Deactivation of FCC catalysts
A review of technology and expertise to rapidly identify FCC unit catalyst contaminates, hydrothermal factors, and other obstacles to catalyst performance
Warren S Letzsch Warren Letzsch Consulting PC
C atalytic cracking replaced thermal cracking for con- verting heavier molecules in crude oil to smaller, more desirable products such as gasoline and diesel. The heart of the process is the catalyst, which is responsible for the conversion and yields generated during the operation. Catalytic cracking follows a carbenium ion cracking mech- anism rather than the less desirable free radical system of thermal cracking. Both reaction mechanisms compete with the conditions found in the cat cracker. The process itself has evolved around the catalysts being developed and used over time. These catalysts deactivate over time in several ways. While some loss of performance is inevitable, it is important to minimise any abnormal loss of activity and/or selectivity. Hydrothermal deactivation There are several ways catalysts could lose activity in the FCC unit due to high temperatures in the presence of steam. These are: • Elevated individual particle temperatures • A higher regenerator mix temperature • Repeated reduction/oxidation cycles. All three are valid mechanisms and do contribute to deactivation. Fluid catalytic cracking (FCC) was introduced in 1942 and became the preferred design due to its ease in circulating catalyst and removing heat. Synthetic silica-alumina cata- lysts were developed for the FCC process, and these early amorphous catalysts had an activity proportional to their surface area. These products started with surface areas ranging from 400 to 550 m²/gm and equated with surface areas of 125 to 180 m²/gm. Laboratory tests showed that high temperatures and the presence of steam would cause the loss of activity and surface area. These conditions are present in the regenerator of the catalytic cracker. Shell 1 conducted experiments with tagged catalyst and found that 80% of the surface area the catalyst lost in the unit occurred in the first 24 hours after the catalyst was added to the unit. Using the regenerator conditions, they found that the loss of surface area could not be dupli- cated in the laboratory. The catalyst was apparently seeing higher temperatures than those measured. To achieve the observed deactivation, high hydrocarbon and oxygen con- centrations were required. Fresh catalyst, with its high sur- face area, could absorb a lot of hydrocarbons in the small pores, resulting in coke concentrations of up to 5 wt% on freshly added catalyst particles.
Oxygen concentrations of 20% are present where air is introduced to the regenerator or in the spent catalyst trans- fer lines that use air to convey the catalyst to the regenera- tor. Figure 1 shows the calculated temperatures more than the spent catalyst/air (oxygen) mix temperature with vary- ing combinations of coke and oxygen. These temperatures (1,550-1,600˚F) are clearly high enough to cause catalyst deactivation. The catalyst stripper plays a role here, and poor strip- ping will lead to higher regenerator temperatures. Design parameters of the stripper include catalyst residence time, temperature, steam rate, and hydrodynamics. Short- circuiting of the catalyst needs to be avoided, while good mixing of the steam and spent catalyst is essential. The burning takes place in milliseconds and is repeated with every cycle through the unit. As the surface area decreases, the rate of deactivation declines since the par- ticles will no longer absorb extra hydrocarbons. These amorphous catalysts were replaced with catalysts con- taining zeolites in 1962. Within a decade, they were used in every unit and all the FCC reactors were modified to operate with shorter catalyst residence times. The older amorphous catalysts were so inactive that reactors were typically designed for two weighted hourly space velocity
450
400
5% Carbon 20% Oxygen
350
300
10% Carbon 5% Oxygen
250
200
5% Carbon 5% Oxygen
150
0
6
10
0
2
4
8
Time (milliseconds)
Figure 1 Particle temperature rise
77
PTQ Q3 2024
www.digitalrefining.com
Powered by FlippingBook