Desulfurization of Coke-Oven Gas
O. I. Platonov
OOO Institut Gipronikel, St Petersburg, Russia
Sourse: Coke and Chemistry number 08, 2007
As we know, coke-oven gas contains ammonia and hydrogen sulfide, which must be removed (to residual contents of <0.03 and <0.5 g/m3 according to European standards). Therefore, the removal of hydrogen sulfide from coke-oven gas by the ammonia method is in increasing use. This method includes the capture of hydrogen sulfide by an aqueous solution of ammonia and subsequent treatment of the acidic gas by the Claus process, to obtain elementary sulfur, which is a commercial product. After this method was introduced at enterprises in North America, two lines for sulfur production from coke-oven gas went into operation in Poland, in 1997 and 1999 . There is also information on the design, construction, and startup of similar units for removing sulfur from coke-oven gas in the Czech Republic (Nova Huta, Ostrava), in France (Dunkirk), China (five units), South Korea, and elsewhere.
Factors encouraging the wide use of the ammonia method of hydrogen-sulfide removal from coke-oven gas include the absence of toxic emissions and the generation of a product (elementary sulfur) capable of withstanding transportation and prolonged storage. This is especially important in view of the global overproduction of sulfur and sulfuric acid and the geographic separation of regions that produce and consume sulfur. Accordingly, there are good prospects for wider use of the ammonia method. In this context, experience in introducing this method in the coke plant at OAO Magnitogorskii Metallurgicheskii Kombinat (MMK) may be of interest for Russian coke specialists. In the present work, we analyze experience at MMK in introducing the least-developed part of the technology: the processing of gas that contains hydrogen sulfide so as to obtain commercial sulfur.
DESULFURIZATION OF COKE-OVEN GAS
The extraction of hydrogen sulfide from coke-oven gas by ammonia liquor is well established in the coke industry and may be regarded as sufficiently welldeveloped. However, the processing of acidic gas containing hydrogen sulfide and ammonia in the coke industry differs from the classical Claus processing of hydrogen sulfide. Profound sulfur extraction in classical Claus equipment involves four or more catalytic-conversion stages, with corresponding capital expenditures. Both thecapital expenditures and the losses of sulfur are reduced in a design employed by Krupp–Koppers (now Thyssen– Krupp EnCoke, Germany).
In this design, after sulfur extraction in a condenser and trapping in a separator, the tail gas from the sulfurremovalunit (SRU) is cooled to ~60°C in a contact cooling unit by excess water from the gas collector, according to patent. This gas is then returned to the incoming coke-oven gas, thereby boosting the reserves of hydrogen sulfide. To reduce the sulfur losses with SO2 in the coolant and simplify the regulation of the process, treatment of the acidic gas by the Claus method is characterized by high Claus ratio (HCR technology). The air supply to the Claus furnace is regulated so that the Claus ratio at the SRU output is CR = [H2S]/[SO2] = 10.
The first SRU of Krupp–Koppers type in the former Soviet countries went into operation in September 2000in purification shop 2 for coke-oven gas at OAO MMK. Operational experience in this shop indicates that the desulfurization technology is highly efficient. The existing system with a closed hydrogen-sulfide cycle differs somewhat from a classical SRU. This must be taken into account in the design of future equipment (shops) for the purification of coke-oven gas, with sulfur production. In the existing system for sulfur removal from acidic gas in purification shop 2 at MMK, the hydrogen-sulfide concentration is low, while the ammonia concentration in the acidic gas is high. According to measurements throughout shop operation, the H2S concentration is 5–8 vol % and the NH3 concentration is 12–17 vol %. This may be attributed to inadequate initial data on the design (the H2S reserves in the coke-oven gas are overestimated by a factor of 2–3) and to thedesign of the section for trapping hydrogen sulfide.