The solubility of n-decane in ethylene and its effect on the oligomerization of supercritical ethylene over heterogeneous catalysts

Highlights

• The solubility of n-decane in subcritical and supercritical ethylene is reported.

• A rigorous thermodynamic model was used to model the experimental solubility data.

• Decane is used as a model compound for coke for the ethylene oligomerization.

• Decane solubility increases greatly beyond the ethylene critical point.

• High reactor temperature causes coke formation even under supercritical conditions.

Abstract

We investigated the effect of coke solubility in supercritical ethylene on the ethylene oligomerization over nickel-based heterogeneous catalysts. The approach uses n-decane as a model compound for coke to simulate coke formed during the catalytic process. We report the solubility of n-decane in ethylene at 30, 50, and 75 ^oC and pressures ranging from 1 to 68 bar; conditions previously screened by our research group for ethylene oligomerization. We reproduced previous literature data on the solubility of n-decane in nitrogen to validate the flow system designed in the present work. The present study is the first to report the solubility of n-decane in ethylene under a broad range of conditions, including subcritical and supercritical conditions. We found that the ethylene – n-decane system deviates from ideality very quickly with increasing pressure. Beyond the ethylene critical point (P = 50.3 bar and T = 9.4 ^oC) the solubility (expressed in terms of mole fraction) of n-decane in ethylene at 30 ^oC reaches a maximum value of 3.0%; which is close to the value observed at 50 and 75 ^oC, under the same pressure. We used the solubility data to estimate the rate of coke dissolution and compared these values with kinetic data previously reported by our group under supercritical conditions. Analysis of coke dissolution rates in supercritical ethylene under different temperatures indicates that the transport of products from the catalyst surface to the bulk of supercritical ethylene controls the catalytic process at supercritical conditions. The approach used in the present work is novel because it accounts for the contribution of coke dissolution rates on the ethylene oligomerization over nickel-based solid catalysts..

Introduction

Ethylene oligomerization is one of the foremost processes used in industry to convert widely available and cheap ethylene into a wide range of crucial chemicals, including surfactants, lubricants, adhesives, gasoline, jet fuel, among many others. This process is currently carried out over homogeneous catalysts [1], [2], requiring the use of co-catalysts, such as methylaluminoxane (MAO) [3], [4], [5]. MAO is a pyrophoric solid and is commonly dissolved in aromatic solvents, such as toluene. The use of toluene as the reaction medium requires expensive separation processes to recover the catalyst and reaction products and has potential safety and environmental implications because of toluene’s toxicity. For these reasons, ethylene oligomerization over heterogeneous catalysts has received attention due to the potential of employing a more environmentally friendly and user-friendly process for the production of light and heavy alkenes [1], [5]. Several works reported the development of novel heterogeneous catalysts for ethylene oligomerization comprising mostly nickel-exchanged aluminosilicates, such as Ni-H-Beta [4], [6], [7], [8], [9], [10], [11], [12], [13], [14], Ni-MCM-41 [7], [15], Ni-SBA-15 [10], [16], [17], [18], and Ni-SIRAL30 [7], [19]. Although these catalysts show good stability and high ethylene conversion, the high selectivity to light alkenes, such as butene and hexene, still hinders the use of heterogeneous catalysts in large-scale ethylene oligomerization processes relative to its homogeneous counterpart.

Catalyst deactivation is another concern with heterogeneous catalysts [20]. Deactivation of heterogeneous catalysts can occur by diverse processes [21]. Usual mechanisms include fouling and poisoning [20], [22], [23]. Fouling is mechanical in nature and results from the deposition of material on the catalyst. Poisoning is a chemical deactivation process caused by strongly adsorbed species on active sites [20], [22], [23]. A common cause of deactivation is the accumulation of coke on the catalyst. Catalyst deactivation requires expensive and complicated regeneration processes [22], which are estimated to cost the industry billions of dollars annually [23].

It has been proposed that supercritical fluids could promote the solvation of coke in heterogeneous oligomerization processes, which in turn reduces catalyst deactivation by coke [9], [10], [24], [25]. A supercritical fluid is any fluid above its critical temperature and pressure, and ethylene has the lowest critical temperature among light alkenes (9.4 ^oC), and a relatively mild critical pressure (50.3 bar). Added to the fact that ethylene is the reactant for the ethylene oligomerization, its mild critical point could make supercritical ethylene a medium for coke mitigation. In previous publications [9], [10], we provided visual evidence for the dissolution of coke in supercritical ethylene during the ethylene oligomerization over the Ni-H-Beta [9]. Our results suggested an increase in the formation of C₁₀₊ alkenes as a consequence of the dissolution of coke molecules in supercritical ethylene and the increased rates of oligomerization under supercritical conditions. In addition to the solvation effect, the oligomerization of supercritical ethylene could shift the product distribution towards high molecular weight alkenes. This is especially desirable to increase the production of liquid hydrocarbons from ethylene oligomerization over heterogeneous catalysts.

While supercritical ethylene can be used under mild conditions to promote coke dissolution during the ethylene oligomerization over heterogeneous catalysts, mild conditions can limit the production of alkenes due to the low extent of chain-growth reactions [26], [27]. Several works reported the increase in ethylene oligomerization rates with increasing temperature and pressure[6], [7], [8], [9], [11], [12], [13], [17], [28], [29]. Tuning the conditions to minimize coke production and simultaneously maximize the extent of chain-growth reactions is necessary to ensure stable catalyst activity and high alkene productivity.

The extent of coke production depends on the rates of coke deposition and coke removal (via dissolution). Information currently available on the quantification of coke deposition and dissolution rates over heterogeneous catalysts is limited. In the present work, we used a systematic approach to quantify the dissolution rates of coke in subcritical and supercritical ethylene. This approach required a solubility study using a coke model compound in subcritical and supercritical ethylene. In previous works, coke produced via subcritical ethylene oligomerization over nickel-exchanged aluminosilicate catalysts has been characterized as aliphatic molecules [7], [9]. Additionally, we characterized the coke formed under supercritical conditions as high molecular weight linear and cyclic alkanes connected via sp³ bonds [9]. We found that the CH₂/CH₃ ratio of the coke formed under supercritical conditions was a function of temperature with values varying between 3.0 and 6.5. Therefore, we chose n-decane as a coke model compound because of its aliphatic nature and its CH₂/CH₃ ratio, which falls within the range of values reported in our previous study (3.0–6.5) [9].

The solubility of liquids, such as n-decane, in compressed gases, has been extensively reported in the literature [30], [31], [32], [33], [34], [35], [36]. Prausnitz et al. studied the solubility of iso-octane, n-decane, carbon tetrachloride, and toluene in nitrogen, hydrogen, and carbon monoxide [32]. They concluded that the solubility deviates from ideality very quickly with increasing pressure, especially for pressures above 10 atm. Diepen et al. studied the solubility of naphthalene in subcritical and supercritical ethylene and reported a sharp increase in solubility beyond the ethylene critical point, which can be attributed to the reduction in the ethylene dielectric constant above the critical point leading to a decrease in the polarity of the supercritical fluid [33]. Kurnik et al. reported the same phenomena for the solubility of 2,3-dimethyl-naphthalene, 2,6-dimethyl-naphthalene, phenanthrene, benzoic acid, and hexachloroethane in carbon dioxide and ethylene [34]. The present work focused on the solubility of n-decane in ethylene because of the potential to minimize coking during the ethylene oligomerization over nickel-based solid catalysts.

The overall objective of this research is to assess and model n-decane solubility in subcritical and supercritical ethylene. The information obtained in the solubility study will be used to provide insight into the solvation of the coke formed over the solid catalyst during the ethylene oligomerization under supercritical conditions, thereby assessing its contribution to the transport of high molecular weight products from the catalyst surface to the gas phase. This investigation is an essential first step in an ongoing broader study on the kinetics of ethylene oligomerization under supercritical conditions.