Fischer reactor conditions

Graver, V., Zhan, X., Engman, J., Robota, H. J., Suib, S. L., and Polverejan, M. 2004. Deactivation of a Fischer-Tropsch catalyst through the formation of cobalt carbide under laboratory slurry reactor conditions. Prepr. Pap.-Am. Chem. Soc. Div. Pet. Chem. 49 192-94. 

Of the technological modifications, Fischer-Tropsch synthesis in the liquid phase (slurry process) may be used to produce either gasoline or light alkenes under appropriate conditions249,251 in a very efficient and economical way.267 The slurry reactor conditions appear to establish appropriate redox (reduction-oxidation) conditions throughout the catalyst sample. The favorable surface composition of the catalyst (oxide and carbide phases) suppresses secondary transformations (alkene hydrogenation, isomerization), thus ensuring selective a-olefin formation.268... 

The CO-hydrogenation reaction, or Fischer-Tropsch (F-T) synthesis reaction, has been thoroughly investigated since its discovery fn the 1920 s [1]. A range of catalysts has been shown to be active for hydrocarbon synthesis and iron [2] and cobalt [3] have found commercial applications in this field. A variety of reactors have been developed to optimize the synthesis reaction [4]. Variations of reactor conditions have been shown to maximize specific products from the broad range of products produced in the reaction [5). 

The Fischer-Tropsch Synthesis (FTS) converts synthesis gas (a mixture of CO and H,) to hydrocarbons. Iron-based catalysts lose activity with time on stream (TOS). The rate of deactivation is dependent on the presence/absence of promoters such as potassium and/or binders such as silica [1.2]. Several possible causes of catalyst deactivation have been postulated [3] (i) Sintering, (ii) Carbon deposition, and, (iii) Phase transformations. With respect to phase transformations, there is considerable disagreement whether the active phase for the FTS is iron oxide or carbide [4,5]. In addition, certain reactor conditions, such as a high partial pressure of water, are known to cause a decline in activity [6]. 

Laboratory Fischer-Tropsch synthesis tests were performed in a slurry-phase Constant Stirred Tank Reactor. The pre-reduced catalyst (20-30 g) was suspended in ca 300 ml molten Fischer-Tropsch wax. Realistic Fischer-Tropsch conditions were employed, i.e. 220 °C 20 bar commercial synthesis gas feed 50 vol% H2, 25 vol% CO and 25 vol% inerts synthesis gas conversion levels in excess of 50%. Use was made of the ampoule sampling technique as the selected on-line synthesis performance monitoring method [23]. 

The unmodified alumina-supported cobalt catalyst (catalyst A) was tested in the pilot plant slug-flow reactor under realistic Fischer-Tropsch conditions. The produced wax was... 

Table 2. Activities and selectivites measured in a fixed-bed reactor at Fischer-Tropsch conditions (T = 473 K, P = 20 bar, Hj/CO = 2.1).

Cobalt has been used as a Fischer-Tropsch catalyst in a variety of forms [8]. Thus it was not surprising to see that both active forms of cobalt powders were moderate Fischer-Tropsch catalysts. Reacting synthesis gas with 2 in batch reactor conditions at elevated pressure and temperatures generated methane as the primary product. The life spans of the catalyst and to a lesser extent the products were affected by whether a support was used or how the cobalt was deposited on the support. Catalytic activity was not especially high and amounted to 4-7 mol of methane/mol of cobalt. 

Within each syncrude type some variation is introduced by the operating conditions of Fischer-Tropsch synthesis, such as pressure and H2 CO ratio, as well as by the Fischer-Tropsch reactor type. These variations cannot be ignored, and ultimately they have an impact on the refinery design. During the subsequent discussion it will become apparent that the selection of the Fischer-Tropsch technology influences not only the refinery design, but also the efficiency with which different products can be produced. 

The properties of these new materials as catalyst support were tested on Fischer-Tropsch process (CO-H2 reaction) in a fixed bed differential reactor. Three materials were tested a) CON, a conventional activated carbon b) SC-155 (G40.60) and c) C-155 (G20.20). All of them were previously iron doped until 5% metallic iron wt/wt was reached. The test conditions were Reaction temperature =270°C H2/CO ratio=3, pressure = latm. The main properties of the tested catalyst supports and their performance in the first hour test are shown in Table 2. SC-155 (G40.60) and C-155 (G20.20) were selected for this test in order to compare materials with near the same specific surface area but with different structural composition, and CON was selected because it is of common use and has very different texture characteristics respect to the other two materials. 

Low-temperature Fischer-Tropsch (LTFT) synthesis runs at temperatures between 200°C and 250°C [23-25]. The chain-growth probability at these conditions is much higher than for the HTFT, and as a consequence, the product distribution extends well into the liquid waxes. LTFT reactors are thus three-phase systems solid catalysts, gaseous reactants, and gaseous and liquid products. Both cobalt and iron... 

Phillips and Timms [599] described a less general method. They converted germanium and silicon in alloys into hydrides and further into chlorides by contact with gold trichloride. They performed GC on a column packed with 13% of silicone 702 on Celite with the use of a gas-density balance for detection. Juvet and Fischer [600] developed a special reactor coupled directly to the chromatographic column, in which they fluorinated metals in alloys, carbides, oxides, sulphides and salts. In these samples, they determined quantitatively uranium, sulphur, selenium, technetium, tungsten, molybdenum, rhenium, silicon, boron, osmium, vanadium, iridium and platinum as fluorides. They performed the analysis on a PTFE column packed with 15% of Kel-F oil No. 10 on Chromosorb T. Prior to analysis the column was conditioned with fluorine and chlorine trifluoride in order to remove moisture and reactive organic compounds. The thermal conductivity detector was equipped with nickel-coated filaments resistant to corrosion with metal fluorides. Fig. 5.34 illustrates the analysis of tungsten, rhenium and osmium fluorides by this method. 

In the design of upflow, three phase bubble column reactors, it is important that the catalyst remains well distributed throughout the bed, or reactor space time yields will suffer. The solid concentration profiles of 2.5, 50 and 100 ym silica and iron oxide particles in water and organic solutions were measured in a 12.7 cm ID bubble column to determine what conditions gave satisfactory solids suspension. These results were compared against the theoretical mean solid settling velocity and the sedimentation diffusion models. Discrepancies between the data and models are discussed. The implications for the design of the reactors for the slurry phase Fischer-Tropsch synthesis are reviewed. 

The range of operating conditions for the 276 experimental run in the 12.7 cm column and 20 experimental runs to date in the 30.5 cm column are shown in Ta ble I. Relevant physical properties of the liquids are listed in Ta ble II., and compared with estimated data for the slurry phase Fischer-Tropsch pilot plant reactor at Rheinpreussen (12). Solid densities were obtained from the literature (13). As received, the isoparaffin (lM) sample was saturated with water. However, this ppm level of water was soon removed during the initial experiments by the dry nitrogen gas. Additional isoparaffin was added when required to maintain the solid concentration weight-percent. All water based runs used humidified air. 

Catalyst activity is also affected by carbon deposits formed by the Boudouard reaction. At usual Fischer Tropsch synthesis conditions, iron catalysts form carbides but little or no deposition of elemental carbon is observed. At temperatures exceeding 270 C carbon deposition becomes critical for the plugging of fixed bed catalyst reactors [22]. Under these conditions the catalyst particles swell upon carbon deposition and also disintegrate. 

The Fischer-Tropsch (FT) synthesis involves catalytic reactions in which CO and H are reacted to form mainly aliphatic straight-chain hydrocarbons (C Hy). The kind of liquid obtained is determined by the process parameters (temperature, pressure), the kind of reactor, and the catalyst used. Typical operation conditions for the FT synthesis are a temperature range of 200-3 5 0 C and pressures of 15-35 bar, depending on the... 

The reactor operates either at low (200-240°C) or high (300-350°C) temperatures and between 1-4 MPa. The product mix changes from longer to shorter chain length molecules as the temperature increases. The product stream from a Fischer-Tropsch reactor is a mix of many components but by selecting the right operating conditions the mix can be adjusted so that the product stream has mostly diesel fuel properties. 

Many of the modern combustion processes can be characterized by relatively low reaction rates compared to the modern catalytic processes operated in chemical reactors [67]. Therefore, these combustion processes do require lower gas velocities and higher solids circulation rates. On the other hand, many catalytic gas-phase reactions, including FCC, Fischer-Tropsch synthesis and oxidation of butane, utilize a relatively high gas velocity in the riser to promote plug flow operating conditions and short contact times between the gas and solids. 

Support modification has been reported earlier in the open literature [5,6,7,8,9]. Zirconia modification of silica supports was used to prevent the formation of unreducible cobalt-silicates [5]. Zr, Ce, Hf, or U modification of titania supports was reported to prevent the formation of cobalt-titanates during regeneration [6]. To increase the porosity of titania supports, they were modified with small amounts of binders, e.g. silica, alumina or zirconia [7]. Lanthanum oxide promotion of alumina was reported to be beneficial for improved production of products with higher boiling points [8], and zirconia modification of alumina supports was carried out to decrease the interaction of cobalt with alumina [9]. All these modified supports were either used for fixed bed cobalt based Fischer-Tropsch synthesis catalysts or they were used for slurry phase cobalt catalysts, but not tested under realistic Fischer-Tropsch synthesis conditions in large scale slurry bed reactors. 

Larger scale Fischer-Tropsch synthesis runs were performed in a pilot plant slug-flow slurry reactor using 3-8kg catalyst as well as in a slurry phase bubble column demonstration unit using 500-1500kg catalyst. The reaction conditions were similar to those in the laboratory CSTR runs. The reactor wax production varied between 5 and 30kg per day for the pilot plant runs and up to 60 bbl/day for the demonstration unit. On-line catalyst samples were taken for particle size distribution measurements and Scanning Electron Microscope analyses. 

The silica modified supported cobalt catalyst. Catalyst B, were tested in a Pilot Plant slurry bubble column reactor under realistic Fischer-Tropsch synthesis conditions and it... 

Another proposal for explaining the two slope distributions is very consistent with the peculiarities of the Fischer Tropsch system The products of Fischer Tropsch synthesis do usually provide a liquid phase and a gaseous phase under reaction conditions.The gaseous compounds leave the reactor normally within a few seconds. The liquid does need a day or more until it elutes from the catalyst bed. Solubility of paraffinic hydrocarbon vapours in a paraffinic hydrocarbon liquid increases by a factor of about 2 for each carbon number of the product (ref. 27). Thus it needs only an increase of a very few carbon numbers of the product molecules to have them leaving the reactor mainly with the gas phase or with the liquid phase. With increasing residence time in the reactor the chance of readsorption increases and correspondingly the probability of chain prolongation increases. The kinetic scheme of this model is shown in Fig. 14. This model is very consistent with the experimental distributions. 

---