Feb-2025

FCC unit stripper design and troubleshooting

Efficient removal of hydrocarbons from spent catalyst depends on multiple variables and is influenced by stripper design, as shown in the following discussion.

Warren Letzsch
FCC Consultant

Article Summary

The purpose of the fluid catalytic cracker (FCC) stripper is to remove hydrocarbons from the spent catalyst before it enters the regenerator or the catalyst transfer line, which takes the spent catalyst to the regenerator. The catalyst leaves the reactor and flows as a mixture of spent catalyst, hydrocarbons and dispersion steam. Steam is injected into the bottom of the stripper to push the hydrocarbons that are in the interstitial space between the catalyst particles back into the reactor, where they are recovered in the FCC gas plant. Hydrocarbons can also be desorbed from the surface of the catalyst and some of the pores. The stripper also promotes further cracking reactions, both thermal and catalytic.

The void space in the stripper can be determined from the density of the catalyst bed, the skeletal density of the catalyst, and the total pore volume (PV) of the catalyst. If a cubic foot of catalyst in the stripper is considered, and the weight of the gases is assumed to be negligible compared to the weight of the catalyst, then:

Catalyst density = Volume of catalyst x Skeletal density          (1)

Volume of the catalyst pores = Weight of the catalyst x Total pore volume          (2)

If the catalyst is not fluidised, then Equation 1 becomes:

Catalyst apparent bulk density = Volume of catalyst x
Skeletal density          (3)

The skeletal density is the density of the solid portion of the catalyst (see Figure 1). Therefore, the density of the catalyst divided by the skeletal density is the fraction of solids in a cubic foot, and the rest is the total void space. From Equation 2, the total catalyst pore volume is calculated, and subtracting from the total void volume gives the interstitial volume.

As an example, a catalyst with a density of 45 lb/ft3 and a skeletal density of 150 lb/ft3 has a total void of 70%. The total pore volume is 0.30 gm/cm3, yielding 0.485 ft3 of pore volume and 0.215 ft3 of interstitial volume.

For every 1,000 lb of catalyst circulated, the void space is:

1,000 lb x 0.70 void space                                                                      
45 lb/ft3

Steam is added to displace the hydrocarbons in the stripper. A pound-mole of steam has a volume of 379 SCF. If the stripper is at 1,000ºF and 22 psig, the volume becomes 425.5 ACF. Each pound of steam would have a volume of 23.6 ACF. These calculations can be made for any catalyst pore volume and density.

A stripper is really a series of mixing stages where the catalyst and hydrocarbons flow down. A tracer such as helium injected into the top of the stripper can be measured as the percentage going out the top and bottom (HO and Hi). This is a measure of mixing efficiency but does not reflect the desorption of the hydrocarbons from the catalyst.

A model of the stripper was constructed and showed that adding all the steam to the bottom of the stripper gave better performance than splitting the steam with 50% to the bottom and 50% added at the middle. Perfect mixing is assumed for each stage. The overall efficiency (Eo) is:

Eo = (1- (Heo/Hei))                       (4)

while each stage has an efficiency of:

Es = (1-Eo)1/n / 100                     (5)

Stripper performance can be calculated from the stage efficiency and the number of stages from Table 1. Eight stages should give excellent performance and can be installed in most applications. A graphical representation of this is shown in Figure 2. It suggests that ideally eight stages with a steam volume of two times the emulsion phase should be adequate.

In gasoil applications, the 2 lb/1,000 lb of catalyst circulated criteria works for disc and doughnut strippers operating within their design guidelines. These were 2-3 lb steam/1000 lb catalyst circulated, a catalyst flux rate of 600-900 lb catalyst/ (ft² minute), and a catalyst residence time of 60-90 seconds.

Stripping tests in the laboratory for benzene and slurry at 500ºC (932ºF) are shown in Figure 3. Both streams are aromatic and represent the hardest to desorb hydrocarbons. Benzene would be essentially inert in the stripper, while the slurry would thermally crack at the long residence times of the catalyst stripper. Having a residence time longer than 90 seconds might give a lower hydrogen in coke but would add to the wet gas compressor load. The dry gas produced is high in hydrogen content and lowers the molecular weight of the gas going to the wet gas compressor.

Higher reactor temperatures increase the desorption rate and reduce the time needed to effectively strip the catalyst. Reaction mix sampling can be used to assess the stripper performance and the amount of cracking occurring downstream of the riser reactor. Cracking in the stripper will increase the delta coke and raise the regenerator temperature.

The yield and operating changes used to calculate the typical impact of a stripper revamp are shown in Table 2. A higher regenerator temperature lowers the catalyst/oil ratio and conversion. Gasoline declines, but decant oil (DO) and dry gas both increase. The slightly lower coke yield is due to the higher heat content (hydrogen) of the coke.

These changes are significant but not large enough to change out a disc and doughnut stripper with a larger diameter to lower the flux rate. Most strippers operating with more than 7 wt % hydrogen in coke are well above the design catalyst flux rates. The pinch points in the disc and doughnut design cause high localised catalyst velocities that impede the rise of small bubbles. They either coalesce into larger bubbles or get swept down the stripper to the catalyst exit. Individual bubbles rise according to their size, as shown in Table 3. Higher values are observed with clusters of bubbles. Replacing the shell of the stripper is very expensive and results in a project having a three- to four-year payback. These payback times seldom get funding. If the cracker is operating at a catalyst circulation limit, a revamp should be considered.

Packings have been found to improve the mixing of the catalyst and steam and run with a constant catalyst velocity based on the superficial velocity of the vessel. An open area of about 96% allows this to happen. High flux rates are well below the rise velocity of a one-inch bubble, and the stripper will perform hydrodynamically. Residence times of the catalyst will be shorter at the higher flux rates, so a higher bed level with heavier feeds may be necessary. Other variables that can be manipulated are the reactor temperature, the pore volume, and the pore structure of the catalyst. More dispersion steam might also be tried.