PFO-11.22 (Wt %)1.60 1 92 100 2 products obtained from thermal cracking in the presence of N and steam. T=Temperature, µ=Viscosity, RI= Refractive Index, ρ =Density, AR=Asphaltene RemovalFigure 7. Distribution of hydrocarbons2 by Group Type 80 Cracked oil-1 (Wt % in N2) (DHAFigure analysis) 7. Distribution in PFO- 2of and hydrocarbons the liquid products by Group obtained Type 80 Cracked oil-1 (Wt % in N2) (DHA analysis) in PFO-2 and the liquid products obtained wt% from thermal cracking in the presence of N2 and steam. obtained liquid products were injected coal, and it2 is also formed by modern 10060 PFO-1 (Wt %) from thermal cracking in the presence of N and steam. 60 directly to the instrument and their results processes for the high-temperature wt% about total Figurecompounds 7. Distribution which were of hydrocarbonscracking (breaking by Group up Typeof large molecules) CrackedPFO-1 (Wt oil-1 %) (Wt % in N2) 1008040 produced after(DHA cracking. analysis) Thus,in PFO -the2 and theof liquid petroleum. products It is obtainedcommercially produced 40 Cracked oil-1 (Wt % in N2) proposed heavyfrom residues thermal were cracking thermal inly the presenceby crystallization of N2 and steam. from the intermediate 80 6020 Cracked oil-1 (Wt % in Steam)cracked successfully. DHA analysis data fraction of condensed coal tar and the 20 of proposed feedstocks and their liquid heavier fraction of cracked petroleum. It 60 400 products, which obtained in both states, was found out that the high- boiling 400 are listed in Table 4. residue of oil contains naphthalene in 20 admixture with small quantities of α- 20 3.2.3. GC-Mass analysis of Naphthalene methyl naphthalene, β- methyl Naphthalene is the most abundant naphthalene and 1,6- di methyl 0 single constituent of coal tar, a volatile naphthalene [23]. 0 product from destructive distillation of Figure 6. Distribution of hydrocarbons by Group Type (DHAFigure analysis) 6. Distribution in PFO- 1of and hydrocarbons the liquid products by Group obtained Type Table 4. DHA analysis of the cracked oils (wt %). (DHA analysis) in PFO-1 and the liquid products obtained from thermal cracking in the presence of N2 and steam. Type PFO-1 PFO-2 1 2 3 4 from thermal cracking in the presence of N2 and steam.

Figure 6. Distribution Figure of hydrocarbons6. Distribution by ofGroup hydrocarbons Type (DHA by analysis) Group inType PFO -1 and Aromaticsthe liquid 86.5 27.9 66 68.4 45.6 54.5 products obtained from(DHAIn order thermal analysis) to cracking evaluate in PFO in -the1 and compoundspresence the liquid of N2 productsandin thesteam. fraction, obtained C14+ 6.5 0.59 11 9.6 3.5 2.5 fromwhichIn order thermal was to crackingevaluateseparated in thecompounds presenceas naphthalene of in N 2the and fraction, steam.during which was separated as naphthalene during Figure 8. GC-Mass analysis results of naphthalene thermal cracking of PFO-1 at 215-225 ºC, GC- FigureOlefines 8. GC -Mass0.34 analysis 20.4 results 3of naphthalene3.6 21 14.2 thermalMass analysis cracking was ofused. PFO Fig.-1 at8 illustrates215-225 ºC,the GrCe-- In order to evaluate compounds in the fraction, Paraffins 1.8 8.9 0.5 1.3 5.4 3.7 sultsMass ofanalysis GC-Mass. was According used. Fig. to 8 theillustrates results theof GCre-- which was separated as naphthalene during sults of GC-Mass. According to the results of GC- Figure Unknowns 8. GC -Mass4.9 analysis 5.5 results 16 of naphthalene13.7 11.4 14.7 thermal cracking of PFO-1 at 215-225 ºC, GC- Naphthenes - 1 2 0.6 1.3 0.9 Mass analysis was used. Fig. 8 illustrates the re- Iso -Paraffins - 35.8 2 2.7 11.8 9.4 sults of GC-Mass. According to the results of GC-

1: oil-1in N2, 2: oil-1in Steam, 3: oil -2 in N2, 4: oil -2 in Steam

E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33 31

Table 3. Cracked oil products characterization results Experiment condition

Gas flow µ @167oF (cSt) RI @100oF (nd) ρ g/cm3 AR% T oC Gas (ml/min)

PFO-1 results

370 100 N2 1.01 1.59 1 92 380 100 N2 1.05 1.59 1 92 390 100 N2 1.03 1.59 1 98 395 100 N2 1.05 1.59 1 95 400 100 N2 1.07 1.59 1 94 410 100 N2 1.07 1.59 1 88 390 150 N2 1.09 1.59 1 95 390 50 N2 1.02 1.58 1 93 390 100 Steam 1.04 1.59 1 90 PFO-2 results 380 100 N2 1.13 1.60 1 91 380 100 N2 1.19 1.60 1 95 380 100 N2 1.26 1.60 1 93 380 100 N2 1.22 1.60 1 92 T=Temperature, µ=Viscosity, RI= Refractive Index, ρ =Density, AR=Asphaltene Removal

Table 4. DHA analysis of the cracked oils (wt %).

Type PFO-1 PFO-2 1 2 3 4 Aromatics 86.5 27.9 66 68.4 45.6 54.5 C14+ 6.5 0.59 11 9.6 3.5 2.5 Olefines 0.34 20.4 3 3.6 21 14.2 Paraffins 1.8 8.9 0.5 1.3 5.4 3.7 Unknowns 4.9 5.5 16 13.7 11.4 14.7 Naphthenes - 1 2 0.6 1.3 0.9 Iso-Paraffins - 35.8 2 2.7 11.8 9.4 1: oil-1in N2, 2: oil-1in Steam, 3: oil -2 in N2, 4: oil -2 in Steam

3.2.4. Outlet gases characterization tenes strongly depend on the specific precipitant process as well as the employed asphaltene sepa- Drawing a comparison between the outlet gases, ration conditions. As mentioned above, the n- obtained from thermal cracking in the presence heptane insoluble fraction content of PFO-1 and of N and steam, as two different carrier gases, a 2 PFO-2 were 31.56 % and 31.7%, respectively. small amount of CO2 was observed. It is also ob- After cracking process, the measured asphaltene served that, when steam was being used, me- content of the water-free liquid product, obtained thane, ethylene and hydrogen production showed using thermal cracking under optimum condi- an increasing trend (Fig. 9). As a result, it can be tions, was 0.68 up to 2 wt%. For each sample, the concluded that thermal cracking with steam is an repeatability of the measured asphaltene content efficient upgrading technique. was within ±0.5% of the reported yield data. The optimum asphaltene removal rates of PFO-1 and PFO-2 were 98% and 96%, at 390 °C and 380 °C, 3.2.5. Precipitation yield measurements for respectively. water-free liquid product

It is found that the yields and properties of the precipitated asphaltenes and the remaining mal- 32 E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33

Acknowledgments 40 35 PFO-2 The authors wish to thank Petrochemical Re- 30 search & Technology Co. of National Petrochemi- PFO-1 cal Co. (NPC-RT) for financially supporting this

25 20 research. wt % 15 10 5 References 0 CH4 C2H4 C3H6 CO2 H2 [1] Andersen, S. I. and Speight, J. G. (2001). "Pe- Gas product troleum resins: separation, character, and role in petroleum." Petroleum science and technolo- Figure 9. Comparison of weight rates of outlet gases (H2, gy, Vol. 19, No. 1, pp. 1-34. CO2, and C1-C3) for different samples in the presence of steam [2] Fan, T. and Buckley, J. S. (2002). "Rapid and accurate SARA analysis of medium gravity crude oils." Energy & Fuels, Vol. 16. No. 6, pp. 1571- 4. Conclusion 1575. This work investigated the effect of different con- [3] Kord, S. and Ayatollahi, S. (2012). "Asphaltene ditions of PFO thermal cracking of two PFO sam- precipitation in live crude oil during natural ples on liquid, solid and gas product yields and depletion: experimental investigation and asphaltene removal efficiency in a newly de- modeling." Fluid Phase Equilibria, Vol. 336, pp. signed experimental setup and, at the same time, 63–70. produced naphthalenes as valuable compounds [4] Leontaritis, K.J. and Mansoori, G.A. (1988). from one of the PFO samples. "Asphaltene deposition: a survey of field The ability to use water as a cheap carrier gas and experiences and research approaches." Journal of high asphaltene extraction efficiency without the Petroleum Science Engineering, Vol. 1, No. 3, pp. use of solvents, no need for catalyst, and simple 229–239. operating system and experimental conditions [5] Mohammadi,A.H. and Richon, D. (2007). "A are are outstanding benefits of this method for monodisperse thermodynamic model for upgrading PFO. estimating asphaltene precipitation." AIChE The results showed that liquid yield increases Journal, Vol. 53, No. 11, pp. 2940–2947. with increasing temperature while the solid [6] Mullins, O., Sheu, E.Y. (1998). Structures and product decreased Dynamics of Asphaltenes, New York, Springer. The obtained results showed That the best ther- [7] Speight, J.G. (1991). The Chemistry and mal cracking condition of PFOs were 390 °C and Technology of Petroleum, New York, CRC Press. 380 °C for the reactor temperature of PFO-1 and PFO-2, respectively; 150 °C for the carrier gas [8] Adebiyi, F.M. and Thoss ,V. (2014). "Organic temperature and 100 ml/min for the carrier gas and elemental elucidation of asphaltene fraction flow rate. In these circumstances, about 70 and of Nigerian crude oils." Fuel, Vol. 118, pp. 426– 53 wt% of the liquid product, 25 wt% of the solid 431. products, and 5 wt% of the gas product are gen- [9] Yong-jun, L. (2012). "Microstructure erated while the asphaltene separation reached about 95 and 96.5%. In addition, viscosity and characterization of asphaltenes from atmospheric molecular weight of the liquid product vary be- residue before and after hydroprocessing." tween 1.79 to 2.82 cSt and 201 to 205 g/mol, re- Journal of Fuel Cheistry and Technology, Vol. 40, spectively. Density and refractive index of liquid product were 1 and 1.6, respectively. No.[10] 9, Lababidi, pp.1086−1091. H.M., Sabti, H. M. and AlHumaidan, F. S. (2014). "Changes in asphaltenes during E. S. Moosavi et al. / Journal of Chemical and Petroleum Engineering, 52 (1), June 2018 / 23-33 33 thermal cracking of residual oils." Fuel, Vol. 117, cess technologies for upgrading of heavy oils and pp. 59–67. residuaet." Fuel, Vol. 86, pp.1216-1231. [11] Duyck, C., Miekeley, Silveira, C. L. P. and [18] Takatusuka,T., Watari, R. and Hayakawa, H. Szatmari, P. (2002). "Trace element (1996). "Renewed attention to the EUREKA pro- determination in crude oil and its fractions by cess: thermal cracking process and related tech- inductively coupled plasma mass spectrometry nologies for residual oil upgrading." Studies in using ultrasonic nebulization of toluene solu- Surface Science and Catalysis, Vol. 100, pp. 293– tions." Spectrochim Acta, Part B, Vol. 57, No. 12, 301. pp. 1979-1990. [19] Shishavan, R., Ghashghaee, M. and Karimza- [12] Park, S.J. and Mansoori, G. A. (1988). deh, R. (2011). "Investigation of kinetics and cra- "Aggregation and Deposition of Heavy Organics in cked oil structural changes in thermal cracking of Petroleum Crudes." Energy Sources, Vol. 10, No. 2, Iranian vacuum residues." Fuel Processing Tech- pp.109-125. nology, Vol. 92, No. 12, pp. 2226–2234. [13] Taylor, S.E. (1992). "Use of Surface-Tension [20] Fesharaki, M., Ghashghaee, M. and Karimza- Measurements to Evaluate Aggregation of deh, R. (2013). "Comparison of four nanoporous Asphaltenes in Organic-Solvents." Fuel, Vol. 71, catalysts in thermocatalytic upgrading of vacuum pp. 1338-1339. residue." Journal of Analytical and Applied Pyroly- sis, Vol. 102, pp. 97-102. [14] Thawer, R., Nicoll, D. C. A. and Dick, G. (1990). "Asphaltene Deposition in Production [21] Sawarkar, A. N., Pandit, A. B., Samant, S. D. Facilities.”SPE Producation Engineering." Vol. 5, and Joshi, J. B. (2007). "Petroleum Residue No. 4, pp. 475-480. Upgrading Via Delayed Coking: A Review." The Canadian Journal of Chemical Engineering, Vol. 85, [15] AlHumaidan, F., Lababidi, H. M. S. and Al- No.1, pp. 1–24. Rabiah, H. (2013). "Thermal cracking kinetics of Kuwaiti vacuum residues in Eureka process." [22] ASTM D3279, Standard Test Method for n- Fuel, Vol. 103, pp. 923–931. Heptane Insolubles, Book of Standards Volume: 04.03, West Conshohocken,2001. [16] Silva, F. B., Guimarães, M. J. O. C. Seidi, P. R. and Garica, M. E. F. (2013). "Extraction And [23] Kashirski, V.G. and Petelina, V.S. (2001). Characterization (Compositional And Thermal) Of "Thermal decomposition of the Kenderlyk Oil Asphaltenes From Brazilian Vacuum Residues." shale under conditions of high-speed heating." Oil Brazilian Journal of Petroleum and Gas, Vol. 7, No. Shale, Vol. 18, No. 1, pp. 85-91. 3, pp. 107-118.

[17] Rana, M.S. and Sámano, V., Ancheyta, J., Diaz,

J.A.I. (2007). "A review of recent advances on pro-