Sasol, process scheme

Consequently, two semicommercial pilot plants have been operated for 1.5 years. One plant, designed and erected by Lurgi and South African Coal, Oil, and Gas Corp. (SASOL), Sasolburg, South Africa, was operated as a sidestream plant to a commercial Fischer-Tropsch synthesis plant. Synthesis gas is produced in a commercial coal pressure gasification plant which includes Rectisol gas purification and shift conversion so the overall process scheme for producing SNG from coal could be demonstrated successfully. The other plant, a joint effort of Lurgi and El Paso Natural Gas Corp., was operated at the same time at Petrochemie Schwechat, near Vienna, Austria. Since the starting material was synthesis gas produced from naphtha, different reaction conditions from those of the SASOL plant have also been operated successfully. 

A generalized scheme of the Sasol process is illustrated in Figure 1. The basic raw materials are coal, water and air. The plant is a complex operation consisting of many interlinked processes. This complexity is, however, not an important factor in the economics when it is borne in mind that the main cost is the production of synthesis gas, which accounts for over 50 % of the total. 

Figure 3 - Original Process Scheme of the Sasol 1 Plant - Sasolburg, South Africa...

This first Sasol LTFT plant had a complex processing scheme. This original concept, shown in Figure 4, showed a high degree of integration to maximize the benefits of the unique characteristics of the FT products (14). 

The Secunda process scheme was conceived to maximize gasoline production - therefore, it includes hydroprocessing and catalytic reformers similar to those used in petroleum refineries. Due to the scale of operation, it includes facilities for the recovered of ethylene, alcohols, ketones, phenols, ammonia and other chemical products. Its twin plant at the same location, Sasol 3, has a very similar configuration. At present all the original Synthol reactors have been replaced by the more efficient Sasol Advanced Synthol (SAS) reactors, with capacities of up to 20 000 bpd per train. 

In 1983 Dr Mark Dry, working for Sasol, proposed a process scheme for maximum diesel production in 1983 (15). While this scheme - shown in Figure 6 - was conceived for a coal based LTFT plant, it is also applicable when natural gas is selected as feedstock thus replacing the coal gasifiers with natural gas reformers. The major difference relative to most LTFT GTL concepts is the inclusion of an oligomerisation unit for the conversion of the light hydrocarbons to liquid fuels which might lead to some 7% additional distillates production compared to equivalent schemes that exclude this process. 

The new Sasol II plant is intended to produce mainly gasoline and is therefore based entirely on Synthol technology. The Sasol II reactors have each a capacity of 2 1/2 times that of Sasol I. A simplified process scheme is shown in Fig. 4/ and Table 2 summarizes the production figures of Sasol II (32). The productivity of Sasol II is more than twice as large as that of all plants operated in Germany in 1944 and even considerably larger than the world FT productivity at that time. How-... 

Fig. 12. Flow scheme for the SASOL I Fischer-Tropsch process. To convert MPa to psig, multiply by 145.

Although the Fischer-Tropsch process (based on coal) has been uneconomic for many years compared to oil-based routes, the peculiar political situation of South Africa allied with its large reserves of cheap coal led it to continue operating this process. Indeed, the experience gained led in the 1970s to the building of a second-generation plant, (the scheme for this SASOL II plant is shown in Fig. 2.4" ) and a SASOL III plant is now on stream. 

A related mechanism for the tetramerization of etfiylene has been deduced from studies on Sasol s chromium-catalyzed process that forms 1-octene. The catalyst is, again, thought to operate by a Cr(I)/Cr(III) cycle such as the one in Scheme 22.17. This mechanism was supported by several studies. For example, Cr(0) complexes of the PNP ligand [Cr(CO) (PNP)] were shown to be catalytically inactive, but the complexes formed from these Cr(0) species and AgX species were found to be active. Moreover, labeling studies showed that the side products were formed from a metallacyclic intermediate, rather than a Cossee enchainment mechanism. This labeling study was analogous to that published by Bercaw discussed above. The octene presumably forms from an unusual nine-membered metallacyclic intermediate. 

The required terminal olefins used as substrates for the hydroformylation, such as 1-pentene or 1-octene, are available in large scales and can be derived either from Sasol s Fischer-Tropsch process or from the shell higher olefins process (SHOP), respectively [43, 44]. Alternatively, trimerization or tetramerization of ethylene affords 1-hexene [45] or 1-octene [46]. Dimerization of butadiene in methanol in the presence of a Pd catalyst (telomerization) is another industrially used access for the manufacture of 1-octene [46]. 1-Octene can also be produced on a large scale from 1-heptene via hydroformylation, subsequent hydrogenation, and dehydration (Scheme 6.2) [44]. This three-step homologation route is also valuable for the production of those higher olefins that bear an odd number of C atoms. (X-Olefins can also be derived from internal olefins by cross-metathesis reaction with ethylene [47]. 

Operation at very low temperatures with very sharp separations results in relatively complex flow schemes. This, combined with the need for low level refrigeration, leads to high plant costs. As a result, most applications of the Rectisol process represent relatively difficult gas treating conditions where other gas treating processes are not suitable for one reason or another. Typical applications are the purification of gas streams in the heavy oil partial oxidation processes of Shell and Texaco and the Lurgi coal gasification process, as used at the Sasol plants in South Africa. 

---