Apr-1997

Claus sulphur recovery options

This article examines some Claus modifications which can alleviate operational difficulties and improve overall sulphur recovery

Gavin McIntyre and Lili Lyddon
Bryan Research and Engineering

Article Summary

Since its inception, the Claus process has been the standard of the sulphur recovery industry, but limitations and problems relating to composition may restrict its effectiveness. Numerous modifications have been applied to the basic process in an effort to develop the optimum system for a certain set of conditions.

The primary consideration in determining the optimum sulphur removal process is the composition of the acid gas stream to be processed. A typical rich feed gas stream to a Claus plant contains at least 50 per cent H2S by volume (50–60 per cent is considered marginal). The traditional Claus plant may be used for primary sulphur recovery from such a rich acid gas stream. Unfortunately, the H2S content of acid gas is sometimes very low (5–50 per cent) with CO2 making up the bulk of the feed.

This lean feed is not sufficient to sustain combustion in the burner of the traditional Claus reaction furnace, so some modification to the process is required. In addition, acid gas feeds may contain undesirable components such as ammonia and hydrocarbons, which cause problems during processing. Again, the traditional Claus process must be modified to handle these contaminants.

For the purposes of this article, acid gas is classified into three different categories based on composition, and the types of Claus or modified Claus processes that may be used to lower sulphur emissions to an acceptable level are discussed.

The first type of feed gas discussed is rich acid gas, or feed containing greater than 50 mole% H2S. The second type is lean acid gas, or feed containing less than 50 mole% H2S. The last type of feed to be considered is acid gas containing ammonia. A process simulator, TSWEET, illustrates how effective various processes are for each of the three acid gas feed types.

Case studies: 1
Rich acid gas feed – H2S content above 50 per cent
Several options are available for consideration when designing the primary sulphur recovery unit to process a rich acid gas feed. Even though many variations of the standard Claus plant may be suitable, one may have a slight advantage over the others.

Depending on the required sulphur removal, a number of configurations could be used for primary sulphur removal, such as a two- or three-bed Claus plant, with or without a direct oxidation bed or cold (sub dew point) bed. These and other variations are compared, based on their ability to effectively process a rich acid gas stream. Figure 1 is a schematic of the basic Claus process. Some of the more common process modifications to the basic Claus, discussed in the following sections, are indicated with dashed lines.

A two-bed Claus plant operating in the 1960s with approximately 93 per cent H2S in the feed is used as the base case for this discussion. A process simulator, TSWEET, was used to obtain an initial model by matching plant data. The model was then modified to show how improvements could be made.

A comparison of plant data to TSWEET predictions is listed in Table 1 to provide a basis for the accuracy of the program’s predictions for further modifications.

This particular plant achieved an overall sulphur recovery of 91.8 per cent, which is low by current industry standards. Upon examination, there are several areas where small operational or design changes could improve recovery. Some of these changes were simulated to show how the plant might operate if it were optimised by today’s standards.

The outlet temperature of the waste heat boiler was decreased from 840°F to 700°F and a sulphur condenser was added to remove all the sulphur formed in the furnace. The condenser improves recovery because the presence of elemental sulphur in the catalyst beds reduces the efficiency of the Claus reaction. Also, the outlet temperature of the first reheater was controlled to maintain a bed outlet temperature of 650°F rather than 727°F.

This lower temperature promotes improved Claus conversion but still ensures destruction of the COS and CS2 formed in the burner. Similarly, the second reheater was controlled to maintain the outlet temperature of the second bed at 30 degrees above the sulphur dew point. The closer to the sulphur dew point the bed operates, the higher the equilibrium conversion.

One final item to note concerning this case is the ratio control of H2S to SO2 in the tail gas. The Claus reaction requires two moles of H2S for every one mole of SO2 for optimum conversion according to the following reactions:
H2S + 11/2O2 Æ SO2 + H2O
2H2S + SO2 € 3S + 2H2O

This plant originally operated at a ratio of 1.2:1, but in the simulation the ratio was changed to 2:1 to optimise the performance of the unit. The ratio is controlled by manipulating the flow rate of inlet air from the blower to the furnace. These seemingly minor changes to operating conditions increased sulphur recovery from 91.8 per cent to 96.1 per cent, a 4.3 per cent improvement over the original design, as shown in Table 1. These modifications are used as the basis for comparisons with other process modifications.