You are currently viewing: Articles


FCCU particulate emissions control with a Shell third stage separator:
a case study

Third Stage Separators (TSS) have been utilised for many years to protect turbo expanders installed on FCCU regenerator flue gas systems.

Edwin H Weaver
Belco Technologies Corporation (Now BELCO Clean Air Technologies)
Viewed : 2656
Article Summary
In more recent years, tremendous improvements have been made in the separation performance of the TSS. Although these were originally targeted at improving the reliability and run time of the turbo expanders, these improvements have allowed the Shell-TSS to reach emission levels that constantly meet the mandated stack particulate emission requirements such as new source performance standards (NSPS) and refinery MACT II. With the increased emphasis on lower particulate emissions from the FCCU, this is very important as in many cases the TSS now provides a viable alternative to an electrostatic precipitator.

The case studies examined follow the typical upgrade of an old Shell-TSS with modern internals for optimum emissions performance. First, the design of an old TSS is examined, along with pertinent implication to the system performance. Next, the modifications to the TSS are described along with the performance and reliability implications of each of the improvements. Other important components, such as the fourth stage separator (FSS) are also examined and the performance implications are discussed. Finally, the actual performance of the upgraded TSS system is examined and compared to mandated emission requirements. It will become clear that the upgraded system meets not only Mact II but has contributed to solve expander-fouling issues.

For many years, the TSS has been used as an effective device to remove catalyst fines from the FCCU regenerator flue gas in order to provide protection for a turbo expander. In this service, it was critical that the TSS removes a sufficient quantity of the catalyst fines so that the remaining catalyst fines in the flue gas would not damage the turbo expander. A typical arrangement of the TSS in this service is shown in Figure 1.

In this arrangement, flue gas from the FCCU regenerator passes through the TSS, where the larger catalyst fines are removed. The TSS underflow, may be routed to a fourth stage separator, but in many cases the underflow is simply returned to the main flue gas, where it is either cleaned by an electrostatic precipitator, a wet EDV scrubber or vented directly to atmosphere. When emissions to the atmosphere are the primary consideration, a FSS is critical to the overall performance of the system. Not only must the TSS operate efficiently, the underflow from the TSS must be efficiently cleaned of catalyst fines in a FSS before the underflow is returned to the flue gas. A typical arrangement of a TSS, designed for minimizing particulate emissions to the atmosphere, is provided in Figure  2. As can be seen in the Figure, the underflow from the TSS is passed through a FSS. The clean overflow from the FSS is returned to the main flue gas. The catalyst fines from the FSS must be collected for disposal. In many cases this is accomplished by collecting the catalyst fines in a hopper for disposal or blending with withdrawn spent catalyst in the spent catalyst hopper.

TSS design considerations
Although several factors are involved in the design of an efficient TSS, the most critical component, in terms of performance is the swirl tube design. Both the TSS and the swirl tube are illustrated in Figure 3. Inside the TSS, the flue gas is divided over multiple swirl tubes. The number of swirl tubes utilised is primarily dependent on the flue gas volume being processed. The original swirl tube design that was utilised in older TSS units was a swirl tube that incorporated a slotted plate at the bottom of the swirl tube. At times the slots in this bottom plate would plug, affecting both reliability and performance. Therefore, much of the development efforts were focused on a swirl tube design that uses an open end1. Of the modifications made to the swirl tube design, prominent are changes to the inlet vane configuration and on optimizing the swirl tube barrel length to diameter ratio. This is a critical part of the design as the catalyst fines must separate from the flue gas and fall down the swirl tube while the flue gas exits out the top of the swirl tube. The effect of barrel length and diameter is illustrated in Figure 4.

The collection efficiency of a commercial TSS installation as a function of particle size can be used to illustrate the effectiveness of the improved TSS design. The collection efficiency is determined by analyzing the particle size distributions and concentration at the TSS inlet and at the outlet. At the TSS inlet, the cumulative particle size distribution shows approximately 50% less than 10 microns and 35% less than 2 microns.

Therefore, there is a significant percentage of coarse particles (greater than 10 microns) and also fine particulate (less than 2 microns). Figure  5 shows the data from this installation. The graph shows the fractional (bi-modal) distribution and the cumulative distribution.

Downstream the TSS, essentially all of the particulates above 5 microns have been removed and even at 2 microns, much of the particles have been removed, as approximately 90% of the remaining particulate is less than 2 microns. Excellent performance has been achieved, leaving only the very small and fine particulate to escape from the TSS. See Figure  6 where the coarse ‘peak’ has disappeared.

This can be made clearer by analysing the grade efficiency, which is calculated from the particle size distributions and mass flows at the TSS inlet and outlet. This is illustrated in Figure  7, where for comparison also data from an older, poor performing unit are shown.

A characteristic of a grade efficiency curve is the 50 % separation point, or cutpoint. For a good operating TSS this is around 2 µm, whereas for the poor operating unit this is as much as 
10 µm.

Treatment of the TSS Underflow
In order to be able to effectively and continuously achieve acceptable air emissions standards it is necessary to effectively treat the underflow from the TSS. This is achieved with a fourth stage separator (FSS). The type of design that is used for the FSS can dramatically affect the overall emissions performance. Even though the amount of flue gas flow from the FSS is very low, typically 3% of the total flue gas flow, due to the fact that it is laden with all of the particulate from the TSS, effective treatment of this stream must be achieved. Figure  8 illustrates the dramatic difference that can be seen in overall emissions by varying the FSS design. If we use 50 mg/Nm3 total stack emissions as a benchmark to indicate that the stack emissions are comfortably below the required emission levels, several interesting things can be learned from this figure. First, if the inlet dust loading to the TSS is low (less than 200 mg/Nm3), we can see that it is possible to achieve the desired outlet emission rate with a 4th stage single cyclone, (designed by Shell). However, at higher dust loads, the only way to ensure reaching the desired emission levels is with the use of a hot gas filter.
Current Rating :  4

Add your rating:

Your rate: 1 2 3 4 5