Feb-2019

Claus waste heat boiler economics Part 1: process considerations

The design of the waste heat boiler has a critical role to play in determining process operations, reliability and economic performance in the sulphur recovery unit.

NATHAN A HATCHER, CLAYTON E JONES, SIMON A WEILAND, STEVEN M FULK and MATTHEW D BAILEY
Optimized Gas Treating, Inc.

Article Summary

Higher sulphur feedstocks offer margin advantages to refiners and gas producers but with challenges to uptime metrics and environmental constraints as downsides. To meet economic objectives, such competing requirements must all be met reliably without equipment failure. One key piece of equipment that can impact uptime is the Claus waste heat boiler (WHB). The WHB operates under harsh conditions with serious reliability challenges. It is one of the most fragile equipment items in the sulphur recovery unit (SRU). Not only does it provide heat recovery from the thermal section of the SRU, but it also affects the unit’s hydrogen balance and COS levels because of recombination reactions. Peak heat flux at the front end of the boiler tubes can lead to over-heating and reliability problems. The design of the WHB plays a critical role in determining the extent of the recombination reactions as well as the peak heat flux, all while balancing capital cost and operational reliability. Part 1 of this two-part series discusses general process considerations including the effects of tube length, pressure drop, and COS/H2 reactions on sulphur recovery. Part 2 examines details of how operating conditions and WHB design affect tube wall temperature and the important reliability implications of that relationship.

The SulphurPro simulator provides a modern heat transfer rate and chemical reaction rate based model of the WHB. The quantitative behaviour of the WHB itself and how it subtly affects SRU performance hinges critically on the recombination reactions that occur at the front of the WHB:

H2 + ½ S2 ⇌ H2S
CO + ½ S2 ⇌ COS

These reactions are so important because not only do they influence sulphur recovery, air demand, and hydrogen production in the SRU, they also affect the heat flux and therefore performance and mechanical longevity of the WHB. These reactions occur towards the front (inlet) side of the WHB and are exothermic. As will be seen in Part 2, the heat associated with them tends to increase heat flux especially in the vicinity of the critical tube-to-tubesheet joint.

Radiative heat transfer, coupled with the exothermic recombination reactions, collectively increase the peak heat flux at the front of the boiler well above predictions from models that ignore or discount these factors. Ceramic ferrules are often used to protect the front end of the tubes from excessive heat, but detailed modelling shows that, at high mass velocity, the tubes can experience greatly elevated tube wall temperatures well downstream of the ceramic ferrules. Details of this modelling effort are presented in Part 2 of this article, lending theoretical support to documented failures in the industry. Also in Part 2, tube wall temperatures, pressure drop, and heat flux predictions from the model are examined down the length of the tubes for oxygen enriched vs air only operations, and the implications of sulphidic corrosion with its resulting effect on boiler tube life and SRU reliability economics are examined with this new information.

Case study: important parameters in WHB design
There are many parameters in Claus WHBs that need to be very carefully considered in the design phase. Many of the parameters are mutually interdependent – changing one affects several others. In a typical WHB design, the tube size and process-side mass velocity will be chosen to meet a fixed outlet temperature specification. The following case study looks at a typical 125 lt/d sulphur plant with two converter stages processing both amine acid gas (AAG) and sour water acid gas (SWAG) under low level oxygen enrichment (see Figure 1). Table 1 shows the conditions of these two acid gas feed streams. The WHB was sized by fixing the process-side mass flux, the tube size and the process outlet temperature. The tube count and tube length were adjusted to meet the target specifications. A matrix of three tube sizes and four mass fluxes was evaluated. The tube sizes were set at 1.5in, 2in and 3in outside diameter and process-side mass fluxes were set at 2, 3, 4, and 5 lb/ft2·s. The results were obtained using the kinetic heat transfer and chemical reaction rate based SulphurPro SRU Simulator (available standalone or integrated within ProTreat Version 6.4).

Table 2 shows values of other design parameters assumed for this application of the model. The boiler produces 350 psig saturated steam from preheated boiler feed water.

Table 3 illustrates how, in meeting an outlet temperature specification of 550°F (288°C), the overall tube length increases with mass velocity at three tube diameters. Because the hot gas is moving faster at the higher mass velocities, less heat is transferred from the gas in a given tube length, thus the overall length of the tube bundle must be increased to meet the outlet temperature target. Besides the amount of cooling provided by the WHB, another parameter of importance is pressure drop through the exchanger. In a sulphur plant, pressure drop is at a premium and must be considered in all aspects of the design. Tube size and length have a large effect on pressure drop. Using smaller tubes decreases the overall length of the tube bundle, but it also increases pressure drop per unit length, so a balance must be struck. Neglecting any single factor of the design can lead to design busts and catastrophic operations failures. The driving factors for any project design are meeting the requirements safely and reliably whilst minimising cost. Sometimes these factors are at odds with each other – the interaction between design parameters can be complex. The base cost of a WHB depends directly on the amount of material needed for construction. Boiler tube length, tube diameter, and tube count all influence the amount of material needed.

As expected, at each tube size, as the mass flux (lb/ft2·s) through the tubes increases, the overall tube length needed to achieve a fixed outlet temperature increases. At the smallest tube size (1.5in) and the lowest mass flux (2 lb/ft2·s), the boiler tube length is least (19.2 ft). But, with smaller boiler tubes, the overall tube count must be increased to process the specified mass flux. A direct consequence of the higher tube count and smaller tube diameter is a greater total surface area per length of tube bundle available for heat transfer so that the shorter tubes will achieve the same specified outlet temperature.

Pressure drop is another factor. It increases with increasing mass flux and decreasing tube diameter. Table 4 shows the calculated pressure drop through the tube bundle. This does not account for the exchanger effluent nozzle exit loss, and it assumes clean tubes (no dirt factor).

The data of Table 3 suggest that the smallest possible tubes and lowest possible mass flux requires the shortest tubes, but the data of Table 4 suggest the largest possible tube size with the lowest mass flux is best. The answer as to which is preferred (at least from a process engineering perspective) involves not just minimising the pressure drop, but also maximising hydrogen production in the SRU and concurrently maximising sulphur recovery efficiency. Also of paramount importance is the consideration of protecting the critical tube-to-tubesheet joint with ceramic ferrules. Allowing for the insulation thickness of the ferrules, the smallest practicable tube is about 1.5in OD. Tubes smaller than this would result in overly severe pressure drop and flow constrictions.