Does your FCC catalyst add up? Manfred Brown, Johnson Matthey Process Technologies

Transcription

1 Does your FCC catalyst add up? Manfred Brown, Johnson Matthey Process Technologies

2 Does your FCC catalyst add up? FCC catalysts and additives are generally considered the second greatest refinery operating expense after crude oil purchases. Manfred Brown, Johnson Matthey, UK, examines the variety of addition devices available on the market today, using real world operating data to assess their accuracy and reliability. Fluid catalytic cracker (FCC) catalysts and additives are generally considered the second greatest refinery operating expense after crude oil purchases. This fact is often quoted, but it is surprising how often their use is inefficient because of how they are added to the FCCU thanks to inadequate addition systems. This often leads to more catalyst being used than necessary; erratic control leads to cat crackers running at non-optimum activity levels, reduced throughputs and inferior product yields. Poor control of additives additions can lead to a number of expensive consequences such as loss of compliance in SO X and NO X emissions or reduced valuable LPG olefins yields. The cost of poorly controlled additions is considerable and erodes refinery operating margins. This article examines the variety of addition devices available on the market today and uses real world operating data to assess how accurate and reliable they really are. Introduction to addition systems The importance of steady catalyst activity at the optimum level in an FCC cannot be understated. It determines the unit s activity and product slate, enhances gasoline, diesel or LPG olefins yields and, ultimately, whether the FCC is profitable or not. Steady activity can only be attained with continuous, controlled fresh catalyst additions. If the addition equipment is unreliable or adds in batches rather than continuously, there will always be periods when the FCC is not at its most profitable. Early designs of FCCs paid scant regard to the provision of catalyst additions. There would be simply a pipe with aeration points running from the fresh cat hopper to the FCC regenerator. Additions were made on a shift by shift basis, the amount being controlled by observation of the regenerator catalyst level. Some units were controlled semi automatically where the dosing valves were opened with timers, but these were prone to blockages, valve failure or, at best, variable addition rates. The author recalls instructions written for such a system that blocked regularly. The solution was to rap sharply on the pipe with a non-sparking implement. There were also catalyst feeders, which were essentially rotary valves. These were not much better than the pipe design and also frequently broke down. None of these early systems gave reliable additions at the amounts required, which typically varies from less than a tonne to tens of tonnes per day. The shot pot The first attempts at good catalyst addition control were based on the shot pot design. Most FCCs had been provided with a manual shot pot with which to add a few kilograms of CO promoter. These consisted of a funnel type vessel with a capacity of a few litres. The funnel was filled with a bag of promoter and it was then blown into the regenerator to control afterburn. Automated versions of these were built (similar to the layout shown in Figure 1) and used for fresh catalyst additions. The shot pot consisted of a small pressure vessel, typically with a catalyst capacity of kg, mounted on or suspended from load cells. The unit was installed underneath the fresh catalyst hopper. The load cells measured the weight of the vessel and its contents. A controller sequenced the associated valves to depressurise the vessel, fill from the fresh cat hopper above, pressurise the vessel then blow the contents into the FCC regenerator. In order for the load cells to weigh the vessel accurately, all connections to it must be through flexible joints, typically rubber. Usually butterfly valves were used for catalyst control and small gate or ball valves for air/vent. The inherent weakness of these systems is the constant and frequent pressure cycling leading to failure of the rubber joints and erosion/ sticking/passing of the valves. After the first year or so of operation, these units usually require frequent maintenance and rarely add at reliable rates. Many variations of this basic design 2

3 Figure 1. Shot pot design Figure 2. INTERCAT JM TM AAS have been tried with the addition of vacuum systems to allow refilling from tote bins, additional vessels to hold bulk amounts, multiple source designs, better valves, etc. Reliability has improved in recent years but they still suffer from mechanical failure, an inevitable consequence of frequent pressure cycling. Another issue with these units is that they rely on summation of the weights of each shot to calculate daily additions. Thus any error in weighing is compounded by the number of shots in a day. Additive addition system design In the late 1980s, Intercat, Inc. introduced a new design of loader for the dosing of additives to the FCC. These used a larger vessel than the shot pot and dosed directly from it whilst maintaining constant pressure. The current design is shown in Figure 2. The vessel has a volume of 50 ft 3 (1.4 m 3 ) and a capacity of 1000 kg of additive. Additions are in the region of kg/d. For higher additions, larger units are available. The whole unit is supported by three load cells so the number of flexible hoses is minimised to two: air supply and product discharge; high quality 1 in. SS hoses are used. In normal operation, a steady flow of carrier air flows through the unit piping, to the FCC, controlled by a 1 in. globe valve. The vessel is kept pressurised at approximately 4 barg, depending on the back pressure from the regenerator. Below the vessel are two valves: a ball valve and an Everlasting Valve. The former is kept open and only closed when abnormal conditions arise such as unexpected weight loss. The second valve controls product flow. This valve is specially designed for Intercat, Inc. by the Everlasting Valve Company and is of a rotating/shearing disk design. It has many proprietary modifications and has been found to be very reliable in this service, typically operating for 10 years or more without maintenance. The valve is opened by the controller (IMS) for a set time, usually approximately 30 seconds and the weight drop noted. The controller then calculates how many such shots are required that day to meet the target addition rate and thus sets the time to wait until the next shot. For example, if the addition rate is to be 240 kg/d and the last shot was 5 kg, then 48 shots are required per day so the loader will wait 30 minutes before the next shot. This calculation is repeated after every shot so variations in back pressure from the regenerator or other unit events will always be accounted for. The fundamental advantages with this design are its simplicity and few pressure cycles. There are only two flex hoses and the only valve that comes into frequent contact with catalyst is the Everlasting Valve. Reinforced seat ball valves are used for all other valves and are on/off only (except for the carrier 3

4 Figure 3. INTERCAT JM AAS with day hopper air globe valve). Whereas a shot pot is expected to pressure cycle 50 or more times a day, the INTERCAT JM design only needs refilling every few days. INTERCAT JM fresh catalyst addition system design Following the success of the INTERCAT JM AAS, systems were designed for fresh catalyst. These are of two varieties: a large version of the AAS and a day hopper design. The former is simply a large AAS; units with capacities up to 120 t are in use. These are refilled directly from bulk trucks in much the same way as fresh catalyst hoppers. The day hopper uses a vessel of 5 or 10 t capacity as an addition system, automatically refilled from the FCC s fresh catalyst hopper. A typical layout is shown in Figure 3. The day hopper operates the same way as the AAS but, when empty, the IMS controller automatically refills the day hopper from the fresh catalyst hopper. These have been very successful in a number of locations, again because of their simplicity and the refill cycles being kept to once a day or so; although there are units successfully in operation loading 25 tpd of fresh catalyst, refilling the day hopper five or six times a day. For comparison, a 50 kg shot pot would have to cycle 500 times/d at these rates. Multi compartment loaders A recent innovation is the multi compartment loader. This is a 200 ft 3 (5.7 m 3 ) AAS vessel that has baffles dividing it into (usually) three compartments, one of 2 t capacity and two of 1 t capacity. These operate in the same way as the AAS above but each compartment has its own outlet, with valves and refilling line. Thus the loader can add up to three different products, be they fresh catalyst or additives. As there is only a single vessel, plot space is far less than three separate loaders and the controller ensures no conflicts can occur between the Figure 4. INTERCAT JM multi compartment loader (MC3) 4

5 additions of the three materials. These units are also being used as fresh catalyst addition systems with the large compartment being auto refilled (as with the CAS above) and the other two used for additives. Precision data Just how accurate are these addition systems? Most units report a high level of accuracy on their screens, but are these numbers real? The only way to check is by reconciling the reported addition history from the addition system with the known (accurately weighed) deliveries of catalyst to the refinery. A study was therefore carried out of a number of addition systems, covering many different designs, and this has revealed some surprising data on the precision and reliability of these systems. Shot pot data Comparing the amounts of material delivered to the refinery with the sum of additions to the FCC regenerator, Table 1 shows the errors of shot pot loaders at eight different refineries. The average error of the examples above is a disappointing 7.8%. There is also a high degree of variability in these data, showing that the true accuracy of these small devices is highly unpredictable and is fundamentally limited by the antiquated design. Single AAS data The data from three single INTERCAT JM loaders were looked at closely to check overall reliability and precision. Different sized units were chosen to see if large loaders are less precise than smaller ones: a 50 ft 3 (1 t), a 500 ft 3 (10 t) and a 1100 ft 3 (25 t). Example 1: 50 ft 3 AAS (1 t capacity) Details: Loaded from tote bins. Adding ZSM-5 additive at ~150 lb/d (68 kg). Data were examined from more than 2.5 years (1013 days) operation. Figure 5 shows the data from this unit. The triangles indicate refills from a tote bin and the diamonds the end of day weight of additive in the loader. When one compares the actual amount of catalyst added over a 2.5 year period with the amount reported by the addition system, the following results are achieved: Total reported additions to FCC = 104,347 lbs. Total refills by tote bin = 104,997 lbs. Loader accuracy = 99.4%. Loader error = 0.62%. In other words, the true amount of catalyst added by this addition system over a 2.5 year period was within 0.6% of the reported amount. This is an impressive result and testament to the quality and reliability of this style of addition system. Single shot pots Error Refinery A 3.6% Refinery B 24.9% Refinery C 5.3% Refinery D 0.7% Multi source shot pots Individual product error Overall error Refinery E 7.9% 5.6% 4.9% Refinery F 18.0% 12.9% 7.0% Refinery G 3.9% 3.1% 1.6% Refinery H 6.5% 6.6% 6.4% 10.0% Table 1. Shot pot errors Figure 5. Example 1: 50 ft 3 refills and additions 5

6 In other words, the true amount of catalyst added by this addition system over a 1.5 year period was within 0.2% of the reported amount. Once again this is a remarkable result, reinforcing the fundamental accuracy of this design of addition system. Example 3: 1100 ft 3 AAS (25 t capacity) Details: Loaded by trucks. Figure 6. Fresh catalyst addition deviations on 8 tpd Example 2: 500 ft 3 AAS (10 t capacity) Details: Loaded from bulk trucks. Adding SO X reduction additive at ~1500 lb/d (680 kg). Data were examined from more than 1.5 years (604 days) operation. Refill data are from the supplier shipment amounts and are therefore guaranteed to be accurate. When one compares the actual amount of catalyst added over a 1.5 year period with the amount reported by the addition system, the following results are achieved: Total reported additions to FCC = 683,085 lbs. Total catalyst deliveries from bulk trucks = 681,500 lbs. Loader accuracy = 100.2%. Loader error = -0.23%. Adding SO X additive at ~1000 lb/d. Data were examined from more than 1.5 years (588 days) operation. Once again, refill data are from the supplier shipment amounts and are therefore guaranteed to be accurate. Comparing totals, one can conclude the following: Total reported additions to FCC = 486,505 lbs. Total catalyst deliveries from bulk trucks = 493,105 lbs. Loader accuracy = 98.7%. Loader error = 1.3%. In other words, the true amount of catalyst added by this addition system over a 1.5 year period was within 1.3% of the reported amount. Once again this is an impressive level of accuracy. Figure 7. Equilibrium catalyst MAT Multi compartment data Multi compartment loader units can show similar low errors to conventional AAS. Data from two MC3 loaders that are using the large 6

7 compartment for fresh catalyst have recently been reviewed. Both show excellent accuracy for fresh catalyst, 0.4% and 0.3% errors respectively, and very good accuracy for additives, 1.4% and 0.8% respectively. Fresh catalyst loader CAS shows the same accuracy and precision as AAS. A very good example was previously published for a BP unit. 1 This was a 70 m 3 system, designed to add up to 12 tpd with occasional additions at 30 tpd. Target daily additions at 8 tpd gave the deviations shown in Figure 6. These show addition accuracy within 0.2%. This resulted in far better control of unit equilibrium catalyst activity (MAT) as shown in Figure 7. This superior control of MAT permitted inferior feeds to be run and/ or higher MATs to be targeted. This project paid for itself in less than a year. Conclusion Control of fresh catalyst and additive additions to an FCC unit is critical for unit throughput, yields, product slate, environmental compliance and, ultimately, profitability. Simple pipe valve and shot pot systems are inaccurate, unreliable, require high maintenance and will reduce the cracking margins on the operating FCC. When running, the error of shot pots is almost 8% when reconciled against catalyst deliveries to the refinery. Would this degree of inaccuracy be acceptable for other major refinery expenditures such as raw oil or energy? Larger INTERCAT JM additive and fresh catalyst addition systems have been shown to consistently operate at less than 2% error, usually well below 1%. This is true of small, large and multi compartment systems. This results in higher in unit catalyst activity, lowers maintenance requirements, improves yields, allows higher throughput and the processing of cheaper feeds. References BROWN, M., and CAMERON, A., "A Fresh Approach", March 2006, Hydrocarbon Engineering. Information contained in this publication or as may be otherwise supplied by Johnson Matthey is believed to be accurate and correct at the time of publication and is given in good faith. JOHNSON MATTHEY GIVES NO WARRANTIES, EXPRESS OR IMPLIED, REGARDING MERCHANTABILITY OR FITNESS OF ANY PRODUCT FOR A PARTICULAR PURPOSE. Each User must determine independently for itself whether or not the Products will suitably meet its requirements. Johnson Matthey accepts no liability for loss or damage resulting from reliance on this information other than damage resulting from the death or personal injury caused by Johnson Matthey s negligence or by a defective product. Freedom under Patent, Copyright and Designs cannot be assumed. 7

8 For further information on Johnson Matthey, please contact your local sales representative of visit our website. INTERCAT is a trademark of the Johnson Matthey group of companies.