Chong Peng, Zhiming Zhou, Xiangchen Fang, Hualin Wang,*

1 Panjin Institute of Industrial Technology, Dalian University of Technology, Panjin 116045, China

2 East China University of Science and Technology, Shanghai 200237, China

3 Dalian Research Institute of Petroleum and Petrochemicals, SINOPEC, Dalian 116045, China

Keywords:Thermodynamics and kinetics Naphthalene hydrogenation Monoaromatics Operating condition Kinetics model

A B S T R A C T Hydrocracking represents an important process in modern petroleum refining industry, whose performance mainly relies on the identity of catalyst. In this work, we perform a combined thermodynamics and kinetics study on the hydrogenation of naphthalene over a commercialized NiMo/HY catalyst. The reaction network is constructed for the respective production of decalin and methylindane via the intermediate product of tetralin, which could further undergo hydrogenation to butylbenzene, ethylbenzene,xylene, toluene, benzene, methylcyclohexane and cyclohexane. The thermodynamics analysis suggests the optimum operating conditions for the production of monoaromatics are 400 °C, 8.0 MPa, and 4.0 hydrogen/naphthalene ratio.Based on these,the influences of reaction temperature,pressure,hydrogen/-naphthalene ratio, and liquid hourly space velocity (LHSV) are investigated to fit the Langmuir-Hinshelwood model. It is found that the higher temperature and pressure while lower LHSV favors monoaromatics production, which is insensitive to the hydrogen/naphthalene ratio. Furthermore, the high consistence between the experimental and simulated data further validates the as-obtained kinetics model on the prediction of catalytic performance over this kind of catalyst.

1. Introduction

Given the increasingly serious environmental problems, the demand for improving the quality of fuel including gasoline and diesel has aroused world-wide interest in recent decades [1-4].To fulfill the stringent fuel quality standards, hydrotreating and hydrocracking process have been widely applied as an important and necessary stage in the oil refining, not only reducing the contents of heteroatoms-containing compounds, such as sulfur and nitrogen,but also decreasing the density and increasing the cetane number [5-10]. During the hydrotreating process, the hydrogenation to saturate aromatic compounds emerges as an essential step to alleviate the steric hindrance of substituted dibenzothiophenes for the subsequent hydrodesulphurization [11,12]. Moreover, the saturation of aromatic compounds is recognized as a prerequisite for the removal of nitrogen [13,14]. As a result, the design and development of efficient catalysts for aromatics hydrogenation remain highly desirable in hydrotreating process, among which naphthalene as a typical aromatic compound has gained considerable attention for its hydrogenation into decalin [15-17].

So far, various kinds of catalysts have been investigated for naphthalene hydrogenation,which could be categorized into noble metal catalyst,such as Pt[18,19],Pd[20,21]and Rh[22],as well as non-noble metal catalyst, such as Ni [23,24], Co [25], W [26,27],and Mo[28,29].Generally,the noble metal catalysts deliver strong capability in the activation and dissociation of hydrogen, giving rise to much higher catalytic activity and selectivity toward naphthalene hydrogenation.However,the high cost,limited abundance,as well as low resistance to poisons of noble metals restrict their real application in industry [30-32]. Compared with noble metal catalysts, the non-noble metal ones exhibit many unique advantages,such as cost-effectiveness,high stability under high temperature and pressure, despite of their moderate activity for this reaction [33-36]. Hence, many efforts have been devoted to improve their catalytic efficiency by the selection of optimal catalyst supports and the introduction of promoters, and a significant increase in the naphthalene conversion of 95.6%and tetralin selectivity of 99.7% was also demonstrated over a 4%NiO-20%MoO3/Al2O3catalyst at 200 °C, 8 h and 6 MPa in the recent work [37].Notably, a systemic understanding of the thermodynamics and kinetics over such kind of bimetallic Ni-Mo is still lacking to establish the reaction network and obtain the relevant thermodynamics and kinetics parameters, and thus guide its further optimization under the industrial conditions.

In this work,a combined thermodynamics and kinetics study of naphthalene hydrogenation over the commercial NiMo/HY catalyst was conducted. A preliminary investigation on the influences of reaction temperature was conducted to establish the reaction network,which further yield the equilibrium constant and conversion for each step. Based on which, the influences of operation conditions including temperature, pressure, as well as hydrogen/naphthalene ratio on the equilibrium conversion and product selectivity were tested. Correspondingly, the kinetics experiments were further conducted to verify the above predictions by thermodynamics calculations under the exclusion of internal and external diffusion effects. As a result, a Langmuir-Hinshelwood model was applied to fit the above kinetics data,which gave good consistence between the experimental and calculated concentrations of product species in naphthalene hydrogenation over the NiMo/HY catalyst.

2. Experimental

The hydrogenation of naphthalene was carried out in the laboratory set up as shown in Fig. 1. Typically, the NiMo/HY catalyst(Granular, FC-24, Sinopec Dalian (Fushun) Research Institute of Petroleum and Petrochemicals) with specific surface area larger than 350 m2·g-1and pore volume larger than 0.28 ml·g-1was grinded and screenedviaan ASTM standard sieve to an average diameter of 0.22 millimeter.A weighed catalyst of 10 g was loaded in the constant temperature zone of the fixed bed reactor,and the top and bottom of the catalyst bed were filled with quartz sand.Then the gas in the reactor was replaced with H2(99.999% purity)and pressured to 5.0 MPa.The catalyst wasin situpresulfided using a 3.0%(mass)CS2(99.9%,Aladdin)/n-dodecane(99%,Aladdin)solution at a flow volume ratio of solution to hydrogen as 1:300. The reaction temperature was raised from room temperature to 230°C with a heating rate of 10°C·min-1and kept at this temperature for another 2 h,following by heating to 370 °C with another 3 h.After cooling down to the reaction temperature,the presulfiding solution was switched to reactant of 5% (mass) naphthalene(99%, Aladdin)/n-dodecane (99%, Aladdin) solution to initialize the reaction. The gas products were analyzed by an Agilent 7890 gas chromatograph equipped with one Porapack Q,one 5A zeolite,and one HP-PLOT Al2O3capillary columns and two thermal conductivity (TCD) and one flame ionization detectors (FID). On the other hand, the liquid products were analyzed by an Agilent 6890 plus gas chromatograph interfaced with an Agilent 5975C MSD.

Fig. 1. Schematic diagram of the experimental setup.

The conversion of naphthalene was calculated by:

whereF0Nand FNrepresent the molar flow rate of naphthalene in the inlet and outlet liquid phase,respectively.Moreover,the selectivity to productiwas calculated by:

whereFirepresents the molar flow rate of productiin the outlet gas and liquid phases.

3. Results and Discussion

3.1. Reaction network of naphthalene hydrogenation

It was found that the gas product (mainly C2-C4alkanes)accounted for less than15%(mass)of the total product throughout this work in the temperature range of 280-330 °C. In comparison,the liquid product of naphthalene hydrogenation mainly involved tetralin, decalin, methylindene, butylbenzene, ethylbenzene,xylene, toluene, benzene, methylcyclohexane, and cyclohexane,which was similar to that of tetralin hydrogenation. Moreover,the hydrogenation of solventn-dodecane was also investigated with much lower conversion(＜5%)to produce chain-alkanes rather than cycloalkanes and arene, thus giving limited influences on the hydrogenation of naphthalene.

Fig. 2 gives the influences of reaction temperature on the conversion of naphthalene and the selectivity of various products in the range of 280-330°C.It can be seen in Fig.2(a)that naphthalene exhibits an almost complete conversion during the temperature range, as a result of its low content in feedstock. Meantime, the selectivity of tetralin gradually decreases from 80% to 25%,consistent with its further conversion into other deep hydrogenation products.Correspondingly,the selectivity of decalin and methylindene as shown in Fig.2(b)firstly increase followed by a continuous decrease, indicating them as the intermediate products. As shown in Fig. 2(c), the selectivity of butylbenzene also exhibits a maximum value around 300°C,while that of the other products including ethylbenzene, xylene, toluene, and benzene continuously increases with the temperature. This suggests the favorable dealkylation of butylbenzene into these products under high temperature.Moreover,Fig.2(d)shows the similar trends of methylcyclohexane and cyclohexane, which could be reasonably treated as the final products. From the point view of thermodynamics, the productions of methylcyclohexane and cyclohexane could be either from the ring-opening and dealkylation reaction of decalin or the hydrogenation of toluene and benzene. In order to justify this, the hydrogenation of toluene and benzene over the NiMo/HY catalyst under the same reaction conditions was also investigated, which exhibited almost no conversion. In other words, this suggests that methylcyclohexane and cyclohexane arise from the ring-opening and dealkylation of decalin.

Based on the above discussion, the reaction network for naphthalene hydrogenation could be summarized in Fig. 3. Typically,the hydrogenation of naphthalene firstly gives the production of tetralin, which could be further converted into decalin and methylindan. The deep hydrogenation of methylindan gives rise to the production of butylbenzene, which could further dealkylate into ethylbenzene, xylene, toluene, and benzene. On the other hand, the ring-opening and dealkylation reaction of decalin could further produce methylcyclohexane and cyclohexane.

3.2. Thermodynamics analysis of naphthalene hydrogenation

Herein,Table 1 lists the parameters for the calculation of Gibbs free energy of formation of various species with temperature.As a result, the Gibbs free energy (ΔrG?T) of formation could be calculated based on the reaction expression as shown in Fig. 3, which further yield the reaction equilibrium constant (Kp) as a function of reaction temperature Fig.4.It can be seen that all these equilibrium constants decrease with the reaction temperature. Specifically, the equilibrium constant follows the order ofKp2＜Kp1＜Kp4＜1 ＜Kp9＜Kp3＜10.This suggest that the hydrogenation, isomerization, and ring-opening of naphthalene are highly affected by the reaction equilibrium, which could be treated as reversible reactions. On the contrary, the equilibrium constants(Kp5,Kp6,Kp7,Kp8,Kp10andKp11)of dealkylation reaction are much higher than 100, which could be treated as irreversible reactions.

Table 1 Correlation of Gibbs energy (kJ·mol-1) of formation of various species with temperature①

Fig. 5 exhibits the influences of reaction temperature on the molar fraction of reaction species,the selectivity of product,as well as unit hydrogen consumption for naphthalene hydrogenation. It can be seen that naphthalene could be almost completely converted into others at equilibrium, and the product is mainly composed of ethylbenzene, xylene, toluene, benzene,methylcyclohexane, and cyclohexane. Moreover, the increase of reaction temperature gives rise to the increased selectivity of monoaromatics and compensated by decreased selectivity of naphthene,consistent with the trend as shown in Fig.2.As a result,the generation of more monoaromatics rather than naphthene would consume less hydrogen from 0.078 to 0.063 with the temperature rising from 250 °C to 450 °C.

Fig. 2. Effect of reaction temperature on (a) conversion of naphthalene, and (b-d) selectivity of various species in the liquid phase during naphthalene hydrocracking.(Condition: Pt = 5 MPa, LHSV = 2 h-1, hydrogen/naphthalene ratio = 30).

Similarly,Fig.6 gives the influences of reaction pressure on the molar fraction of reaction species,the selectivity of product,as well as unit hydrogen consumption for naphthalene hydrogenation.Because the naphthalene hydrogenation is a molecular number reduction reaction and the production of naphthene consumes more hydrogen than that of monoaromatics, the selectivity of naphthene increases with the reaction pressure while that of monoaromatics decreases as shown in Fig. 6(b). Correspondingly,the unit hydrogen consumption also increases with the reaction pressure as depicted in Fig. 6(c).

Fig. 3. Reaction network of naphthalene hydrogenation over NiMo/HY.

Fig. 4. Variation of equilibrium constant with temperature in naphthalene hydrogenation.

Furthermore, Fig. 7 shows the influences of hydrogen/naphthalene ratio on the molar fraction of reaction species,the selectivity of product, as well as unit hydrogen consumption for naphthalene hydrogenation. It can be seen that with the increase of hydrogen/naphthalene ratio to 4, the molar fraction of naphthalene sharply decreases to almost zero, indicating the complete conversion of naphthalene. Correspondingly, the fraction of monoaromatics reaches the maximum at the hydrogen/naphthalene ratio of 4, while that of naphthene is almost zero. After that, the selectivity of monoaromatics decrease with the hydrogen/naphthalene ratio, compensated by the increased selectivity of naphthene. Moreover, the unit hydrogen consumption remains 0.06gH2/gNAin the hydrogen/naphthalene ratio range of 1-4,which further increases with that as shown in Fig. 7(c). Based on the above trends, it can be concluded the favorable production of monoaromatics when the hydrogen/naphthalene ratio less than 4, which competes with that of naphthene when the hydrogen/-naphthalene ratio larger than 4.

Fig. 6. Variation of (a) molar fraction of various species, (b) product selectivity and (c) unit hydrogen consumption with total pressure in naphthalene hydrocracking.(Condition: T = 350 °C, hydrogen/naphthalene ratio = 5).

Fig. 7. Variation of (a) molar fraction of various species, (b) product selectivity and (c) unit hydrogen consumption with initial hydrogen/naphthalene ratio in naphthalene hydrogenation. (Condition: T = 350 °C, Pt = 5 MPa).

Based on the above thermodynamics analysis for naphthalene hydrogenation to produce monoaromatics, the optimal operation conditions could be derived as: (i). reaction temperature around 400 °C. Although further increasing reaction temperature could slightly increase the yield of monoaromatics, it would consume much more energy for heating and cause the formation of coke over catalyst surface to deactivate catalyst; (ii) reaction pressure around 8-10 MPa.Although the lower pressure favors the production of monoaromatics, it would result in poor solubility of H2in solvent for naphthalene hydrogenation.Moreover,further increasing pressure has limited influences on the yield of monoaromatics while consumes more hydrogen; (iii) hydrogen/naphthalene ratio around 4, which gives the highest yield of monoaromatics. It is worth to note that the hydrogen/naphthalene ratio of 4 refers to that in liquid phase, and thus the hydrogen in gas phase should be in a large excess.

Fig.5. Variation of(a)molar fraction of various species,(b)product selectivity and(c)unit hydrogen consumption with reaction temperature in naphthalene hydrogenation(Condition: Pt = 5 MPa, hydrogen/naphthalene ratio = 5).

3.3. Kinetics analysis of naphthalene hydrogenation

To begin with, the influences of both internal diffusion and external diffusion were investigated. Firstly, five kinds of catalysts with different particles sizes,i.e., 1.42, 0.63, 0.34, 0.22, and 0.17 mm, were prepared and tested for this reaction under the same reaction conditions.As shown in Fig.8(a),the yield of tetralin increases with the decrease of catalyst particle size to 0.34 mm.Afterwards,the yield of tetralin remains almost unchanged despite of the decreased catalyst particle size. Hence,the effect of internal diffusion could be eliminated when catalyst particle size smaller than 0.34 mm. On the other hand, the influence of space velocity was also investigated by varying the amount of catalyst,i.e., 5 and 10 g. As shown in Fig. 8(b), the yields of tetralin for these two amounts of catalysts are almost identical,indicating the elimination of external diffusion. Hence, the catalyst with an average size of 0.22 mm and loading of 10 g was chosen for the kinetics study in this work.

The influences of reaction temperature,pressure,LHSV,and H2-/naphthalene on the products are shown in Fig.9.It can be seen in Fig. 9(a) that the yield of tetralin continuously decreases with the increase of reaction temperature, while those of the final product 1-Ar including benzene, toluene, xylene, ethylbenzene and butylbenzene, as well as cycC6 such as methylcyclohexane and cyclohexane increase. Moreover, the increase of reaction pressure from 4-6 MPa in Fig.9(b)would promote the conversion of tetralin into other products.The cycC6 rather than 1-Ar product were also found to be the main product, consistent with the above thermodynamics analysis. Fig. 9(c) suggests that the increase of LHSV would inhibit the conversion of tetralin into either 1-Ar or cycC6.Furthermore, the productions of tetralin, 1-Ar and cycC6 were found to be insensitive to the H2/naphthalene ratio, possibly due the excess of H2in the feedstock.

As well known that, the naphthalene hydrogenation catalysts usually involve two kinds of active site:one is the metal active site for the hydrogenation and dehydrogenation, and the other one is the acid site for the isomerization, ring-opening, and dealkylation.Hence, the reaction mechanism mainly involves the adsorption of naphthalene and hydrogen, consecutive hydrogen addition steps and desorption of products, thus a Langmuir-Hinshelwood model was applied to fit the experimental data for a description of this reaction. The reaction network in Fig. 3 could be simplified as shown in Fig. 10, where m and a represent the metal active site and acid site, respectively. Based on this, the reaction rate expressions for the hydrogenation of naphthalene into tetralin and then decalin could be written as:

Fig. 8. Effects of (a) catalyst particle size and (b) LHSV on the yield of tetralin in naphthalene hydrogenation.

Fig. 9. Effects of (a) temperature, (b) pressure, (c) LHSV, (d) H2/naphthalene ratio on the yield of product in naphthalene hydrogenation.

Fig. 10. Simplified reaction network of naphthalene hydrogenation over NiMo/HY.

Fig. 11. Comparison of experimental and calculated concentrations of product species in naphthalene hydrogenation over NiMo/HY.

wherekm,1andkm,2represent the reaction rate constant for the hydrogenation of naphthalene into tetralin,and that of tetralin into decalin, respectively.Keq,1andKeq,2are the corresponding equilibrium constants.Kadm,jrepresents the adsorption constant of speciesjon metal active site.

On the other hand,the reaction rate expressions of tetralin into methylindene, decalin into cycC6, methylindene into butylbenzene, and butylbenzene into BTXE could be written as:

whereka,1,ka,2,ka,3andka,4represent the reaction rate constants of tetralin into methylindene, decalin into cycC6, methylindene into butylbenzene, and butylbenzene into BTXE, respectively.Keq,3andKeq,4represent the reaction equilibrium constants of tetralin into methylindene, and methylindene into butylbenzene, respectively.Kada,jrepresents the adsorption constant of speciesjon acid site.Moreover, the reaction rate constant could be written based on the Arrhenius equation:

wherek0m,jandk0a,jrepresent the pre-exponential factor of speciesjon metal active site and acid site,respectively,andEam,jandEaa,jrepresent the corresponding activation energy.Similarly,the adsorption equilibrium constant could be written based on the quantitative structure- reactivity relationship (QSRR):

whereNAR,jandNSC,jrepresent the number of aromatic ring and saturated carbon within speciesj, respectively.

Furthermore, the mass balance within the reactor gives rise to the following equations:

whereFl,Wandc1,inrepresent the volume flow rate,the amount of catalyst, the concentration of naphthalene in feedstock, respectively. Substituting Eqs. (1)-(6) into Eqs. (11)-(17) gives rise to the kinetics model,which could be further fitted to the experimental data. As a result, Table 2 lists the estimated kinetic parameters for naphthalene hydrogenation over the NiMo/HY catalyst. It can be seen that the activation energy for tetralin hydrogenation to decalin (76.1 kJ·mol-1) is almost two times higher than that of naphthalene hydrogenation to tetralin (37.4 kJ·mol-1), indicating the favorable naphthalene hydrogenation with respect to tetralin hydrogenation. Moreover, the reaction rate constant for naphthalene hydrogenation to tetralin is much higher than that of tetralin hydrogenation to decalin, consistent with previous studies by Girgis and Gates [38].

Table 2 Estimated kinetic parameters for naphthalene hydrogenation over NiMo/HY

Table 3 Adsorption constants of various species on metal and acid sites of NiMo/HY

Table 3 summarizes the adsorption constants of various species on metal active site and acid sites of the NiMo/HY catalyst.It can be seen that the adsorption equilibrium constant on acid site is much higher that on the corresponding metal active site,consistent with the strong acidity and relatively poor hydrogenation activity for industrial catalyst. Fig. 11 suggests the limited deviation less than 4.8% between the measured molar fraction and the calculated molar fraction based on the kinetics data in Fig. 9, further validating the as-obtained kinetics model in the description of naphthalene hydrogenation over the NiMo/HY catalyst.

4. Conclusions

In this work,a combined thermodynamics and kinetics study on naphthalene hydrogenation is performed over a NiMo/HY catalyst.The reaction network is constructed for the separate production of monoaromatics and naphthene via the intermediate product of tetralin. Based on these, the optimum operating conditions for the productions of monoaromatics by thermodynamics calculations are suggested to be 400°C,8.0 MPa,and 4.0 hydrogen/naphthalene ratio.The influences of reaction temperature,pressure,hydrogen/-naphthalene ratio, and liquid hourly space velocity (LHSV) are investigated to fit the Langmuir-Hinshelwood model, which fits well with the experimental data.The insights revealed here might pave the way for rational design of catalytic hydrogenation of naphthalene.

Acknowledgements

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (91934301), The China Postdoctoral Science Foundation (2019M661409 and 2020T130190), Doctoral Start-up Foundation of Liaoning Province (2019-BS-054), Liaoning Revitalization Talents Program (XLYC1807245), The Open Project of State Key Laboratory of Chemical Engineering (SKL-ChE-18C04), and Dalian High-Level Talent Innovation Program(2017RQ085).