Crude Slate, FCC Slurry Oil, Recycle, and Operating Conditions Effects on H-Oil ® Product Quality

: This paper evaluates the inﬂuence of crude oil (vacuum residue) properties, the processing of ﬂuid catalytic cracking slurry oil, and recycle of hydrocracked vacuum residue diluted with ﬂuid catalytic cracking heavy cycle oil, and the operating conditions of the H-Oil vacuum residue hydrocracking on the quality of the H-Oil liquid products. 36 cases of operation of a commercial H-Oil ® ebullated bed hydrocracker were studied at different feed composition, and different operating conditions. Intercriteria analysis was employed to deﬁne the statistically meaningful relations between 135 parameters including operating conditions, feed and products characteristics. Correlations and regression equations which related the H-Oil ® mixed feed quality and the operating conditions (reaction temperature, and reaction time (throughput)) to the liquid H-Oil ® products quality were developed. The developed equations can be used to ﬁnd the optimal performance of the whole reﬁnery considering that the H-Oil liquid products are part of the feed for the units: ﬂuid catalytic cracking, hydrotreating, road pavement and


Introduction
The ebullated bed vacuum residue H-Oil ® hydrocracking proved commercially to be able of achieving 93% conversion of vacuum residue into gas (15.2%), naphtha (10.2%), diesel (47.2%), vacuum gas oil (25.1%), and unconverted hydrocracked vacuum residue, also known as vacuum tower bottom product (VTB) (5.85%) [1,2]. However, the naphtha, the diesel, the vacuum gas oil, and the VTB from H-Oil ® are not finished marketable products and they require further processing. The naphtha and the diesel are hydrotreated to near zero sulphur level. The vacuum gas oil (VGO) is catalytically cracked. It was found that the properties of the H-Oil ® VGO varied in a wide range, depending on H-Oil ® feed structure and operation severity which affected the H-Oil ® VGO reactivity during its processing in the fluid catalytic cracking (FCC) [3,4]. The H-Oil ® feed structure consisted of straight run vacuum residue, FCC slurry oil (SLO) and recycle of partially blended fuel oil (PBFO). The PBFO is prepared from around 70% VTB and 30% FCC heavy cycle oil (HCO).

Results and Discussion
Investigations have shown that density and Kw (Watson characterization factor) of heavy oils very well correlate with their contents of saturates [1,8], hydrogen, and aromatic carbon [9][10][11]. Therefore, density and Kw can be used as indicators for aromaticity and hydrogen deficiency of the heavy oils. Figure 1 presents graphs of the relations of density with Kw, and hydrogen content of the mixed H-Oil ® feed, straight run vacuum residual oils (SRVROs), and H-Oil ® atmospheric tower bottom product (ATB), and VTB. These data show a very strong relation between density, Kw, and hydrogen content for the H-Oil ® ATB, and VTB, and a weaker relation for the SRVROs, and the H-Oil ® mixed feed. The mixed feed demonstrates a lower slope of decreasing of Kw with enhancement of density than the SRVROs. Since Kw depends on average boiling and density [9] this phenomenon can be explained with a lower average boiling point of the mixed feed. The addition of FCC SLO and recycle of partially blended fuel oil (PBFO) to the straight run vacuum residue indeed decreases the average boiling point of the mixed feed. It is difficult to find a reasonable explanation why the correlations of Kw, density and hydrogen content for the H-Oil ® residual oil products ATB, and VTB are stronger than those of the mixed feed, and the SRVROs.
The relations between 135 characterizing parameters for the 36 studied cases were investigated by the use of intercriteria analysis (ICrA). More information about the application of ICrA the reader can find in our recent studies [1,3]. ICrA defines the values of positive and negative consonance (µ) of the studied criteria (parameters) [1,3]. The meaning of µ = 0.75 ÷ 1.00 denotes a statistically meaningful positive relation, where the strong positive consonance exhibits values of µ = 0.95 ÷ 1.00, and the weak positive consonance exhibits values of µ = 0.75 ÷ 0.85. Respectively, the values of negative consonance with µ = 0.00 ÷ 0.25 means a statistically meaningful negative relation, where the strong negative consonance exhibits values of µ = 0.00 ÷ 0.05, and the weak negative consonance exhibits values of µ = 0.15 ÷ 0.25 [1,3].
of density than the SRVROs. Since Kw depends on average boiling and density [9] this phenomenon can be explained with a lower average boiling point of the mixed feed. The addition of FCC SLO and recycle of partially blended fuel oil (PBFO) to the straight run vacuum residue indeed decreases the average boiling point of the mixed feed. It is difficult to find a reasonable explanation why the correlations of Kw, density and hydrogen content for the H-Oil ® residual oil products ATB, and VTB are stronger than those of the mixed feed, and the SRVROs. The relations between 135 characterizing parameters for the 36 studied cases were investigated by the use of intercriteria analysis (ICrA). More information about the application of ICrA the reader can find in our recent studies [1,3]. ICrA defines the values of positive and negative consonance (μ) of the studied criteria (parameters) [1,3]. The meaning of μ = 0.75÷1.00 denotes a statistically meaningful positive relation, where the strong positive consonance exhibits values of μ = 0.95÷1.00, and the weak positive consonance exhibits values of μ = 0.75÷0.85. Respectively, the values of negative consonance with μ = The data in Table 1 confirm that for the studied 36 cases the density, and the Kw are equivalent substitutes of the contents of aromatic carbon, and hydrogen content for the H-Oil ® gas oils. The consonances µ of Kw and aromatic carbon content for HAGO (heavy Processes 2021, 9, 952 4 of 17 atmospheric gas oil), LVGO (light vacuum gas oil), and HVGO (heavy vacuum gas oil) are 0.00. The consonances µ of density and hydrogen content for HAGO, LVGO, and HVGO are also 0.00. The average number of aromatic rings predicted by the aromatic ring index (ARI) strongly correlates with density of HAGO, LVGO, and HVGO (µ = 0.98-0.99). The ARI of H-Oil ® gas oils was found to affect conversion and coke yield during catalytic cracking of H-Oil ® heavy oils [3].  Table 2 presents the range of variation of the properties of the mixed feed and of the products: naphtha, diesel, heavy atmospheric gas oil (HAGO), light vacuum gas oil (LVGO), heavy vacuum gas oil (HVGO), ATB, and VTB for the studied 36 cases. These data indicate that the properties of mixed feed and of the liquid products vary in a rather wide range. Properties of the liquid products from H-Oil ® are important because they control the reactivity of these streams during their further refining in processes like FCC and hydrotreatment [3,4,12,13] to produce finished marketable products. It was found in our earlier studies that the lower the Kw of H-Oil ® gas oils the lower their crackability in FCC is [3]. The higher the density, and the aromatics content in the H-Oil ® diesel the lower its reactivity during hydrotreatment [12][13][14]. It was reported in [1] and in [5,6,15] that the properties of H-Oil ® VTB affect the process of production of road asphalt whose feed contains H-Oil ® VTB. Therefore, understanding the factors controlling H-Oil ® liquid products properties can allow optimization of the whole refinery performance.   Table 3 shows some of the statistically meaningful relations between the H-Oil ® feed properties, H-Oil ® operating conditions and H-Oil ® product properties established by the use of ICrA. It is evident from these data that the H-Oil ® mixed feed Kw very strongly correlates with VTB density; ATB Kw, and HVGO Kw. The influence of the H-Oil ® mixed feed Kw on the LVGO, HAGO, and diesel Kw factors decreases with reduction of molecular weight (average boiling point) of these three products ( Figure 2). Figure 2 shows that there is a dependence of the consonance of mixed feed Kw and Kw of H-Oil ® liquid products: diesel, HAGO, LVGO, HVGO, ATB, and VTB on the average boiling point of the liquid products. These data indicate that quality of the H-Oil ® mixed feed affects mostly the quality of the hydrocracked heavy oil products, and the lighter products like diesel are weaker dependent on the H-Oil ® residual feedstock quality, while the naphtha quality is not affected at all from the H-Oil ® feed quality. The lighter products like diesel and naphtha are primary and secondary products and the secondary cracking reactions most probably decrease the dependence of their quality on the original vacuum residue feedstock quality.    The data in Table 3 show that the mixed H-Oil ® feed quality expressed by Kw controls the H-Oil ® VTB properties since it is known that the H-Oil ® VTB density strongly correlates with Concarbon (micro carbon) content [1] and as we will see later in this work it also correlates with softening point and viscosity. Thus, quality of the H-Oil ® VTB will be strongly affected by the Kw of the feed, and from crudes which contain vacuum residue fractions with a lower Kw may be expected during H-Oil ® hydrocracking to be produced VTB with a higher density.
where:   It is evident from the data in Figure 3 that the Kw of the mixed H-Oil ® feed gradually decreases from Case 1 to Case 36. The blended SRVROs Kw for the studied 36 cases varied between 11.90 (Kw of Urals crude oil, the main crude oil for LNB refinery for this study) and 11.22 (Kw of the crude oil blend 41%Urals/34.5%Kirkuk/24.5%El Bouri; Case 32). The lowest Kw of the mixture blended SRVROs-FCC SLO was that of Case 32 and it was 10.97. The lowest Kw of the mixed H-Oil ® feed was that of case 32, and it was 10.07. As apparent from the data in Figure 4 the sum of the FCC SLO and the recycle of PBFO can The Kw of the mixture blended SRVROs-FCC SLO was computed by Equation (10) and the calculated Kw of the blended SRVROs, and the Kw of FCC SLO that varied between 9.6 and 9.8.
It is evident from the data in Figure 3 that the Kw of the mixed H-Oil ® feed gradually decreases from Case 1 to Case 36. The blended SRVROs Kw for the studied 36 cases varied between 11.90 (Kw of Urals crude oil, the main crude oil for LNB refinery for this study) and 11.22 (Kw of the crude oil blend 41%Urals/34.5%Kirkuk/24.5%El Bouri; Case 32). The lowest Kw of the mixture blended SRVROs-FCC SLO was that of Case 32 and it was 10.97. The lowest Kw of the mixed H-Oil ® feed was that of case 32, and it was 10.07. As apparent from the data in Figure 4 the sum of the FCC SLO and the recycle of PBFO can reach 43% of the fresh blended SRVRO feed. Considering that it has a substantially lower Kw (9.7 for FCC SLO, and 10.4 for the PBFO) it becomes clear that its effect on the mixed H-Oil ® feed Kw will be appreciable. By the use of multiple linear regression for the studied 36 cases two equations were obtained relating Kw factors of FCC SLO and PBFO recycle to H-Oil ® mixed feed Kw (Equation (2)), and Equation (3)   Equations (2) and (3) exhibit that for the studied 36 cases the H-Oil ® mixed feed Kw predominantly depends on the shares of FCC SLO and of PBFO recycle. Understandably the FCC SLO has a bigger negative impact on the H-Oil ® mixed feed Kw than that of the recycle because the FCC SLO has a lower Kw than that of the recycle. The influence of the blended SRVROs Kw on the H-Oil ® mixed feed Kw seems to be negligible, because after inclusion of the blended SRVROs Kw in Equation (3)   Equations (2) and (3) exhibit that for the studied 36 cases the H-Oil ® mixed feed Kw predominantly depends on the shares of FCC SLO and of PBFO recycle. Understandably the FCC SLO has a bigger negative impact on the H-Oil ® mixed feed Kw than that of the recycle because the FCC SLO has a lower Kw than that of the recycle. The influence of the blended SRVROs Kw on the H-Oil ® mixed feed Kw seems to be negligible, because after inclusion of the blended SRVROs Kw in Equation (3) the relative average error of Equation (3) is slightly improved in comparison with that of Equation (2) (from 0.80 down to 0.74%).
The relation of the H-Oil ® mixed feed to VTB density can be expressed by Equation (4) VTB D15 = −0.178FeedKw + 3.074 R = 0.992, av. rel. error = 0.3% Interestingly the data in Table 3 also show that the feed Kw statistically meaningful intermediary negatively correlates with the hydrocracking reaction temperature. This at first glance strange correlation can be explained with the fact that the higher Kw vacuum residual oil feeds are lighter, and contain more saturates which negatively impact colloidal stability of the H-Oil ® feed and as a consequence require lower reaction temperature to keep the ATB sediment content within the acceptable limits [1].
In order to evaluate the influence of H-Oil ® unit through-put, hydrocracking reaction temperature, and shares of FCC SLO, and of PBFO recycle in the H-Oil ® mixed feed on HVGO quality expressed by the Kw a multiple linear regression of the data was performed. Equation Equation (5) indicates that HVGO Kw increases with enhancement of throughput, and reduction of reaction temperature, FCC SLO, and PBFO recycle contents in the mixed feed. Increasing H-Oil ® feed rate decreases reaction time, that in turn diminishes the secondary cracking reactions and as a consequence a higher amount of aliphatic hydrocarbons from the HVGO boiling range are preserved, and they are known to have a higher Kw. As temperature increases, the rates of thermal cracking reactions increase more rapidly than the hydrogen addition counterparts [17], that in turn gives HVGO product with a lower amount of preserved aliphatic hydrocarbons leading to a product with a lower Kw. The FCC SLO, and the recycle of PBFO increase the aromaticity of the feedstock and from the more aromatic feedstock during hydrocracking a more aromatic lower Kw HVGO is obtained.
The relation between Kw of HVGO and Kw of LVGO is given by the regression Equation (6).
The relation of Kw of LVGO and Kw of HAGO is presented by the regression Equation (7).
H AGOKw = 1.011LVGOKw R = 0.970, av. rel. error = 0.45% The H-Oil ® diesel quality expressed by its cetane index was found to depend on through-put, reaction temperature, and FCC SLO content in the H-Oil ® mixed feed. This dependence is given in the regression Equation (8).
It is evident from Equation (8) that similarly to the H-Oil ® HVGO (Equation (5)) the H-Oil ® diesel cetane index (CI) increases with enhancement of throughput, and decreasing of reaction temperature, and FCC SLO content in H-Oil ® mixed feed. The dependence of diesel CI on these variables, however, is lower than that of the H-Oil ® HVGO which can be seen from the lower accuracy of the prediction of Equation (8), ten times as low as that of Equation (5). This suggests that other factors not included in Equation (8) can also affect the hydrocracked diesel fraction cetane index. The inclusion of the recycle of PBFO does not improve the accuracy of prediction that suggests that it does not have impact on H-Oil ® diesel cetane index. The diesel fraction is difficult to crack at the hydrocracking conditions, although its secondary hydrocracking is documented in several researches [18][19][20][21]. The fact that the H-Oil ® diesel cetane index decreases with augmentation of reaction temperature and extending of reaction time (feed through-put reduction) suggests that the diesel may undergo secondary cracking reactions which reduce the aliphatic hydrocarbons content in the diesel and increase the aromatics content. The higher aromatics content was found in our earlier study to correlate with a lower cetane index [22].
As mentioned earlier in this research the H-Oil ® VTB density strongly correlates with Concarbon (micro carbon) content. Since the measurement of the viscosity of the H-Oil ® VTB samples featured with high density and high Concarbon content was difficult to perform due to their high melting point solutions with FCC HCO containing 30% FCC HCO with kinematic viscosity of 11.6 mm 2 /s were prepared and their viscosity was measured. An ICrA matrix of the H-Oil ® VTB properties studied in this work density, Concarbon content (CCR), kinematic viscosity of blends 70%VTB/30%FCC HCO, and softening point was prepared and shown in Table 4. As evident from the ICrA matrix in Table 4 all four studied H-Oil ® VTB properties density, Concarbon content (CCR), kinematic viscosity of blends 70%VTB/30%FCC HCO, and softening point statistically meaningful strongly correlate with each other. Figure 5 exhibits graphs of the dependences of density, viscosity, and softening point of H-Oil ® VTB on Concarbon content. These data clearly indicate that viscosity, and softening point of the H-Oil ® VTB exponentially increase with enhancement of Concarbon content and density. The relation of Concarbon content to density for the H-Oil ® VTB and for the straight run vacuum residual oils shown in Figure 5a indicates that for the same value of density the H-Oil ® VTB has a higher Concarbon content. Since the density correlates with the total aromatic structures content, and the Concarbon content correlates with the number of condensed aromatic rings [1] one may conclude that at the same content of aromatic structures the H-Oil ® VTB could contain a higher amount of condensed aromatic rings.  As the H-Oil ® VTB having higher density, and higher Concarbon content possesses a higher softening point and it is more brittle undercutting of HVGO in the vacuum distillation column has been applied to decrease softening point and Fraas breaking point, and increase penetration to use this material as a feed for production of road asphalt [1,6]. In this study instead of undercutting H-Oil ® HVGO we explored the feasibility to improve softening point of the harder H-Oil ® VTB samples by blending them with H-Oil ® . Figure 6 shows that the softening point of the H-Oil ® VTB linearly decreases with augmentation of HVGO content in the blend H-Oil ® VTB-HVGO (Figure 6a), and that the dependence of the slope of decreasing the softening point of the blend VTB-HVGO on the softening point of the pure VTB can be described by a second order polynomial (Figure 6b).  As the H-Oil ® VTB having higher density, and higher Concarbon content a higher softening point and it is more brittle undercutting of HVGO in the vac tillation column has been applied to decrease softening point and Fraas break and increase penetration to use this material as a feed for production of road asp In this study instead of undercutting H-Oil ® HVGO we explored the feasibility to

Materials and Methods
36 different cases of the operation of the LNB H-Oil ® ebullated vacuum resid drocracking (EBVRHC) with crude slate (this is the crude slate processed in LNB ref share (per cent of total fresh vacuum residue feed) of FCC SLO, and of VTB rec shown in Figure 4 were studied. The variation of operating conditions and net conv for the studied 36 cases is summarized in Table 5. A simplified diagram of the L Oil ® hydrocracker where the investigations were performed is presented in Figu commercial supported Ni-Mo catalyst was employed throughout the study and fo of the cases a nano-dispersed catalyst was also used.

Materials and Methods
36 different cases of the operation of the LNB H-Oil ® ebullated vacuum residue hydrocracking (EBVRHC) with crude slate (this is the crude slate processed in LNB refinery), share (per cent of total fresh vacuum residue feed) of FCC SLO, and of VTB recycle as shown in Figure 4 were studied. The variation of operating conditions and net conversion for the studied 36 cases is summarized in Table 5. A simplified diagram of the LNB H-Oil ® hydrocracker where the investigations were performed is presented in Figure 7. A commercial supported Ni-Mo catalyst was employed throughout the study and for some of the cases a nano-dispersed catalyst was also used.     The net vacuum residue 540 • C+ conversion was estimated by the equation: where: EBRHCFeed540 • C+ = mass flow rate of the EBVRHC feed fraction boiling above 540 • C, determined by high temperature simulated distillation, method ASTM D 7169 of the feed and multiplied by the mass flow rate of the feed; EBRHCProduct540 • C+ = mass flow rate of the EBVRHC product fraction boiling above 540 • C, determined by high temperature simulated distillation, method ASTM D 7169 of the liquid product multiplied by the flow rate of the liquid product.
The methods used to characterize the mixed H-Oil ® feed, and the liquid products: naphtha, diesel, HAGO, LVGO, HVGO, VTB, ATB are summarized in Table 6. The Kw [9] was estimated based on information about density and distillation characteristics by the use of Equation (10) [11].
The molecular weight of the studied H-Oil ® mixed feed, HAGO, LVGO, HVGO, ATB, and VTB was estimated by the correlation of Goosens [23] (Equation (13)): The correlation developed by Abutaqiya [24] was employed to estimate the average aromatic ring numbers in the average hydrocarbon structure of the investigated EBVRHC heavy oils, designated as ARI. ARI is estimated by Equations (14) and (15).
where: MW = molecular weight of EBVRHC heavy oils, g/mol; FRI = function of refractive index where, n D 20 = refractive index at 20 • C. The refractive index was estimated from density at 15 • C by the correlation developed in our earlier research [25] and shown in Equation (16).

Conclusions
135 parameters including H-Oil ® operating conditions and H-Oil ® feed and liquid product properties were evaluated by the use of Intercriteria analysis. It was found that the crude oils containing vacuum residue fractions with a lower Kw factor during ebullated bed hydrocracking produce hydrocracked vacuum residue with a higher density, higher Concarbon content, higher viscosity, and higher softening point. The addition of FCC slurry oil and recycle of partially blended fuel oil to the straight run vacuum residual oils decreases the H-Oil ® mixed feed Kw that in turn leads to production of higher density hydrocracked vacuum residue, lower Kw gas oils, and lower cetane index diesel. The augmentation of H-Oil ® reaction temperature enhances density and decreases Kw of VTB, and H-Oil ® gas oils, and reduces the cetane index of diesel. The magnification of through-put amplifies the H-Oil gas oil Kw and diesel cetane index. All investigated factors controlling the properties of the liquid H-Oil ® products: hydrocracked vacuum residue, hydrocracked gas oils, and hydrocracked diesel were found to have no impact on the properties of hydrocracked naphtha.
The developed in this work correlations can be used to evaluate the influence of crude oil properties, H-Oil ® operating conditions, and the processing of FCC slurry oil, and recycle of partially blended fuel oil on the quality of the H-Oil ® products: diesel, HAGO, LVGO, HVGO, and VTB. This information can be used to assess the impact H-Oil feed properties and operating conditions on the performance of the other refinery units which process the H-Oil ® products mentioned above and to find the parameters which provide the optimal performance of the whole refinery.

Conflicts of Interest:
The authors declare no conflict of interest.