Catalytic Cracking of Non-Hydrotreated, Hydrotreated and Sulfuric Acid-Treated Vacuum Gas Oils

Abstract

The quality of the catalytic cracking feed can affect the conversion by 35%, while the activity of the catalyst can influence the conversion by 11% and the reaction temperature by 15%. The pivotal role of feed quality justifies the investigations directed to better understanding which components of the feed impinge the conversion, yields, selectivity and properties of the catalytic cracking products. In this research, two virgin vacuum gas oils, a hydrotreated vacuum gas oil and five sulfuric acid-treated vacuum gas oils were cracked on a commercial equilibrium catalyst (Nova DAO) in a micro-activity (MAT) unit at different catalyst-to-oil ratios to obtain the conversion, yields, and selectivities at the point of maximum gasoline yield. The treatment of one of the virgin and the hydrotreated vacuum gas oils with sulfuric acid decreased the heavy aromatics from 22.6 to 0.0 wt.% and resins from 2.7 to 0.0 wt.%. Intercriteria analysis of the experimental cracking data revealed that the reduction and removal of the heavy aromatic compounds from the vacuum gas oil had a profound effect on conversion, yields, and gasoline quality. It led to conversion enhancement from 70.8 to 86.1 wt.% and a reduction of gasoline research octane number by two points. The conversion at the maximum gasoline yield was confirmed to be very well predicted by a correlation that includes the empirical parameters aromatic carbon and hydrogen contents with %AAD of 0.7 wt.% and maximum absolute deviation of 2.3 wt.%.

1. Introduction

Fluid catalytic cracking was introduced in 1942 [1] in response to the need for high-octane gasoline, used at the time as a fuel for aircraft. Since then, it has evolved as the main producer of high-octane gasoline [2], middle distillate producer [3,4], ethylene and propylene producer [5,6,7], bio-fuel producer [8,9,10], and waste (scrap tire, plastics) processor [11,12,13]. Its resilience in processing heavy oils with diverse quality has allowed a high penetration of FCC technology in petroleum refining [14,15,16]. Worldwide, approximately more than 600 FCC units out of 825 refineries are operated in petrochemical complexes, converting vacuum gas oil (VGO) and high-boiling residues into lighter fuel products and chemical feedstocks. Typically, FCC unit feeds all over the world are mixtures of atmospheric and vacuum gas oils as the major components, atmospheric residues, coker and visbreaking gas oils, hydrocracking residues, hydrotreated gas oils and residues, furfural extracts, demetalised oil (DMO), etc. [16]. According to Global News Wire (2024), it is expected that the global refinery fluid catalytic cracker units (FCCU) capacity will increase from 14,153 mbd in 2023 to 16,870 mbd in 2028 at an average annual growth rate (AAGR) of 3.5% [17]. It has been observed over the years that the single variable that has the greatest effect on the performance of the FCC process is the feed quality [18,19,20]. Navarro et al. [16] and Fisher [18] reported that the conversion of any FCC feed at the point of maximum gasoline yield (called maximal conversion) can vary between 50 and 85 wt.%. The FCC feed conversion from an equilibrium FCC catalyst obtained at the same operating conditions with a reference feed can vary between 64 and 75 wt.% as communicated in [21,22]. Considering that the increase of reaction temperature by 10 °C leads to a conversion raising of about 3 wt.% [23], one may expect that for 50 °C reaction temperature augmentation (for example, from 500 to 550 °C,) the conversion enhancement would equal to 15 wt.%. All of this is too far from the 35 wt.% difference in conversion; that is only due to the distinction in feed quality for FCC. This can explain why so many studies have been dedicated to exploring the effect of feedstock quality on FCC process performance [16,18,19,20,21,22,23,24,25,26,27,28,29].

Navaro et al. [16] have investigated more than one hundred FCC feed samples, originating from various geographic locations and consisting of selected FCC refinery streams. All of these feeds were characterized for density, simulated distillation characteristics, refractive index, Conradson carbon, total and basic nitrogen, sulfur, metal, hydrogen, saturates, aromatics, resins, and asphaltenes (SARA) contents. Other specialized techniques, such as UV-vis and mass spectrometry for the selected sample, were also employed to investigate the hydrocarbon types and the chemical composition. All feeds were cracked in the advanced cracking equipment (ACE) [30] pilot plant using two distinct commercial FCC catalysts. They found that hydrogen content is the best descriptor for predicting maximal conversion. For the range of hydrogen contents in the FCC feeds from 9.6 to 14.8 wt.%, the maximal conversion fluctuation is between 50 and 85 wt.% [16].

Fisher [18] analyzed nine very different FCC feeds for their density, refractive index, aniline point, kinematic viscosity at 40 and 100 °C, total and basic nitrogen contents, sulfur, and Conradson carbon contents. He employed mass spectrometry to quantify the hydrocarbon types. Then he cracked the FCC feeds on a moderate octane catalyst Davison, Nova D (Davison standard activity of 70 vol.%) in a laboratory micro-activity unit (MAT, ASTM D 3907 [31]) at 490 °C, at different catalyst-to-oil ratios (CTOs) ranging between 1.8 and 9 wt./wt. (100 s residence time), to obtain the yield structure at the maximal conversion. He ascertained that the maximal conversion could be predicted from the mass spectrometry data for the content of saturated plus mono-nuclear aromatic hydrocarbons, so-called gasoline precursors. He found the yields of products at this conversion point to correlate with the maximal conversion for the feeds with gasoline precursor levels between 42.4 and 85.0 wt.%, Fisher observed that the maximal conversion varied between 50 and 85 wt.%.

Harding et al. [19] separated the sour imported heavy gas oil (SIHGO) into four boiling fractions <371 °C, 371–427 °C, 427–482 °C, and >482 °C, and three fractions by chemical type using ASTM D 2007 [32] (saturates, aromatics and polar) and enriched the original SIHGO with these seven fractions in a different ratio. In this way, they obtained ten different FCC feeds, which were cracked on a low matrix rare earth ultrastable Y catalyst (REUSY) in a laboratory MAT (ASTM D 3907 [31]) unit at 525 °C, 30 s residence time, and different catalyst-to-oil ratios. Employing an approximate linear combination technique, they determined that the conversion and yields were only slightly affected by changes in boiling range, while they were strongly influenced by the chemical type. At a constant CTO of 5.0 wt./wt. the saturates exhibited conversion of 93.8 wt.%, the aromatics 53.6 wt.%, and the polars 8.8 wt.%.

Ng et al. [20] proposed a feed grading method with consideration of only concentrations of gasoline precursors, measured by mass spectrometry, total nitrogen, and microcarbon residue (MCR). They cracked ten different feeds on an equilibrium Dimension 60 catalyst (MAT activity of 72 wt.%) in a laboratory MAT unit (ASTM D 3907 [31]) at 510 °C, catalyst contact time of 30 s, and CTO variation between 3.8 and 8.3 wt./wt. For the feeds with gasoline precursor levels between 47.7 and 97.0 wt.% Ng et al. [24] reported maximal conversion variation in the range 63.3–84.7 wt.%. They found a very strong correlation between the FCC feed gasoline precursor level and the maximal gasoline yield (R² = 0.968), but not such a strong correlation, they determined between the gasoline precursor level and the maximal conversion [24], in contrast to the research of Fisher [18].

Xu et al. [25] investigated the catalytic cracking of 24 deasphalted oils (DAOs) that was performed in a laboratory in a confined fluid bed reactor on commercial catalyst RHZ-200 at a reaction temperature between 480 and 560 °C, CTO between 4 and 10 wt./wt., and residence time between 1.4 and 5 s. They characterized the 24 FCC feeds by using liquid chromatography, so-called saturates, aromatics, and resins (SAR) analysis. The researchers established that the yield of gasoline was promoted by the contents of saturate and aromatic compounds (the effect of saturates being twice that of aromatics) and depressed by the resins. The yield of diesel was boosted mainly by aromatics, to a much lesser extent by saturates and was depressed by resins. The main contributors to coke formation during the DAO catalytic cracking were the resins, followed by aromatics (5 times less effective than resins), and the least coke-making saturates (10 times less effective than resins).

Bollas et al. [26] investigated the catalytic cracking of ten feeds, which they characterized by the n-d-M method (ASTM D3238 [33]), whereof they determined the aromatic carbon and the average carbon number, calculated from the molecular weight of the FCC feed. These feed parameters, along with the total nitrogen and sulfur contents, were suitably incorporated to forecast the influence of feedstock quality on the conversion and coke yield of the fluid catalytic cracking process. They tested the proposed method with thirteen different FCC feedstocks and found an excellent accuracy of prediction of conversion (R² = 0.993) and a lower accuracy of prediction of coking tendency (R² = 0.912).

Lappas et al. [27] studied the catalytic cracking of ten feeds with great variability in their properties (specific gravity variation between 0.893 and 1.003) with the aim of determining the effect of feedstock quality on gasoline composition. They cracked the feeds on a commercial equilibrium catalyst (MAT activity of 70 wt.%) in a circulating riser pilot plant at a reaction temperature in the range of 537–560 °C, with CTO variation between 4 and 16 wt./wt. and partial pressure of hydrocarbons 68.9–82.7 kPa in order to investigate the effect of feedstock physical properties on the gasoline yield and composition, the feed conversion, and the coke yield. They ascertained that feed conversion, coke yield, gasoline yield, and composition were strongly affected by the type of FCC feedstock. The FCC conversion was established to have a strong function of three feed properties: mean average boiling point, specific gravity, and basic nitrogen content. They observed that from a paraffinic feed, olefinic gasoline is produced, while from an aromatic FCC feedstock, aromatic gasoline is obtained. The gasoline aromatic content was found to depend on the mean average boiling point, specific gravity, refractive index, basic Nitrogen, and Sulfur contents. Meanwhile, the gasoline olefin content was determined to be affected by the mean average boiling point, specific gravity, K_W-characterizing factor, and basic nitrogen and sulfur content.

Lappas et al. [28] performed 145 tests with various feeds whose property variation was as follows: 0.8097 ≥ specific gravity ≤ 1.0079; 1.3266 ≥ Refractive index (70 °C) ≤ 1.5313; 10.1 ≥ Hydrogen by NMR (wt%) ≤ 15.3; 0.0 ≥ Micro carbon residue (wt%) ≤ 18.9; 0.0003 ≥ Sulfur (wt%) ≤ 3.35; 0.001≥ Total Nitrogen (wt%) ≤ 0.356. They cracked these feedstocks in a circulating riser pilot plant and in an ACE [30] unit. Despite a good agreement for comparative ranking of feedstock quality and catalyst activity observed, there were significant differences in the absolute yields obtained at constant conversion. The results explained by the authors are differences in reactor operating parameters and their effect on catalyst performance and product selectivity.

In earlier research [29], the catalytic cracking of ten gas oils and one deasphalted hydrocracked vacuum residue obtained from ebullated-bed vacuum residue hydrocracking in a laboratory ACE [30] unit was examined. The cracking experiments were performed on a commercial equilibrium catalyst (micro-activity of 73 wt.%) at a reaction temperature of 527 °C; catalyst time-on-stream 30 s; catalyst-to-oil ratio variation between 3.5 and 7.5 wt./wt. The studied heavy oils were characterized by empirical methods, K_W-characterization factor [34], n-d-M method (ASTM D3238 [33]), aromatic ring index (ARI) [35,36], and the Conoco Philips correlations of the aromatic carbon and hydrogen contents [37]. Stratiev et al. [29] found that the conversion at the maximum catalyst-to-oil ratio of 7.5 wt./wt., which for the examined feeds varied between 45.6 and 70.5 wt.%, was predicted by the K_W-characterization factor with a squared correlation coefficient R² of 0.99. The product yields were determined to depend on the feed carbon atoms, aromatic ring index, paraffinic and aromatic carbon content calculated by the n-d-M method, and the total aromatics content measured by liquid chromatography.

All of the abovementioned investigations tried to relate the FCC feed quality to conversion, yields, and gasoline characteristics. The distinct researchers used diverse FCC feed characterizing methods, from mass spectroscopy to SAR analysis and different empirical correlations. Neither has attempted to relate the conversion, yields, and gasoline characteristics with data from group hydrocarbon composition of the FCC feeds, nor has a summary been provided of which empirical method for characterization and which physicochemical properties are the best descriptors of FCC conversion product selectivities and gasoline quality. For that reason, eight vacuum gas oils: SIHGO, non-hydrotreated vacuum gas oil (VGO) derived from Russian Export Blend crude oil (REB), hydrotreated REB VGO (HTREB), and five vacuum gas oils obtained from SIHGO, and HTREB after sulfuric acid treatment were characterized by diverse empirical methods and by the group hydrocarbon composition as saturates, light aromatics, middle aromatics, heavy aromatics and resins and cracked in a laboratory MAT unit (ASTM D 3907) [31]. The treatment with sulfuric acid mainly decreased the content of heavy aromatics from 22.6 in the SIHGO to 0.0 wt.% at the expense of increasing the saturated content from 46.6 to 73.8 wt.%. In addition, the resin content in the SIHGO was reduced from 2.7 to 0.0 wt.% as a result of the sulfuric acid treatment. In this way, the effect of the heavy aromatics and resins on the catalytic cracking of vacuum gas oils was investigated. Besides, the meaning of empirical parameters such as aromatic carbon (C_A) and hydrogen (H) contents, which were found to be excellent descriptors of the feed behavior in the FCC process [38,39], was verified in this research.

The aim of this study is to investigate the effect of heavy aromatics, resins, and the empirical feed characterizing parameters such as K_W, ARI, C_A, and H on FCC conversion at the point of maximum gasoline yield and on the reactivity of the FCC feed.

2. Materials and Methods

Two virgin vacuum gas oils, SIHGO, also investigated in [19,40], REB VGO, and the one obtained by hydrotreatment of the third VGO employed in this work, HT, REB, were cracked in a laboratory MAT unit on a commercial FCC catalyst (Nova DAO), whose properties are summarized in

Table 1. Physical and chemical properties of a commercial equilibrium catalyst (Nova DAO) used in this study.

Chemical Composition, wt.%
Al2O3	47.2
Na2O	0.17
Re2O3	1.34
Fe	0.68
V, ppm	1074
Ni, ppm	328
Physical properties
Catalyst fractional composition, wt.%
0–20 μ	<1
0–40 μ	6
0–80 μ	65
0–105 μ	87
0–149 μ	97
Average particle size, μ	71
Apparent bulk density, g/cm3	0.92
Pore volume, cm3/g	0.3
Total surface area, m2/g	66
Zeolite surface area, m2/g	37
Matrix surface area, m2/g	29
Zeolite unit cell size, nm	2.439
Micro-activity, wt.%	65

at 527 °C reaction temperature, catalyst contact time of 30 s, and catalyst-to-oil ratio variation between 1 and 6 wt./wt. The details of the experimental procedure and the modeling of selectivity data obtained from micro-activity testing of FCC catalysts and feeds are described in [41]. Conversion is defined as 100-LCO-HCO. Light cycle oil (LCO) and heavy cycle oil (HCO) are the yield fractions in the cracking products in wt.% of feed with cut points of 215 °C< LCO < 338 °C HCO.

The composition of gasoline, obtained during the cracking experiments, expressed as normal paraffins, iso-paraffins, olefins, naphthenes and arenes (PIANO-analysis) was analyzed using a gas chromatograph HP5880A (Hewlett-Packard, Palo Alto, CA, USA) equipped with a flame ionization detector (FID) and a pre-fractionation column (20* OV-101 on 80/100 Chromosorb W-HP), which allows only the hydrocarbons boiling in the gasoline range to be analyzed on the main column (0.21 mm inner diameter × 50 m, 0.5 m cross-linked methylsilicon).

The research and motor octane numbers of the gasoline product are calculated from the gasoline composition using an octane model that takes into account the nonlinear mixing characteristics of the hydrocarbons [42].

The vacuum gas oils SIHGO and HTREB were treated with sulfuric acid (95–97%, density of 1.84 g/cm³, from E. Merck, Darmstadt, Germany). The procedure of H₂SO₄ treatment was as follows: The VGOs are heated to 60 °C by mixing them with an inert gas (nitrogen or CO₂) fed through a glass tube. At this temperature, the acid is gradually added. After the acid has been added, the mixture is stirred for 30 min at a temperature of 50–60 °C. It is then allowed to stand for 12 h (practically until the acidic tar is completely separated from the fraction). The resulting fraction is decanted from the acidic tar and washed thoroughly with water until the aqueous extract reacts neutrally. The VGOs are then heated to 140 °C, and at this temperature, 10 wt. % of bleaching clay is added to remove the acidic residue. Stirring is again carried out with inert gas for 30 min. The mixture is then cooled to 90–105 °C and filtered through a Buechner funnel under vacuum (water pump).

Table 2. Preparation of VGOs treated with H₂SO_4.

VGOs Treated with H2SO4	Raw Material for Treatment with H2SO4	wt. % Added H2SO4	wt. % Removed Acid Tar
SIHGO (6% H2SO4)	SIHGO	6	22
SIHGO (16% H2SO4)	SIHGO (6% H2SO4)	10	22
HTREB (6% H2SO4)	HTREB	6	22
SIHGO (46% H2SO4)	SIHGO (16% H2SO4)	30	34
SIHGO (76% H2SO4)	SIHGO (46% H2SO4)	30	34

summarizes the preparation of H₂SO₄-treated VGOs

The group hydrocarbon composition of all studied in this work eight vacuum gas oils was determined by using liquid adsorption chromatography on silica gel following the procedure: The sample oil in amount of about 8 g is diluted in n-hexane in ratio 1:3 and charged to a glass percolation column containing 80–85 g silica gel (silica gel 60 Fluka (Darmstadt, Germany), particle size 35–70 mesh ASTM). After the whole sample quantity soaks the silica gel, 350–450 mL of n-hexane (99% Fluka) is charged to the column for desorption of saturates, 200 mL of benzene (98.5–99.9%) for desorption of aromatics, and 200 mL ethanol (99.7%)-benzene mixture 50 to 50 by volume for desorption of resins. Then, 100 mL ethanol is charged to the column for final washing. The effluent is collected in glass bottles (20–25 pieces). The solvents are completely removed from the recovered n-hexane, benzene, and ethanol-benzene fractions by distillation and residues are weighed. The content of different hydrocarbon groups in the sample is determined by refraction (n_d²⁰) and yields of the different fractions. Saturates are these fractions that have n_d²⁰ ≤ 1.49. Light aromatics are the fractions that have n_d²⁰ between 1.49 and 1.53. Middle aromatics are the fractions that have n_d²⁰ between 1.53 and 1.59. Heavy aromatics are the fractions that have n_d²⁰ ˃ 1.59. The resin fraction follows the heavy aromatics fraction, and its refraction cannot be determined because of its dark color. The total weight of all the recovered fractions must be equal to at least 97% of the sample charged. If this recovery is not obtained, the test is repeated.

The empirical feed characterizing parameters aromatic carbon and hydrogen contents were calculated by using the Total method as described in [43] and shown as Equations (1) and (2).

$$C_A=-814.136+635.192RI-129.266SG+0.013MW-0.340S-6.872\mathrm{l}\mathrm{n}\left(VIS\right)$$

(1)

where:

C_A = Aromatic carbon content, wt. %;

RI = Refractive index at 20 °C (n_d²⁰);

MW = Molecular weight, g/mol

VIS = Kinematic viscosity ate 98.9 °C

$$H=52.825-14.260RI-21.329SG-0.0024MW-0.052S+0.757\mathrm{l}\mathrm{n}\left(VIS\right)$$

(2)

The molecular weight (MW) of investigated vacuum gas oils was calculated by using the correlation of Goosens [44] (Equation (3)).

$$MW={\displaystyle \frac{0.010770{\times T}_{50\%}^{\left[1.52869+0.06486\times \mathit{ln}\left(\frac{T_{50\%}}{1078-T_{50\%}}\right)\right]}}{d}}$$

(3)

where:

T_50% = Boiling temperature of evaporate at 50 wt.% from the simulated distillation curve, °C;

d = Density of VGO at 20 °C, g/cm³.

The aromatic ring index (ARI) of VGOs is calculated by the correlation of Abutaqiya et al. [35,36], as shown in Equation (4).

$$\mathrm{A}\mathrm{R}\mathrm{I}=\mathrm{f}\left(\mathrm{M}\mathrm{W},{\mathrm{F}}_{\mathrm{R}\mathrm{I}}\right)={\displaystyle \frac{2[\frac{\mathrm{M}\mathrm{W}}{{\mathrm{F}}_{\mathrm{R}\mathrm{I}}}–(3.5149\mathrm{M}\mathrm{W}+73.1858\}}{(3.5074\mathrm{M}\mathrm{W}-91.972-(3.5149\mathrm{M}\mathrm{W}+73.1858)}}$$

(4)

The function of the refractive index can be estimated by Equation (5) [35,36].

$${\mathrm{F}}_{\mathrm{R}\mathrm{I}}={\displaystyle \frac{{RI}^2-1}{{RI}^2+2}}$$

(5)

K_W-characterizing factor of studied VGOs is calculated by using Equation (6).

$$K_w={\displaystyle \frac{\sqrt[3]{1.8\times (T_{50}\left(°C\right)+273.15)}}{specificgravity}}$$

(6)

Table 3. Physical and chemical properties of the eight vacuum gas oil feeds for FCC experiments.

		SIHGO	REB	HTREB	SIHGO (6% H2SO4)	SIHGO (16% H2SO4)	SIHGO (46% H2SO4)	SIHGO (76% H2SO4)	HTREB (6% H2SO4)
Density at 15 °C	g/cm3	0.9154	0.9183	0.8997	0.9009	0.8995	0.8945	0.8661	0.8779
Kinematic viscosity at 37.8 °C	mm2/s	27.70		49.40	27.10	26.30	49.10	22.10	23.40
Kinematic viscosity at 98.9 °C	mm2/s	4.10	7.66	5.70	4.00	4.00	5.60	4.00	3.90
Refractive index at 20 °C		1.51	1.51	1.50	1.51	1.50	1.50	1.48	1.49
Molecular eight	g/mol	315	361	356	319	322	328	339	368
ARI		1.57	1.65	1.47	1.46	1.36	1.33	0.88	1.04
KW-characterizing factor		11.68	11.83	12.03	11.86	11.89	11.97	12.36	12.34
Aromatic carbon content (CA)	wt.%	22.3	20.4	18.3	20.9	18.1	12.0	9.8	16.2
Hydrogen content (H)	wt.%	11.8	12.0	12.5	12.2	12.3	13.0	13.3	12.7
Sulfur	wt.%	2.60	1.70	0.44	2.20	1.90	0.44	0.80	1.20
Total Nitrogen	wt.%	0.09	0.12	0.09	0.02	0.01	0.01	0.01	0.01
Basic Nitrogen	wt.%	0.04	0.045	0.045	0.00	0.00	0.00	0.00	0.00
Group hydrocarbon compostion
Saturates	wt.%	46.60	47.30	60.00	49.60	53.10	65.30	73.80	59.40
Ligh Aromatics	wt.%	16.10	16.80	17.50	16.30	16.90	18.30	15.40	21.60
Middle aromatics	wt.%	11.70	9.50	8.80	12.30	11.90	14.50	9.80	11.70
Heavy aromatics	wt.%	22.60	24.20	12.70	18.30	15.40	0.00	0.00	6.10
Resins	wt.%	2.70	2.20	1.10	2.70	1.80	1.00	0.00	1.50
Simulated distillation ASTM D-2887 [45]	°C
IBP		214	30	292	219	227	241	226	307
10%		309	401	368	310	313	316	313	373
30%		363	419	406	363	366	368	367	410
50%		407	442	433	406	408	411	411	435
70%		451	463	462	450	451	453	454	463
90%		507	494	502	505	506	506	509	502
FBP		571	503	560	576	574	569	582	555

summarizes the physical and chemical characteristics of the VGO feeds to be cracked in this research.

The evaluation of relationships between the characteristics of the VGOs to be cracked and the maximal conversion, yields and product selectivities was performed by using inter-criteria analysis (ICrA) [46]. Details of the theory and application of ICrA are presented in [47]. The ICrA approach calculates two intuitionistic fuzzy functions, μ and υ, whose values define the degree of the relationship between the criteria. For μ = 0.75 ÷ 1.00 and υ = 0 ÷ 0.25, a region of statistically meaningful positive consonance is determined, while at μ = 0 ÷ 0.25 and υ = 0.75 ÷ 1.00, an area of statistically meaningful negative consonance is derived. All other cases are considered to be dissonant. A strong consonance is considered at values of μ = 0.95 ÷ 1.00, υ = 0.00 ÷ 0.05 (positive), μ = 0.00 ÷ 0.05, υ = 0.95 ÷ 1.00 (negative), while a weak consonance is deemed at μ = 0.75 ÷ 0.85, υ = 0.15 ÷ 0.25 (positive), μ = 0.15 ÷ 0.25, υ = 0.75 ÷ 0.85 (negative). Two software packages for ICrA were established and are freely available as open source from https://intercriteria.net/software/ and detailed in [48,49,50].

Before the ICrA evaluation, all variables are normalized using the normalization formula (Equation (7)).

$$X_{new}={\displaystyle \frac{X-X_{min}}{X_{max}-X_{min}}}$$

(7)

3. Results and Discussion

3.1. Effect of Sulfuric Acid Treatment on the Properties of Treated Vacuum Gas Oils

The data in Table 3 indicate that the sulfuric acid treatment predominantly removes the heavy aromatics from the VGO without affecting the boiling point distribution and molecular weight. Figure 1a exhibits the presence of a very strong linear relation between the content of heavy aromatic compounds and the quantity of H₂SO₄ used to treat the VGO (R² = 0.9932). Figure 1b also displays that the content of resins decreases when the quantity of H₂SO₄ used to treat SIHGO increases. For the resins, the relation with the H₂SO₄ treating rate is not as strong as that for the heavy aromatics (R² = 0.9717), but it can still be considered very strong. Actually, the treatment with 6% H₂SO₄ does not lead to a reduction of resin content, and in the case of HTREB, it even increases from 1.1 to 1.5 wt.%. From this data, one may conclude that the treatment of VGO with sulfuric acid mostly removes the heavy aromatic compounds. The employment of Equations (4) and (5) and the data for the refractive index of the three aromatic fractions, light, middle, and heavy, were used as a criterion to separate them in the process of liquid chromatography, along with the data for the molecular weight allowed to calculate their ARI. For the light aromatics, ARI was 1.5; for the middle, it was 2.5; and for the heavy, it was 3.0. Thus, one may suggest that H₂SO₄ preferentially reacts with the aromatic compounds, which have an ARI of 3.0, and they remained in the acidic tar following the procedure described in the Section 2. Another interesting observation from the data in Figure 1a is the stronger impact of the H₂SO₄ treatment on HTREB (higher slope) compared to that of the SIHGO treatment. A higher reactivity of HTREB-heavy aromatic compounds towards reacting with H₂SO₄ under the conditions of the procedure described above could be expected from the limited data of Figure 1. Another interesting observation from the sulfuric acid treatment of the SIHGO and HTREB is that after the first treatment (addition of 6% H₂SO₄) in both VGOs, the basic nitrogen components are completely removed. This finding is in line with that reported by Caeiro et al. [51], who treated an industrial FCC gas oil with a stoichiometric amount of H₂SO₄ (95%), and the result of the treatment they communicated was rejection of high nitrogen content. As the resin content does not decrease after the addition of 6% H₂SO₄ but only that of heavy aromatic compounds (see Figure 1a,b), one may suggest that the basic nitrogen components are concentrated in the heavy aromatic fraction, that has a refractive index at 20 °C higher than 1.59 and ARI of 3.0.

3.2. Catalytic Cracking of Studied Eight Vacuum Gas Oils

Figure 2 presents a graph of the dependence of VGO conversion on the catalyst-to-oil ratio.

The data in Figure 2 indicate that the least reactive VGO under the cracking conditions studied is REB, while the most reactive is SIHGO (76% H₂SO₄). As a measure of the reactivity of the studied VGOs, the catalyst-to-oil ratio interpolated at a constant conversion of 65 wt.% was used (

Table 4. Catalyst-to-oil ratio interpolated at constant conversion of 65 wt.%.

Vacuum Gas Oils	SIHGO	REB	HTREB	SIHGO (6% H2SO4)	SIHGO (16% H2SO4)	SIHGO (46% H2SO4)	SIHGO (76% H2SO4)	HTREB (6% H2SO4)
Catalyst-to-oil ratio interpolated at 65 wt.% conversion, wt./wt.	3.8	4.2	3.7	2.2	1.9	1.3	1	1.5

).

By regression of feed characterizing data and that of Table 4, Equation (7) was developed.

$$\begin{gathered} CTOat65\%conv=5.9715+40.82\times BN+0.0606\times C_A-0.4193\times H \\ \mathrm{R}=0.999,\mathrm{s}\mathrm{t}.\mathrm{error}=0.08\mathrm{w}\mathrm{t}./\mathrm{w}\mathrm{t}. \end{gathered}$$

(8)

where,

CTO at 65% conv = Catalyst-to-oil ratio interpolated at 65 wt.% conversion, wt./wt.;

BN = Content of basic nitrogen in the FCC VGO feed, wt.%;

C_A = Aromatic carbon content in FCC VGO feed, calculated by Equation (1), wt.%;

H = Hydrogen content in FCC VGO feed, calculated by Equation (2), wt.%.

Equation (8) suggests that the reactivity of a FCC VGO feed depends on three characterizing parameters: aromatic carbon, hydrogen, and basic nitrogen contents. The higher the aromatic carbon and basic nitrogen contents, and the lower the hydrogen content, the higher the catalyst-to-oil ratio needed to achieve the VGO conversion of 65 wt.%. The sulfuric acid-treated vacuum gas oils (SIHGO 6% H₂SO₄ and HTREB 6% H₂SO₄), whose basic nitrogen has been completely removed, but whose group hydrocarbon composition has not been considerably affected in comparison to the parent SIHGO and HTREB, are cracked much more easily than the original gas oils. For a catalyst-to-oil ratio of 2 wt./wt. the difference in conversions between SIHGO and SIHGO 6% H₂SO₄ is 16.7 wt.% and that between HTREB and HTREB 6% H₂SO₄ is 24.8 wt.%. Whereas at the highest catalyst-to-oil ratio of about 5 wt./wt. this difference is much lower 3.6 wt.% for SIHGO gas oils, and 4.5 wt.% for the HTREB gas oils. This observation can be explained by the poisoning effect of basic nitrogen components on the Brønsted acid sites responsible for cracking [51].

The difference is bigger for low catalyst-to-oil ratios because at low catalyst-to-oil ratios, the quantity of basic nitrogen components per mass of catalyst is higher, and poisoning will be more pronounced. At higher catalyst-to-oil ratios, the influence of basic nitrogen components on crackability is less pronounced because their relative amount in the feed compared to the catalyst mass is lower. After the removal of basic nitrogen, the effect of hydrocarbon composition, quantified by the contents of aromatic carbon and hydrogen, on the reactivity of FCC gas oils becomes more prominent.

Figure 2 shows a plot of the dependence of gasoline yield on the catalyst-to-oil ratio for the eight vacuum gas oils under study. It is evident from this data that for the sulfuric acid-treated VGOs, the maximum gasoline yield is found between CTO of 3 and 4 wt./wt. A calculation of conversion for the sulfuric acid-treated VGOs from the data in Figure 1 for a catalyst-to-oil ratio of 3.5 wt./wt. may define the conversion where overcracking occurred. In previous studies [38,39], it was found that the empirical parameters C_A and H can be used to calculate the conversion where overcracking occurred using Equation (9).

$$OvercrackingFCCconversion\left(wt.\%\right)=-0.36441\times C_A+6.6859\times H$$

(9)

Table 5. Observed and calculated by Equation (9) overcracking conversion of the studied eight vacuum gas oils.

Sulfuric Acid Treated VGOs	SIHGO (6% H2SO4)	SIHGO (16% H2SO4)	SIHGO (46% H2SO4)	SIHGO (76% H2SO4)	HTREB (6% H2SO4)	SIHGO *	REB *	HTREB *
Observed overcracking conversion at 3.5 wt./wt. CTO, wt.%	71.9	74.9	81.7	85.8	79.5	70.8	70.5	76.5
Calculated by Equation (8) overcracking conversion, wt.%	74.0	75.6	82.5	85.4	79.0	70.8	72.8	76.9

* Note: The observed conversions of SIHGO, REB, and HTREB VGOs are considered at CTO between 4.7 and 5.1 wt./wt.%, which was the maximum CTO for these FCC feeds.

summarizes the data observed and calculated by Equation (9) overcracking conversions of the investigated VGOs. It should be noted here that the overcracking of SIHGO, REB, and HTREB VGOs was not possible to define exactly because no reduction of gasoline yield can be seen during conversion enhancement, similar to that evident for the sulfuric acid-treated VGOs from the data in Figure 3. In this case, for these three VGOs, the conversion attained at the maximum CTO was assumed to be representative of their overcracking conversion. It can be expected that the overcracking conversion of SIHGO should not be higher than that of SIHGO (6% H₂SO₄) and that the maximal conversion of HTREB would not be higher than that of HTREB (6% H₂SO₄) because the removal of heavy aromatics by the sulfuric acid from both SIHGO and HTREB ought not to lead to obtaining a higher overcracking conversion from that of SIHGO (6% H₂SO₄) and that of HTREB (6% H₂SO₄). Thus, the conversion level of SIHGO of 70.8 wt.% was observed at CTO of 4.7 wt./wt. may be safely expected to be the overcracking point or very close to it. The same speculation for the overcracking point of HTREB may define the conversion level of 76.5 wt.% observed at CTO of 5.3 wt./wt. as the overcracking conversion or very close to the maximal conversion value.

Comparing the calculated by Equation (9) overcacking conversions with those apparent for the sulfuric acid-treated VGOs, an excellent agreement can be seen that can be considered as a reasonable basis to assume that the reported overcracking conversions for SIHGO, REB and HTREB should be very close to the true values. Equation (9) was developed for FCC feeds whose overcracking conversions varied between 50 and 85 wt.% [38], and it showed an adequate prediction for the data reported in [20] and [24]. The data set of the sulfuric-treated VGOs in this work also shows very good predictability.

The results generated in this research suggest that the feed characterizing parameters, aromatic carbon, hydrogen, and basic nitrogen contents, are sufficient to be used for forecasting the FCC feed reactivity and the conversion at maximum gasoline yield.

Figure 4 displays a chart of the dependence of dry gas yield on the catalyst-to-oil ratio.

There is a band with about 0.5 wt.% width that shows an increasing trend of dry gas yield with the enhancement of CTO for all studied feeds. It is difficult to discern a feed with a notable tendency to make more dry gas than the other feeds, probably because the dry gas is considered a product of thermal cracking rather than catalytic cracking.

Figure 5 exhibits a plot of C₃ fraction yield dependence on CTO for the eight VGOs.

In contrast to the dry gas yield, the C₃ fraction yield discerns that REB is the FCC feed that makes the least C₃ fraction yield, while SIHGO (76% H₂SO₄) is the VGO that produces the highest yield of C₃ fraction. This observation is in line with the reactivity of the examined FCC feeds. REB is the least reactive, while SIHGO (76% H₂SO₄) is the most reactive, and for that reason, REB makes the least C₃ fraction yield, while SIHGO (76% H₂SO₄) produces the highest yield of C₃ fraction.

Figure 6 presents a graph of the relationship between C₄ fraction yield and CTO for the studied VGOs.

Similar to the C₃ fraction yield, the C₄ fraction yield distinguishes that REB is the FCC feed that makes the least C₄ fraction yield, while SIHGO (76% H₂SO₄) is the VGO that produces the highest yield of C₄ fraction. The explanation of this finding is the same as that mentioned above for the C₃ fraction yield variation.

Figure 7 indicates a diagram of the dependence of LCO yield on CTO for the studied VGOs.

The data in Figure 7 indicates that SIHGO makes the highest yield of LCO, while SIHGO (76% H₂SO₄) makes the lowest yield of LCO. SIHGO yields a higher amount of LCO than the REB and the HTREB. A possible reason for this observation may be the higher content of material boiling below 338 °C (20 wt.% in SIHGO versus 6 wt.% in HTREB and 4 wt.% in REB). The content of the material boiling below 338 °C in a FCC feed cannot be the single reason for registering different LCO yields. For example, REB, despite its lower content of LCO (4 wt.%) relative to HTREB (6 wt.% of LCO), makes more LCO than HTREB, which probably owes to the lower reactivity of REB, which results in a lower extent of conversion of the LCO boiling range material in the REB into gasoline, gas and coke. Indeed, Harding et al. [19] showed that the material boiling below 370 °C makes more LCO than the original SIHGO, which can be the explanation for the highest yield of LCO observed during the cracking of SIHGO in the current study.

Figure 8 indicates how the HCO yield varies with changing the CTO for the studied feeds. Non-sulfuric acid-treated VGOs (SIHGO, REB and HTREB) showed much higher HCO yield due to their lower reactivity. SIHGO demonstrates a lower yield of HCO relative to REB and HTREB, probably because of the lower content of the material boiling above 338 °C. With the other feeds, the higher the reactivity, the lower the yield of HCO.

Figure 9 shows a plot of the linear dependence of coke yield on CTO for the studied VGOs. It is evident from this data that SIHGO and REB exhibit the highest yield of coke, while SIHGO (76% H₂SO₄) demonstrates the lowest coke yield.

Figure 10 shows LPG olefinicity variation as a function of CTO for the eight VGOs.

The data in Figure 10 displays that LPG olefinicity decreases with CTO enhancement for all feeds, being higher for the non-sulfuric acid-treated VGOs (SIHGO, REB and HTREB) and the lowest for SIHGO (76% H₂SO₄). The higher the feed reactivity, the lower the LPG olefinicity because at higher conversion, a bigger proportion of secondary hydrogen transfer reactions leading to lower olefinicity will take place.

Table 6. Product yields, conversion, gasoline composition, GC-RON, and GC-MON were obtained from the catalytic cracking of eight VGOs at the point of maximal gasoline yield.

Product Yields, Gasoline Composition and Octane Numbers		SIHGO	REB	HTREB	SIHGO (6% H2SO4)	SIHGO (16% H2SO4)	SIHGO (46% H2SO4)	SIHGO (76% H2SO4)	HTREB (6% H2SO4)
Dry gas (C2-fraction)	wt.%	2.03	2.7	2.35	1.56	2.02	1.4	1.5	1.57
C3 fraction	wt.%	5.75	4.83	6.01	5.42	6.32	7.02	7.36	6.0
C4 fraction	wt.%	10.25	9.97	11.11	10.02	10	13.29	14.5	11.45
Gasoline (CN)	wt.%	49	48.3	53.8	52.2	54	58.0	60.6	58.0
LCO	wt.%	19.7	16.8	17.6	19.2	17.7	13.9	10.9	14.2
HCO	wt.%	9.5	13.5	5.9	8.8	7.4	4.29	3.3	6.3
Coke	wt.%	3.72	4.0	3.16	2.55	2.56	2.1	1.84	2.48
Δ Coke	wt.%	0.76	0.69	0.61	0.68	0.61	0.51	0.45	0.57
Maximal Conversion	wt.%	70.8	70.5	76.5	71.9	74.9	81.7	85.8	80.4
Gasoline n-paraffins	wt.%	4.2	4	4.1	3.9	3.9	4.1	4.4	4.2
Gasoline iso-paraffins	wt.%	36.8	34.3	37.8	36.5	36.9	40.4	44.3	41.9
Gasoline aromatics	wt.%	35.2	34	33	36.5	37.4	30.4	28.7	29.8
Gasoline naphthenes	wt.%	9.4	9.6	9.9	11	10.5	10.7	9.7	10.5
Gasoline olefins	wt.%	14.5	18	15.3	12.1	11.3	14.4	12.9	13.6
GC-MON		79.5	80.5	80.7	80.3	80.5	79.6	79.5	79.7
GC-RON		89.5	90	89.7	88.6	88.8	87.9	87.2	87.8

summarizes the conversion, product yields, gasoline composition and gas chromatographic research octane number (RON) and motor octane number at the maximal conversion (overcracking point conversion).

In order to determine the presence or absence of statistically meaningful relations between the FCC feed characterizing parameters (data in Table 3) and the experimental cracking data from Table 6, intercriteria analysis (ICrA) evaluation was performed.

Table 7. μ-values of ICrA evaluation of the feed characterizing data (Table 3) and cracking data at maximum gasoline yield (Table 6).

μ	D15	RI	ARI	KW	CA	H	Sulfur	Total N	Basic N	Sat	L. Aro	M. Aro	H. Aro	Resins	C2	C3	C4	CN	LCO	HCO	Coke	Δ Coke	Conv
D15	1.00	0.93	0.93	0.07	0.86	0.07	0.68	0.68	0.54	0.11	0.32	0.43	0.89	0.79	0.57	0.07	0.07	0.00	0.86	0.96	0.86	0.93	0.04
RI	0.93	1.00	0.93	0.07	0.93	0.07	0.71	0.68	0.54	0.11	0.32	0.46	0.86	0.82	0.54	0.11	0.14	0.04	0.93	0.93	0.86	0.93	0.07
ARI	0.93	0.93	1.00	0.14	0.86	0.14	0.64	0.75	0.61	0.18	0.39	0.39	0.86	0.75	0.61	0.11	0.14	0.04	0.86	0.93	0.93	0.86	0.07
KW	0.07	0.07	0.14	1.00	0.14	0.93	0.21	0.18	0.18	0.89	0.68	0.43	0.11	0.14	0.46	0.82	0.86	0.89	0.14	0.07	0.18	0.07	0.86
CA	0.86	0.93	0.86	0.14	1.00	0.07	0.79	0.64	0.50	0.11	0.32	0.46	0.86	0.89	0.54	0.11	0.14	0.11	0.93	0.86	0.86	0.93	0.07
H	0.07	0.07	0.14	0.93	0.07	1.00	0.18	0.14	0.14	0.96	0.68	0.50	0.04	0.07	0.39	0.89	0.93	0.89	0.14	0.07	0.11	0.00	0.93
Sulfur	0.68	0.71	0.64	0.21	0.79	0.18	1.00	0.46	0.32	0.14	0.29	0.57	0.75	0.82	0.54	0.29	0.25	0.29	0.71	0.71	0.64	0.75	0.25
Total N	0.68	0.68	0.75	0.18	0.64	0.14	0.46	1.00	0.86	0.18	0.32	0.18	0.71	0.54	0.54	0.11	0.14	0.07	0.61	0.68	0.71	0.64	0.07
Basic N	0.54	0.54	0.61	0.18	0.50	0.14	0.32	0.86	1.00	0.18	0.29	0.07	0.57	0.39	0.50	0.11	0.14	0.07	0.46	0.54	0.64	0.50	0.07
Sat	0.11	0.11	0.18	0.89	0.11	0.96	0.14	0.18	0.18	1.00	0.64	0.46	0.07	0.04	0.36	0.86	0.89	0.86	0.18	0.11	0.14	0.04	0.89
L. Aro	0.32	0.32	0.39	0.68	0.32	0.68	0.29	0.32	0.29	0.64	1.00	0.54	0.32	0.32	0.57	0.57	0.61	0.64	0.32	0.32	0.43	0.32	0.61
M. Aro	0.43	0.46	0.39	0.43	0.46	0.50	0.57	0.18	0.07	0.46	0.54	1.00	0.39	0.54	0.25	0.54	0.50	0.54	0.46	0.46	0.32	0.43	0.57
H. Aro	0.89	0.86	0.86	0.11	0.86	0.04	0.75	0.71	0.57	0.07	0.32	0.39	1.00	0.82	0.64	0.07	0.04	0.07	0.79	0.93	0.86	0.89	0.04
Resins	0.79	0.82	0.75	0.14	0.89	0.07	0.82	0.54	0.39	0.04	0.32	0.54	0.82	1.00	0.57	0.14	0.11	0.18	0.82	0.82	0.75	0.86	0.14
C2	0.57	0.54	0.61	0.46	0.54	0.39	0.54	0.54	0.50	0.36	0.57	0.25	0.64	0.57	1.00	0.43	0.39	0.43	0.46	0.61	0.64	0.57	0.39
C3	0.07	0.11	0.11	0.82	0.11	0.89	0.29	0.11	0.11	0.86	0.57	0.54	0.07	0.14	0.43	1.00	0.96	0.93	0.18	0.11	0.07	0.07	0.96
C4	0.07	0.14	0.14	0.86	0.14	0.93	0.25	0.14	0.14	0.89	0.61	0.50	0.04	0.11	0.39	0.96	1.00	0.89	0.21	0.07	0.11	0.07	0.93
CN	0.00	0.04	0.04	0.89	0.11	0.89	0.29	0.07	0.07	0.86	0.64	0.54	0.07	0.18	0.43	0.93	0.89	1.00	0.11	0.04	0.07	0.07	0.96
LCO	0.86	0.93	0.86	0.14	0.93	0.14	0.71	0.61	0.46	0.18	0.32	0.46	0.79	0.82	0.46	0.18	0.21	0.11	1.00	0.86	0.79	0.86	0.14
HCO	0.96	0.93	0.93	0.07	0.86	0.07	0.71	0.68	0.54	0.11	0.32	0.46	0.93	0.82	0.61	0.11	0.07	0.04	0.86	1.00	0.86	0.89	0.07
Coke	0.86	0.86	0.93	0.18	0.86	0.11	0.64	0.71	0.64	0.14	0.43	0.32	0.86	0.75	0.64	0.07	0.11	0.07	0.79	0.86	1.00	0.86	0.04
Δ Coke	0.93	0.93	0.86	0.07	0.93	0.00	0.75	0.64	0.50	0.04	0.32	0.43	0.89	0.86	0.57	0.07	0.07	0.07	0.86	0.89	0.86	1.00	0.04
Conv	0.04	0.07	0.07	0.86	0.07	0.93	0.25	0.07	0.07	0.89	0.61	0.57	0.04	0.14	0.39	0.96	0.93	0.96	0.14	0.07	0.04	0.04	1.00
Bottoms cracking	0.04	0.11	0.11	0.89	0.18	0.89	0.29	0.11	0.11	0.86	0.71	0.54	0.07	0.18	0.43	0.86	0.89	0.93	0.18	0.04	0.14	0.11	0.89
Gasoline n-paraffins	0.21	0.29	0.29	0.64	0.29	0.61	0.36	0.25	0.29	0.57	0.43	0.36	0.25	0.25	0.39	0.64	0.68	0.64	0.29	0.21	0.36	0.29	0.61
Gasoline iso-paraffins	0.04	0.11	0.11	0.89	0.18	0.89	0.29	0.14	0.14	0.86	0.64	0.46	0.07	0.14	0.43	0.93	0.96	0.93	0.18	0.04	0.14	0.11	0.89
Gasoline aromatics	0.75	0.79	0.71	0.21	0.79	0.21	0.71	0.46	0.36	0.25	0.39	0.61	0.71	0.79	0.54	0.25	0.21	0.25	0.79	0.79	0.68	0.75	0.29
Gasoline naphthenes	0.39	0.36	0.36	0.54	0.36	0.61	0.39	0.29	0.14	0.57	0.64	0.71	0.36	0.43	0.36	0.57	0.54	0.57	0.43	0.43	0.29	0.32	0.61
Gasoline olefins	0.68	0.68	0.75	0.39	0.61	0.39	0.39	0.68	0.64	0.43	0.50	0.29	0.61	0.50	0.64	0.36	0.39	0.29	0.61	0.68	0.68	0.61	0.32
GC-MON	0.57	0.57	0.64	0.43	0.57	0.36	0.39	0.50	0.39	0.39	0.64	0.32	0.57	0.50	0.61	0.25	0.29	0.32	0.57	0.57	0.68	0.57	0.29
GC-RON	0.86	0.86	0.93	0.21	0.79	0.21	0.57	0.75	0.64	0.25	0.46	0.32	0.79	0.68	0.68	0.18	0.21	0.11	0.79	0.86	0.89	0.79	0.14

Note: D15 = density at 15 °C, g/cm³; RI = refractive index; ARI = aromatic ring index; C_A = aromatic carbon content, wt.%; H hydrogen content, wt.%; Sat = content of saturates, wt.%; L. Aro = content of light aromatics, wt.%; M. Aro = content of middle aromatics, wt.%; H. Aro = content of heavy aromatics, wt.%; C₂ = yield of dry gas; C₃ = yield of C₃ fraction, wt.%; C₄ = yield of C₄ fraction, wt.%; CN = yield of gasoline (cracked naphtha), wt.%; LCO = yield of LCO, wt.%; HCO = yield of HCO, wt.%; Coke = yield of coke, wt.%; Conv. = conversion at the maximal gasoline point (overcracking), wt.%; GC-RON = gas chromatographic research octane number; GC-MON = gas chromatographic motor octane number.

and

Table 8. υ -values of ICrA evaluation of the feed characterizing data (Table 3) and cracking data at maximum gasoline yield (Table 6).

υ	D15	RI	ARI	KW	CA	H	Sulfur	Total N	Basic N	Sat	L. Aro	M. Aro	H. Aro	Resins	C2	C3	C4	CN	LCO	HCO	Coke	Δ Coke	Conv
D15	0.00	0.04	0.04	0.89	0.11	0.89	0.25	0.07	0.07	0.86	0.64	0.50	0.04	0.14	0.39	0.89	0.89	0.96	0.11	0.00	0.07	0.07	0.93
RI	0.04	0.00	0.07	0.93	0.07	0.93	0.25	0.11	0.11	0.89	0.68	0.50	0.11	0.14	0.46	0.89	0.86	0.96	0.07	0.07	0.11	0.04	0.93
ARI	0.04	0.07	0.00	0.86	0.14	0.86	0.32	0.04	0.04	0.82	0.61	0.57	0.11	0.21	0.39	0.89	0.86	0.96	0.14	0.07	0.04	0.11	0.93
KW	0.89	0.93	0.86	0.00	0.86	0.07	0.75	0.61	0.46	0.11	0.32	0.54	0.86	0.82	0.54	0.18	0.14	0.11	0.86	0.93	0.79	0.89	0.14
CA	0.11	0.07	0.14	0.86	0.00	0.93	0.18	0.14	0.14	0.89	0.68	0.50	0.11	0.07	0.46	0.89	0.86	0.89	0.07	0.14	0.11	0.04	0.93
H	0.89	0.93	0.86	0.07	0.93	0.00	0.79	0.64	0.50	0.04	0.32	0.46	0.93	0.89	0.61	0.11	0.07	0.11	0.86	0.93	0.86	0.96	0.07
Sulfur	0.25	0.25	0.32	0.75	0.18	0.79	0.00	0.29	0.29	0.82	0.68	0.36	0.18	0.11	0.43	0.68	0.71	0.68	0.25	0.25	0.29	0.18	0.71
Total N	0.07	0.11	0.04	0.61	0.14	0.64	0.29	0.00	0.00	0.61	0.46	0.57	0.11	0.21	0.25	0.68	0.64	0.71	0.18	0.11	0.04	0.11	0.71
Basic N	0.07	0.11	0.04	0.46	0.14	0.50	0.29	0.00	0.00	0.46	0.36	0.54	0.11	0.21	0.14	0.54	0.50	0.57	0.18	0.11	0.04	0.11	0.57
Sat	0.86	0.89	0.82	0.11	0.89	0.04	0.82	0.61	0.46	0.00	0.36	0.50	0.89	0.93	0.64	0.14	0.11	0.14	0.82	0.89	0.82	0.93	0.11
L. Aro	0.64	0.68	0.61	0.32	0.68	0.32	0.68	0.46	0.36	0.36	0.00	0.43	0.64	0.64	0.43	0.43	0.39	0.36	0.68	0.68	0.54	0.64	0.39
M. Aro	0.50	0.50	0.57	0.54	0.50	0.46	0.36	0.57	0.54	0.50	0.43	0.00	0.54	0.39	0.71	0.43	0.46	0.43	0.50	0.50	0.61	0.50	0.39
H. Aro	0.04	0.11	0.11	0.86	0.11	0.93	0.18	0.11	0.11	0.89	0.64	0.54	0.00	0.11	0.32	0.89	0.93	0.89	0.18	0.04	0.07	0.04	0.93
Resins	0.14	0.14	0.21	0.82	0.07	0.89	0.11	0.21	0.21	0.93	0.64	0.39	0.11	0.00	0.39	0.82	0.86	0.79	0.14	0.14	0.18	0.07	0.82
C2	0.39	0.46	0.39	0.54	0.46	0.61	0.43	0.25	0.14	0.64	0.43	0.71	0.32	0.39	0.00	0.57	0.61	0.57	0.54	0.39	0.32	0.39	0.61
C3	0.89	0.89	0.89	0.18	0.89	0.11	0.68	0.68	0.54	0.14	0.43	0.43	0.89	0.82	0.57	0.00	0.04	0.07	0.82	0.89	0.89	0.89	0.04
C4	0.89	0.86	0.86	0.14	0.86	0.07	0.71	0.64	0.50	0.11	0.39	0.46	0.93	0.86	0.61	0.04	0.00	0.11	0.79	0.93	0.86	0.89	0.07
CN	0.96	0.96	0.96	0.11	0.89	0.11	0.68	0.71	0.57	0.14	0.36	0.43	0.89	0.79	0.57	0.07	0.11	0.00	0.89	0.96	0.89	0.89	0.04
LCO	0.11	0.07	0.14	0.86	0.07	0.86	0.25	0.18	0.18	0.82	0.68	0.50	0.18	0.14	0.54	0.82	0.79	0.89	0.00	0.14	0.18	0.11	0.86
HCO	0.00	0.07	0.07	0.93	0.14	0.93	0.25	0.11	0.11	0.89	0.68	0.50	0.04	0.14	0.39	0.89	0.93	0.96	0.14	0.00	0.11	0.07	0.93
Coke	0.07	0.11	0.04	0.79	0.11	0.86	0.29	0.04	0.04	0.82	0.54	0.61	0.07	0.18	0.32	0.89	0.86	0.89	0.18	0.11	0.00	0.07	0.93
Δ Coke	0.07	0.04	0.11	0.89	0.04	0.96	0.18	0.11	0.11	0.93	0.64	0.50	0.04	0.07	0.39	0.89	0.89	0.89	0.11	0.07	0.07	0.00	0.93
Conv	0.93	0.93	0.93	0.14	0.93	0.07	0.71	0.71	0.57	0.11	0.39	0.39	0.93	0.82	0.61	0.04	0.07	0.04	0.86	0.93	0.93	0.93	0.00
Coke sel.	0.07	0.11	0.04	0.82	0.11	0.89	0.29	0.04	0.04	0.86	0.57	0.61	0.07	0.18	0.36	0.93	0.89	0.93	0.18	0.11	0.00	0.07	0.96
Bottoms cracking	0.93	0.89	0.89	0.11	0.82	0.11	0.68	0.68	0.54	0.14	0.29	0.43	0.89	0.79	0.57	0.14	0.11	0.07	0.82	0.96	0.82	0.86	0.11
Gasoline n-paraffins	0.64	0.61	0.61	0.25	0.61	0.29	0.57	0.43	0.32	0.32	0.46	0.57	0.61	0.61	0.50	0.25	0.21	0.25	0.61	0.68	0.57	0.57	0.29
Gasoline iso-paraffins	0.93	0.89	0.89	0.11	0.82	0.11	0.68	0.64	0.50	0.14	0.36	0.50	0.89	0.82	0.57	0.07	0.04	0.07	0.82	0.96	0.82	0.86	0.11
Gasoline aromatics	0.21	0.21	0.29	0.79	0.21	0.79	0.25	0.32	0.29	0.75	0.61	0.36	0.25	0.18	0.46	0.75	0.79	0.75	0.21	0.21	0.29	0.21	0.71
Gasoline naphthenes	0.54	0.61	0.61	0.43	0.61	0.36	0.54	0.54	0.54	0.39	0.32	0.21	0.57	0.50	0.61	0.39	0.43	0.39	0.54	0.54	0.64	0.61	0.36
Gasoline olefins	0.29	0.32	0.25	0.61	0.39	0.61	0.57	0.11	0.00	0.57	0.50	0.68	0.36	0.46	0.36	0.64	0.61	0.71	0.39	0.32	0.29	0.36	0.68
GC-MON	0.32	0.36	0.29	0.50	0.36	0.57	0.50	0.21	0.18	0.54	0.29	0.57	0.32	0.39	0.32	0.68	0.64	0.61	0.36	0.36	0.21	0.32	0.64
GC-RON	0.11	0.14	0.07	0.79	0.21	0.79	0.39	0.04	0.00	0.75	0.54	0.64	0.18	0.29	0.32	0.82	0.79	0.89	0.21	0.14	0.07	0.18	0.86

Note: D15 = density at 15 °C, g/cm³; RI = refractive index; ARI = aromatic ring index; C_A = aromatic carbon content, wt.%; H hydrogen content, wt.%; Sat = content of saturates, wt.%; L. Aro = content of light aromatics, wt.%; M. Aro = content of middle aromatics, wt.%; H. Aro = content of heavy aromatics, wt.%; C₂ = yield of dry gas; C₃ = yield of C₃ fraction, wt.%; C₄ = yield of C₄ fraction, wt.%; CN = yield of gasoline (cracked naphtha), wt.%; LCO = yield of LCO, wt.%; HCO = yield of HCO, wt.%; Coke = yield of coke, wt.%; Conv. = conversion at the maximal gasoline point (overcracking), wt.%; GC-RON = gas chromatographic research octane number; GC-MON = gas chromatographic motor octane number.

summarize μ-values and ν –values obtained from ICrA. Concerning the yields at the point of maximum gasoline yield, one can see that the dry gas (C₂) yield has no statistically significant relation to any FCC feed characteristic, which could be a result of the fact that the dry gas is a product of thermal cracking. The C₃ fraction yield has negative consonances with the contents of heavy aromatics and resins. Regressing the data allowed us the develop the following correlation for C₃ fraction yield, as shown in Equation (10).

$$\begin{gathered} C_3\left(overcrackingyield\right)=13.567-0.03784\times HeavyAro-0.01821\times MW-0.5190\times Resins \\ \mathrm{R}=0.984,\mathrm{s}\mathrm{t}.\mathrm{error}=0.19\mathrm{w}\mathrm{t}.\% \end{gathered}$$

(10)

where:

$C_3\left(overcrackingyield\right)=$ The yield of C₃ fraction at the overcracking point, wt.%;

$HeavyAro=$ Content of heavy aromatics in FCC feed, wt.%;

$MW=$ Molecular weight of FCC feed, g/mol;

$Resins=$ Content of resins in FCC feed, wt.%.

Equation (10) indicates that the increase of heavy aromatics and resin contents and that of molecular weight of FCC feed leads to a reduction in the C₃ fraction yield.

The data in Table 7 and Table 8 indicate that the yields of C₄ fraction and gasoline (cracked naphtha = CN) have strong negative consonances with the heavy aromatics content in the FCC feed, which is very well illustrated in Figure 11a,b. The higher the FCC feed heavy aromatics content is, the lower the yields of C₄ fraction and gasoline are.

Regarding the yields of LCO and HCO, the following regressions were developed (Equations (11) and (12)).

$$\begin{gathered} LCO\left(overcrackingyield\right)=11.547+0.28384\times HeavyAro+0.0792\times 338°Cyield \\ \mathrm{R}=0.883,\mathrm{s}\mathrm{t}.\mathrm{error}=1.66\mathrm{w}\mathrm{t}.\% \end{gathered}$$

(11)

where,

$LCO\left(overcrackingyield\right)=$ LCO yield at the point of maximum gasoline yield, wt.%;

$338°Cyield=$ Content of fraction boiling below 338 °C in FCC feed, wt.%

$$\begin{gathered} HCO\left(overcrackingyield\right)=4.6398+0.2304\times HeavyAro+0.7899\times Resins-0.0948\times 338°Cyield \\ \mathrm{R}=0.937,\mathrm{s}\mathrm{t}.\mathrm{error}=1.5\mathrm{w}\mathrm{t}.\% \end{gathered}$$

(12)

The yield of coke at the overcracking point and the Δ Coke (this is the coke laid down on the catalyst) was found to best correlate to the contents of hydrogen and heavy aromatics, as exemplified in Equations (13) and (14).

$$\begin{gathered} Cokeyield=1.9896\times EXP(0.025\times HeavyAro) \\ \mathrm{R}=0.883,\mathrm{s}\mathrm{t}.\mathrm{error}=0.39\mathrm{w}\mathrm{t}.\% \end{gathered}$$

(13)

$$\begin{gathered} ∆Coke=58.134\times EXP(-0.365\times H) \\ \mathrm{R}=0.979,\mathrm{s}\mathrm{t}.\mathrm{error}=0.029\mathrm{w}\mathrm{t}.\% \end{gathered}$$

(14)

The bottom cracking, defined as ${\displaystyle \frac{YieldofLCO}{(YieldofLCO+yieldofHCO}}\times 100$, was found to depend mainly on the content of heavy aromatic compounds in the FCC feed as shown in Figure 12.

Some characteristics of gasoline were found to be related to the feed ARI. Figure 13 indicates the relationships between FCC feed ARI and the iso-paraffin content in the gasoline (Figure 13a) and its research octane number (Figure 13b).

As apparent from the data in Table 7 and Table 8, the gasoline aromatic content has a weak positive consonance with the feed aromatic carbon content (μ = 0.79; υ = 0.21), and the gasoline olefin content has a weak positive consonance (μ = 0.75; υ = 0.25) with feed ARI. The gasoline n-paraffin and naphthene contents are in dissonance with any of the feed characteristics studied.

All results obtained in this research underline the significant role the heavy aromatic content plays in the catalytic cracking of vacuum gas oils. The heavy aromatics, as shown in the data in Table 3, can be diminished not only by sulfuric acid treatment but also by hydrotreatment. HTREB has been obtained by hydrotreatment of REB in a commercial FCC feed hydrotreater. An earlier study [52] showed how the severity of hydrotreatment in commercial FCC feed hydrotreaters can affect the quality of HTREB. As evident from the data in Table 3, as a result of the hydrotreatment of REB, 47.5% of the heavy aromatics components are converted to other FCC feed components. The heavy aromatics in REB, HTREB, and SIHGO may have different impacts on the FCC process performance. In order to assess how the heavy aromatics in the three VGOs (SIHGO, REB and HTREB) influence conversion and product yields, the approximate linear combination technique described by Harding et al. [19] was applied to calculate the yield distribution of the heavy aromatics at a constant catalyst-to-oil ratio of 4.0 wt./wt. from the cracking experiments with SIHGO, SINGO (6%H₂SO₄), HTREB(6%H₂SO₄), and REB.

Table 9. Conversion and product yields at a catalyst-to-oil ratio of 4.0 wt./wt. calculated for the heavy aromatics, which are contained in the vacuum gas oils SIHGO, REB, and HTREB.

Conversion and Yields of Products at CTO of 4.0 wt./wt., wt.%	SIHGO (HARO)	REB (HARO)	HTREB (HARO)
C2	2.4	2.7	−0.4
C3	2.6	4.4	−2.5
C4	3.1	6.9	−3.6
CN	22.2	41.0	21.3
LCO	32.1	21.1	26.2
HCO	33.1	19.7	56.6
Coke	4.1	4.0	2.4
Conversion	35.1	59.1	17.2

summarizes the results of this comparison.

The data in Table 9 clearly indicate that the heavy aromatic components in the three studied vacuum gas oils have different crackability. The most crackable are the heavy aromatics from REB. During hydrotreatment, the heavy aromatics in REB are hydrogenated, and their content is significantly reduced at the expense of augmentation of saturate content (see the data of REB and HTREB in Table 3). Despite the significant reduction in the heavy aromatics and simultaneous increase in saturate content, the difference in reactivity between HTREB and REB is not as big as that between HTREB and HTREB (6% H₂SO₄) and that between SIHGO and SIHGO (6% H₂SO₄) (see the data in Figure 1 and Table 4). This suggests that the hydrogenation of the heavy aromatics during the FCC feed hydrotreatment does not significantly affect the removal of the most inhibiting species in the FCC process–the basic nitrogen compounds [53]. The low hydrodenitrogenation extent observed in the FCC feed hydrotreatment [54], along with the noted conversion of the neutral nitrogen species in basic ones [55,56,57] during the gas oil hydrotreatment, can explain why the difference in reactivity of REB and HTREB is not very big. The removed heavy aromatics from REB in the process of its hydrotreatment, as shown in the data in Table 9, exhibit much higher crackability than the heavy aromatics removed by treatment with 6% H₂SO₄ of both SIHGO and HTREB. Obviously, the treatment with 6% H₂SO₄ of both SIHGO and HTREB removes these species, which have a great influence on the VGO reactivity in the process of catalytic cracking, and their reactivity is considerably lower than that of the whole VGO. Based on the data in Table 9, one may suggest that the heavy aromatic fraction from HTREB contains a higher level of inhibitors for catalytic cracking than that in SIHGO. This observation coincides with the higher slope of heavy aromatics removal, which is evident for HTREB in comparison with SIHGO (see Figure 1a). The data in Table 9 also indicate that the heavy aromatic fraction of HTREB suppresses the formation of gaseous products (dry gas, C₃ fraction, and C₄ fraction). A similar suppressing effect was reported by Harding et al. [19], where the polar components were found to suppress the C₄ fraction and gasoline formation. Therefore, it can be deduced that the heavy aromatics contained in the various VGOs may have diverse effects on conversion level and product distribution during VGO catalytic cracking.

4. Conclusions

The hydrotreatment and sulfuric acid treatment of the vacuum gas oils REB, HTREB, and SIHGO decreased the contents of heavy aromatics and resins. A linear dependence was revealed between the amount of sulfuric acid used to treat SIHGO and HTREB and the contents of heavy aromatics and resins. The reactivity in the catalytic cracking of the studied vacuum gas oils was found to depend on three feed-characterizing parameters: the contents of aromatic carbon, nitrogen and basic nitrogen. The overcracking point of the sulfuric acid-treated VGOs was found to be at a catalyst-to-oil ratio of 3.5 wt./wt.%, while for the other three VGOs (REB, HTREB and SIHGO), the overcracking point was not possible to determine under the conditions studied. The correlation developed earlier to relate the empirically determined contents of aromatic carbon and hydrogen predicted the overcracking conversion of the sulfuric acid-treated VGOs with an average absolute deviation of 0.7 wt.% and maximal absolute deviation of 2.3 wt.%. The heavy aromatics in the VGOs were established to suppress the formation of FCC products: C₃ fraction, C₄ fraction, and gasoline, and decreased the bottoms cracking. On the other hand, it favored the production of LCO, HCO, and coke. The coke on the catalyst (Δ Coke) was found to depend on feed hydrogen content only, but not on the total nitrogen and basic nitrogen contents. The heavy aromatics contained in the three VGOs, REB, HTREB and SIHGO, exhibit quite a different crackability, differing by a factor of three. Hydrotreatment, while very efficient in reducing heavy aromatics, is not as capable of conversion enhancement as that achieved by sulfuric acid treatment, probably because of the inability of hydrotreatment to remove the most inhibiting basic species in the FCC VGO feed.

The FCC gasoline octane number and the content of iso-paraffins were established to depend on the aromatic ring index of the FCC feed.