The present study investigates the oxidative dehydrogenation of propane-butane (C3–C4) fraction over mono (Cr or Mo) and bi-metal (Cr-Mo) loaded SiO2 catalysts. The catalysts were prepared by sequential impregnation method at 500oC calcination temperature. Experiments were performed by feeding C3–C4 fraction, oxygen, nitrogen, and steam into a continuous flow quartz reactor at an atmospheric pressure (P = 1 atm.), reaction temperatures between 500 – 650oC, gas hourly space velocity (GHSV) within 100 – 400 h-1, and at reaction time (t r) = 2h. Appropriate water vapor addition to the feed sinificantly minimizes oxidation into coke deposits on the catalyst surface, and also prevents further olefin conversion into undesirable product. The physicochemical properties were evaluated by BET, XRD, IR, and EPR characterization techniques. The major oxidation products are ethylene, propylene, isobutylene, butylene. This paper reports that the total yield of olefins (Ʃ C2-C4) = 66.0 % was achieved at 83.5 % conversion level of C3–C4 at 630oC. The results indicate that the addition of Mo to catalysts of Cr/SiO2 modifies its catalytic activity for the ODH reaction. Mono-metallic catalysts (Cr/SiO2 and Mo/SiO2) were prepared for comparison purposes.
Keywords: oxidative de-hydrogenation of C3–C4, C3–C4 conversion, selectivity to olefin, olefin yield, olefin production.
Olefins are the most important starting materials for modern industrial organic chemistry. Over the past several decades, the manufacturing capacities and importance of low molecular-weight olefin (ethylene, propylene, isobutylene, and butylene) have undergone remarkable transitions, as well as continue to serve as a fundamental basis for the petrochemical and refining industries. The catalytic de-hydrogenation of alkanes has a considerable industrial impact because it represents a route to obtain olefins from feed-stocks of low-cost saturated hydrocarbons. Traditional technologies for olefins production involve catalytic dehydrogenation of alkanes via either the steam cracking or fluid catalytic cracking method. Dehydrogenation reactions appear very simple: their thermodynamic and kinetic characteristics have, nevertheless, contributed to make the development of technologies that allow for a reliable and efficient industrial application, rather complex i.e. the overall efficiency of olefin production by the catalytic dehydrogenation method is greatly impeded by a few drawbacks such as: (i) thermodynamic limitations, owing to the highly endothermic and the substantial energy requirement to activate the reactant molecules; (ii) coke deposits build up on the reactor wall thus, lowering heat transfer, increasing pressure and corrosion. Consequently, the existing commercial processes for olefins production are very energy-intensive, exhaustive, and not cost-effective. While these two routes are very well developed, increasing the capacity of these processes is only possible to some extent, as changing regulation limits the use of by-products (notably aromatic molecules) in fuels. For these reasons, the present industrial capacity for C2-C4 olefins production via these traditional processes is expected to be insufficient, and therefore, cannot meet the fast-growing demand of olefins in the international market. Since producers seek to leverage their existing assets and the available internal streams to find an optimum solution for meeting the demands of olefins, oxidative de-hydrogenation, which involves coupling the reagent mixture with oxidant such as oxygen [Eq. (1) ‒ (2)], has been widely studied, as a potentially attractive route to circumvent the thermodynamic limitations, eliminate coking, and therefore, extend catalyst lifetime.
C3H8 + 1/2O2 → C3H6 + H2O ∆H f 0 = ‒ 95.5 kJ/mol (1) C4H10 + 1/2O2 → C4H8 + H2O ∆H f 0 = ‒ 95.5 kJ/mol (2) There are, however, a number of current challenges preventing oxidative de-hydrogenation from being widely implemented. The difficulties inherent in oxidative dehydrogenation reactions revolve around selectivity control because all equivalent C‒H bonds have an equal bonding energy, and therefore an equal chance of reacting. When two C‒H bonds of neighboring carbons are split, a double bond is formed and alkanes are converted to alkenes. Thereby, oxygen addition to alkane feeds exposes the synthesized olefins to further oxidation conditions that results into the formation of environmentally-damaging and economically-useless carbon oxides (CO and CO2), consequently decreasing the yield of alkenes [Eq. (3) ‒ (4)]. C3H6 + 3O2 → 3CO + 3H2O ∆H f 0 = ‒ 219 kJ/mol (3a) C3H6 + 9/2O2 → 3CO2 + 3H2O ∆H f 0 = ‒ 1867.5 kJ/mol (3b) C4H8 + 4O2 → 4CO + 4H2O ∆H f 0 = ‒ 518 kJ/mol (4a) C4H8 + 6O2 → 4CO2 + 4H2O ∆H f 0 = ‒ 2716 kJ/mol (4b) Therefore, the design of effective catalytic systems that are sufficiently active, exhibit high selectivity, be periodically-regenerated under severe conditions, and yet operate at temperatures that minimize oxygenation of the desired products, are key performance demands for cost-effective production of olefins. Nevertheless, the mechanism of carbon filament formation and catalyst de-activation resulting from the decomposition of hydrocarbons on catalyst metal particles has been extensively studied in the past.[5 – 11] Thereby, if the cracking of lower alkanes is to be utilized for olefin production in a continuous process, the addition of steam (as a diluent) at 550oC has been demonstrated to restore the catalytic activity even after complete catalyst deactivation, and enhance higher equilibrium conversion. During the past decade, the design and synthesis of bi-metallic catalysts have attracted considerable attention because they show multiple functionalities and prominent catalytic activity, selectivity, and stability over monometallic catalyst[12 ‒ 16] i.e. bi-metallic catalytic systems can achieve chemical transformations that are unprecedented with because different components of the catalyst have a particular function in the overall reaction mechanism.[17 ‒ 19] For instance, the use of chromium-based catalyst enhances the presence of amorphous silica dioxide phase, whereas molybdenum exhibits excellent catalyst attrition resistance, facilitates easy products desorption from the catalyst surface, maintain optical defect concentration, and blocks non-selective sites. Thereby, the new physical and chemical properties derived from synergistic effects between the two metals are highly desirable for catalytic applications. Incorporation of hetero-atoms into a silica framework has been reported to increase thermal stability  and may also increase the acidity of the support.  Transition metal oxides have a tremendous importance in the field of heterogeneous catalysis, serving as either catalysts or as supports for other catalytically active species. For instance, amorphous silica is an important support in catalytic technology due to its thermal stability, tunable porosity, outstanding specific surface area, and tremendous metal(s)-support interaction. [19, 22 – 24] Generally, it is believed that for metals on an irreducible oxide support: the strength of metal-metal bond is significantly larger than that of the metal-support bond. In addition, depending on the particular metal-oxide system, oxidation and reduction at elevated temperatures are essential steps for the preparation of high surface area supported catalysts; notwithstanding, these treatments can cause various morphological alterations such as thermal-sintering,[25 – 26] encapsulation,[27 – 28] inter-diffusion,[29 – 31] and alloy formation. In particular, alloy formation from metals supported on silica has received considerable attention because of its adverse influence to significantly alter catalytic activity and selectivity.[33 – 34] A unique feature of supported metal oxide catalysts is that the active component should be exclusively present as a surface phase, 100 % dispersed, below monolayer coverage, and that there is no spectroscopic complication from the co-existence of bulk crystalline phases. In addition, It is generally accepted that the activation of a hydrocarbon molecule on oxide catalysts takes place on a centre involving a M–O acid–base couple, oxide ions (O): playing the role of a hydrogen atom abstraction centre, cationic centers (M): facilitating the electron transfer. However, in spite of the numerous studies[19 – 34] on metals supported on SiO2 at elevated temperatures, there are still controversial and unresolved issues regarding: (i) the nature of the metal-support interaction between metals and SiO2; (ii) the morphological changes that occur during the high temperature reduction of metals supported on SiO2; (iii) the role of oxygen vacancies in the inter-diffusion of metals into SiO2; (iv) the extent to which silicides are formed by the direct interaction between metals and SiO2; (v) the role of the silicon substrate, frequently used to prepare SiO2 thin films, in metal silicide formation; (vi) the composition of metal silicides (if formed); (vii) the mechanism of silicide formation the between metals and SiO2. Notwithstanding, since catalyst activity and selectivity[35 – 37] are highly dependent on the size, shape, and nature of oxide support, it is therefore, of considerable importance to investigate and define the optimal conditions for catalyst preparation, pre-treatment and activation. An atomistic understanding of catalyst systems is essential to the rational design of improved catalyst systems. To the best of our knowledge, there has been only one report in the literature, so far, describing the synthesis of bi-functional Cr-Mo/SiO2 catalyst and its application for the catalytic oxidative desulfurization of diesel fuel. The objective of this contribution was to verify the morphological alterations after catalyst synthesis, and study the effects of Cr, Mo, and (Cr-Mo) supported silica catalysts on the oxidative dehydrogenation of C3–C4 fraction to olefin, at temperatures between 500 – 650oC, atmospheric pressure (P = 1 atm.), gas hourly space velocity (GHSV) within 100 – 400 h-1, and at reaction time (t r) = 2h in a continuous flow quartz reactor.
2.1 Materials and Methods:
Tetraethyl orthosilicate (TEOS, 99 %) was obtained from Sigma-Aldrich. Ethanol (C2H5OH, 99.5 %) was obtained from Acros Organics. Citric acid (C6H8O7) and hydrochloric acid (HCl, 37 %) were obtained from Fischer Scientific. Propane-butane fraction (C3–C4, 99.9 %) was supplied by Naftogas, Kiev. Oxygen (O2, 99.99 %) and Nitrogen (N2, 99.99 %) were obtained from Azot chemical company; and distilled water (H2O, 1 dm3).
2.2 Catalysts Preparation:
Wet impregnation was used in the preparation of co-impregnated catalysts. The quantitative metal precursors (Сr, Mo, or Cr–Mo) were dissolved in 15 mL de-ionized water. The pH of the aqueous solution was adjusted to 2–3 by adding 0.5 mol/L citric acid. 21 mL of tetraethyl orthosilicate (TEOS) and 22 mL of ethanol (C2H5OH) were added to the solution to produce the TEOS-C2H5OH solution. Then, the aqueous metallic salt solution was dropped slowly into the alcohol solution, and the mixed solution was stirred at room temperature for 1 h to produce sol. The sol was placed at room temperature for 1 h and aged at 40oC for 1 h to produce gel. The gel was dried under ambient atmosphere at 120oC for 12 h, and the catalyst precursor was obtained. Finally, the precursor was calcined at 500oC for 4 h. The amorphous silica (surface area = 200 m2/g) was carefully treated by washing in 2 M of HCl (hydrochloric acid) to remove any volatile impurities adsorbed on the surface (Fig. 2). (i) The 10 % wt. Cr mono-metallic catalyst was prepared by incipient wetting of amorphous silica with an aqueous solution of chromium chloride (CrCl2); (ii) The 15 wt. % Mo mono-metallic catalyst was prepared by incipient wetting of amorphous silica with an aqueous solution of ammonium heptamolybdate ([NH4]6Mo7O24.6H2O); and (iii) The 15 wt. % Cr-Mo bi-metallic catalyst was prepared by slowly dropping aqueous solutions of chromium chloride and ammonium heptamolybdate ([NH4]6Mo7O24.6H2O) in a shaking water-bath at 20oC. After the impregnation, the resulting mixture was stirred, filtered, and then dried at 150oC. Then both mono and bi-metallic catalyst sample were pressed, and then sieved to appropriate sizes for catalytic evaluations. Metal(s) loading was varied between 10 – 15 % wt.
2.3 Catalysts Treatment:
0.1g of each catalyst was calcined in air at 500oC for 6 h to remove any volatile impurity adsorbed on the surface, followed by reduction in 10 % H2/ 90% Ar at 500oC for 6 h to minimize coking and enhance its dehydrogenation activity without influencing the secondary reaction of light olefins production. The total flow rate of the H2/Ar mixture was 50cm3/min. The samples were then cooled at room temperature for 30mins, and stored in inert atmosphere to avoid catalyst degradation.
2.4 Catalysts Test:
The catalytic activity of the samples for the decomposition of C3–C4 fraction was investigated in a continuous flow quartz fixed-bed reactor (6 L vol.). The catalysts sample (0.15 g) was packed in the reactor and activated with flowing N2 at 500oC for 2 h. After which the flow rate of reactant (C3–C4) and N2 was maintained at 4.0 ml/min and 96 ml/min, respectively. The N2 adsorption isotherms of calcined materials were measured at liquid nitrogen temperature (‒196oC). The gas carrier was passed through a molecular sieve trap before being saturated with C3–C4. The gas product samples were analyzed by gas chromatograph.
2.5 Oxidative De-hydrogenation of C3–C4:
The C3–C4 feed composition was analyzed by gas chromatography «Chrom‒5». It was established that the total composition equals 100 % by volume: propane = 20, i -butane = 60, and n -butane = 20. The oxidative dehydrogenation experiment of C3–C4 fraction was carried out in a continuous flow quartz fixed-bed reactor (6 L vol.). The experimental set up is shown in Fig. 1.
Abbildung in dieser Leseprobe nicht enthalten
Fig. 1 Schematic diagram of the pilot unit for dehydrogenation of C3–C4 fraction
Prior to each run, the reactor was purged with N2 for about 10 min and then de-coked using 15 % O2: 85 % N2 mixture to ensure that the reactor walls and the coupon were coke free. This was accomplished by visually observing the appearance of the coupon through an observation hole in the furnace and by monitoring weight of the coupon during the decoking process. If the appearance of the coupon was transparent and non-luminous, and its weight did not decrease with time, the coupon was assumed to be coke free. The reactor was again purged with N2 for about 10 min, after which the hydrocarbon reactants and steam were introduced. The primary reason for N2 purge before and after decoking experiments was to minimize the accumulation of potentially explosive mixtures in the reactor. Each run was repeated at least five times to ensure reproducibility and to assess the range of experimental errors associated with the experiments. In order to determine the catalytic specie with the best performance, three series of experiment were performed. In the 1 st experiment, C3-C4–oxygen–vapor mixture on the 10 % wt. Cr/SiO2 catalyst was fed into the reactor. The propane-butane fraction, oxygen, and nitrogen were supplied from a pressure cylinder through reduction valves. The reactant gases consisting of C3–C4, O2, and some additional N2 carrier gas were then mixed with steam and transported to the reactor through electrically heated lines at a flow-rate desired (± 5%) for the given experiment. A molecular sieve was used at the entrance of the reactor with the objective of retaining impurities coming from the feeding line. The flow rate of C3–C4 fraction, O2, and N2 was regulated by mass flow controller that was calibrated before the experiments. The reactant mixture was subjected to thermal treatment at temperature range within T = 500 – 650oC, atmospheric pressure (P = 1 atm.), and for (t r) = 2 h total reaction time. The flow of bulk gas through the reactor was 100 cm3/min, and 1 L/min for water. The average residence time of reactant mixture in the reactor was about 10 s. The gas hourly space velocity of 100 – 400 h–1 was varied to obtain different level of C3–C4 conversion. The catalyst activity was maintained by regeneration after every experimental hour using a N2/O2 mixture. During the regeneration, the output stream from the reactor was checked for carbon dioxide. The reaction products were obtained at the bottom of reactor, cooled, separated into individual components, and analyzed at 10 min. intervals by gas chromatograph–mass spectrometer (GC–MS). In the 2 nd experiment, C3-C4–oxygen–vapor mixture on the 15 % wt. Mo/SiO2 catalyst was fed into the reactor. In the 3 rd experiment, C3-C4–oxygen–vapor mixture on 15 % wt. (Cr-Mo)/SiO2 catalyst was fed into the reactor. The same experimental condition described in 1st experiment was applied for the 2nd and 3rd experiments.