At the National Petroleum Refiners Association's question and answer session on refining and petrochemical technology, engineers and technical specialists from around the world gather each year to exchange experience and information on refining and petrochemical issues.
Fluid catalytic cracking (FCC) catalysts were a topic of great interest at the most recent NPRA Q&A session, held Oct. 11-13, 1994, in Washington, D.C. The discussions of FCC catalysts included questions about:
- Reduction of olefins in FCC naphtha
- Tolerance of FCC catalysts to oxygen enrichment
- Use of mild hydrocracking catalyst in an FCC feed hydrotreater.
At this renowned meeting, a panel of industry representatives answers presubmitted questions. Moderator and NPRA technical director Terrence S. Higgins then invites audience members to respond or ask additional questions on the subjects under discussion.
OLEFINS IN FCC NAPHTHA
How can olefins be reduced in FCC naphtha to meet reformulated gasoline standards?
MENEGAZ:
The typical FCC naphtha in the U.S. is 25-40 LV % olefins. We envision refiners will meet the reformulated gasoline (RFG) standards for naphtha olefins via one of three approaches. The first is to minimize naphtha olefin formation, similar to how West Coast refiners currently manage bromine number. This would include the following:
- Use a moderate riser temperature.
- Maximize the catalyst-to-oil ratio to recrack naphtha olefins.
- Increase the catalyst rare earth content to maximize hydrogen transfer reactions which saturate olefins.
- Minimize the catalyst matrix content consistent with bottoms upgrading goals.
- Increase fresh catalyst additions to promote overcracking and higher hydrogen transfer. A higher proportion of fresh catalyst in inventory will increase the equilibrium unit cell size which, in turn, promotes olefin saturation.
- Increase hydrodesulfurization severity to increase hydrogen availability for hydrogen transfer reactions and to compensate for the higher delta coke associated with increasing catalyst activity and rare earth content.
- Use ZMS-5 to crack naphtha olefins. This effect is most pronounced on C7 olefins and higher. This may slightly increase C5 and C6 olefins.
The second route is to separate out the C5 olefins and send them to an alkylation or tertiary amyl methyl ether (TAME) unit. The C5 olefins are about a third of the naphtha olefin content. Acid consumption in alkylating C5s is minimized by making a sharp cut in the depentanizer to avoid cyclopentenes and by selective hydrogenation of the dienes. However, we envision that many refiners will actually increase naphtha olefins as a side effect of maximizing C4 olefins for the alkylation and methyl tertiary butyl ether (MTBE) feeds.
The third route is to hydrotreat the FCC naphtha, which reduces olefin content at the expense of lower octane an higher vapor pressure. Pilot plant work on a West Coast naphtha indicated an (R+M)/2 loss of 5 numbers for the C5 cut and 12 numbers for both the C6 and C7 Cuts.
MCLEAN:
I would agree with a lot of what Mr. Menegaz said. I would re-emphasize that this can be optimized by use of a high rare earth, high hydrogen transfer, high activity catalyst, and achieving conversion through catalyst activity rather than reactor temperature or catalyst-to-oil ratio.
I would also extend the question and say that a more difficult problem exists today for refiners who are trying to maximize their light olefin yields while at the same time minimizing the gasoline olefins. The use of a ZSM-5 additive may in fact help in this situation since ZSM-5 will crack with the C6 and heavier olefins in the gasoline range down to materials in the C3 to C5 fraction, and this is particularly beneficial if you do have separate C5 olefin processing such as TAME or alkylation, as Mr. Menegaz discussed.
Also, when you do this, you get a dilution effect. That is, you are adding back more fractions of lower olefin and high octane components, such as alkylate and ethers, so you really have to look at this in terms of what the impact is on the total pool and not just what comes directly out of the FCC unit (FCCU).
ARMBRESTER:
We agree with most of the comments that have been made already. In addition, if you are increasing catalyst rare earth content, our research indicates that you can reduce olefin production possibly as much as 15%.
In terms of downstream processing, if you are going to separate out the C5 olefins and use them for TAME production, then the olefin content of the gasoline can be reduced even further if you isomerize that C5 olefin stream initially and then send it to the TAME unit.
Regarding hydrogenation or saturation of the olefins, the work that we have done indicates that if you go to complete saturation, you could drop the octane number by as much as 10 numbers.
CUNEO:
I agree with the comments about the operating variables in the cracking unit made by Mr. Menegaz. Our experience is very similar. One thing he did not mention was increasing catalyst-to-oil ratio, which can also help increase catalytic hydrogen transfer and reduce olefins.
On the issue of hydrotreating the olefins, yes, they can be fully saturated. If it is desired to hydrotreat the olefins or the light naphtha from the FCCU only for sulfur removal, this can be done in modern low-pressure, high-space-velocity units with much lower octane loss than quoted-on the order of around 2 numbers.
FCC OXYGEN
ENRICHMENT
In last year's NPRA Q&A Session, it was stated that oxygen enrichment necessitates increased catalyst additions. Are some catalysts more tolerant of oxygen enrichment than others? Has anyone developed a good test to determine which catalysts can better withstand the oxygen enrichment environment?
MCLEAN:
The use of oxygen enrichment will accelerate the catalyst deactivation rate due to a number of related effects. Bulk regenerator temperatures will increase due to the smaller nitrogen dilution effect. Catalyst particle temperatures will probably increase by even more than what is indicated by the bulk bed temperature, due to accelerated burning kinetics, although there is not any good way to directly measure this.
Since hydrothermal catalyst deactivation is caused by the presence of steam at high temperatures, the effect of concentrating the steam partial pressure formed from hydrocarbon burning with lower nitrogen dilution will also be present. The higher temperature and higher localized oxygen concentration will also accentuate vanadium mobility and, therefore, vanadium destruction of the zeolite.
All of these things occurring simultaneously will increase the catalyst deactivation rate. Commercial experience from one unit has shown deactivation rate increases by 40% with the use of oxygen enrichment. This increases the importance of activity and hydrothermal stability in catalyst selection.
High levels of both zeolite and active, stable matrix are employed to withstand these effects, without needing to go to excessive catalyst addition rates. If the base case catalyst is high zeolite-to-matrix ratio, it will probably be prudent to consider lowering it, since there is a limit to how much zeolite you can add, and higher matrix catalysts will hold up to the severe environment with more stability.
Concerning test methods, the discussion of how to properly steam deactivate FCC catalysts in the laboratory has been a source of many divergent opinions for a long period of time. There are probably as many opinions as there are companies represented in the room today.
Certainly, for the case with oxygen enrichment, a more severe steaming treatment will be required to simulate what the regenerator operation will do, with some combination of higher temperature, time, and/or steam partial pressure.
It is always a good practice to make sure the deactivation severity used in testing matches reasonably well in terms of microactivity test activity, zeolite and matrix properties and surface area retentions, with the known base catalyst in the commercial unit. We have, in fact, worked with several oil company research and development laboratories which have adjusted their deactivation conditions when they are doing evaluations based on units which do use oxygen enrichment.
SAYLES:
I agree with Mr. McLean's comments. I would like to add that the reason we would consider oxygen injection is for additional carbon burning. Ordinarily, high-coke units would be processing a high-metals feed.
During the investigation of oxygen enrichment, I found the increased likelihood of deactivating the catalyst due to increasing nickel and vanadium mobility interesting. Since higher regenerator temperatures are likely, the catalyst will further deactivate. Care should be taken to minimize these combined effects.
KENT DAVIS (GRACE DAVISON):
We have worked with a refiner in trying to determine an appropriate deactivation method to model FCCUs that experience high catalyst particle temperatures in the regenerator. Modifications to a standard steam-deactivation protocol include changing the oxygen partial pressure and both the time and temperature of the deactivation.
The modified techniques do seem to be effective in identifying catalysts that have the highest activity and/or activity retention after the more severe deactivation. The downside, however, is that the protocols ignore the characteristics of the catalyst that are contributing high particle temperatures. These include inherent selectivity and physical traits such as catalytic coke make and catalyst strippability.
RANDALL HULL (BOC GASES):
We have a customer who has been using oxygen continuously in the regenerator for over 10 years at fairly high levels-24 to 26% enrichment of the air. The leverage effect of oxygen addition is fairly great. For example, at 2% enrichment (21 to 23 mole %) you get about a 10% increase in coke burning capacity.
This particular customer is operating in full combustion, and has a full overhead flue gas handling system. The refiner was under a severe velocity limitation, and the additional oxygen was backing out about double the volume of air for each volume of oxygen added to the regenerator. Although the customer has not commented on catalyst deactivation per se, they have seen less afterburn as a result of the reduced velocities. They have also seen improved yields because of lower coke-on-regenerated-catalyst values and less catalyst fouling of the overhead system.
Regarding the temperature increase, which directionally would lead to deactivation: If you do not change any of your other operating conditions on your FCCU, adding oxygen will definitely change the heat balance. We have seen roughly an increase of 5-10 F. in the dense bed temperature, for every 1% enrichment level above 21 %. Refiners typically readjust the heat balance by reducing feed preheat or, in some cases, if possible, increasing the catalyst-to-oil ratios.
Also, if the refiner has additional capacity in the vapor recovery wet gas system, we have seen refiners increase their recycle of light cycle oils and naphthas to, in effect, take care of the additional heat removal required from the regenerator.
FRANK ELVIN (COASTAL CATALYST TECHNOLOGY INC.):
In the 1970s, a lot of the regenerator designs went from 10 to 35 psig. This resulted in doubling the oxygen partial pressure of the inlet air going into the regenerator. This did not result in additional hydrothermal deactivation, because the regenerator temperatures were kept pretty much the same.
Similarly, if you double your oxygen partial pressure due to oxygen enrichment and maintain your regenerator temperatures constant, you should not see additional hydrothermal deactivation.
In a well designed regenerator, the oxygen content of the flue gas and the oxygen content in the regenerator gas are at the point of spent catalyst return, less than 3% oxygen, whether you have 20% oxygen in the inlet air or you have 40% oxygen in the inlet air.
I have experience with a Gulf Coast FCCU that was using between 500 and 1,000 tons/day of oxygen, which was approximately 20,000 scfm. We enriched the air from 21 to 40%, and maintained temperature with heat removal in the regenerator, and did not see deactivation as a result of oxygen enrichment.
WARREN S. LETZSCH (STONE & WEBSTER ENGINEERING CORP.):
How you contact the catalyst and the air is going to be very important. If you have countercurrent regeneration, the higher oxygen concentration and the highly coked up catalyst particles are really not going to see one another.
If the design is a sweep air system, spent catalyst is coming right out of the stripper, and using the air to lift the catalyst back up to the regenerator. If that is where the oxygen enrichment is done, you could see quite a difference. In a system with a high mix-zone temperature, I think you are going to tend to see more deactivation.
Typically, the regenerator temperature is going to be around 7-9 F. for every percent oxygen that you raise above 21%, and this, as was stated earlier, is because of the dilution effect (lack of nitrogen).
I cannot emphasize enough, and it was mentioned before, the idea of mixing the oxygen so that it is coming in uniformly. it should be pointed out that you can get into severe metallurgical problems on your unit if you take the oxygen concentrations too high. I hope nobody is really contemplating 40% oxygen; concentrations up to 27% are typically used in the unit,
FCC FEED HYDROTREATING
What has been the experience with using mild hydrocracking catalyst on an FCC feed hydrotreater to achieve incremental refinery production of low sulfur diesel? Mat are the limitations with this type of approach?
ARMBRESTER:
Ashland has no commercial experience with this processing scheme, but we have performed laboratory testing of mild hydrocracking of FCC feeds. These tests have shown a 5-10% yield of diesel, with less than 0.05% sulfur content.
Relatively severe conditions are required, with low liquid hourly space velocity being the most beneficial. The minimum reactor inlet temperature appears to be around 725 F. These high severity conditions will result in a relatively short catalyst life, and hydrogen consumption can be expected to be high, at about 500 scf/bbl or greater.
While this operation is helpful in increasing low sulfur diesel production, the hydrotreater reactor temperature that results in the optimum FCC operation may not necessarily be the highest temperature attainable, which could limit the amount of diesel produced. In addition, the stripper and overhead system may require modification to handle the increased volume of naphtha and diesel.
Mild hydrocracking of FCC feeds also results in an improved light cycle oil (LCO) product from the FCC unit, which could then be blended into the diesel hydrotreater feed streams. Although the LCO product has a lower sulfur content, the cetane number or aromatics level may still limit the amount of LCO that can be blended into the diesel pool.
D'AURIA:
We have been involved in about 10 mild hydrocracking projects, that were revamps of hydrotreaters, with the unconverted material being used as FCC feedstocks. All of these projects were driven by vacuum gas oil (VGO) conversion rather than production of low sulfur diesel. The cetane of the diesel product may limit this application, as cetane upgrading is not great in mild hydrocracking.
The converted product generally is less than 100 ppm sulfur, and the unconverted material to the FCC is in the range of 0.1-0.3 wt % sulfur.
Other limitations that must be considered when converting an existing hydrotreater are the available catalyst volume, the design pressure, and the reactor quench capability.
SAFA GEORGE (CRITERION CATALYST CO. L.P.):
Mild hydrocracking catalyst systems have been developed and are in commercial operations in a number of units to achieve incremental production of low sulfur diesel fuel. The overall conversion is dependent on crude type, catalyst system selected (alumina, amorphous silica alumina, or zeolite based), and operating conditions, including liquid hourly space velocity, hydrogen partial pressure, hydrogen availability, and operating temperature.
The resulting 650+ F. conversion varies between 20 and 50%. Mild hydrocracker limitations include desired cycle life, hydrogen availability, and FCC feed requirements.
Benefits include production of low sulfur diesel, with sulfur typically varying between less than 50 ppm at start of run, to less than 200 ppm at end of run and kerosine/jet stream for blending and upgrading of the FCC feed such as the reductions in sulfur, metals, Conradson carbon residue, nitrogen, and polynuclear aromatics.
PANKAJ H. DESAI (AKZO NOBEL CHEMICALS):
We have been selling mild hydrocracking catalysts for the last several years. This application has been mainly in Europe, where refiners are looking for incremental conversion.
As the gentleman from the panel mentioned, the main objective is to obtain incremental VGO conversion, and in all cases the diesel produced is indeed low in sulfur. In the early '80s, we published a paper, titled "Mild Hydrocracking of FCC Feeds Produces Yield Benefits in Mid-Distillates, Gasoline" (OGJ, July 22, 1985, p. 106).
PAUL VANCE (ACREON CATALYSTS):
Mild hydrocracking in an FCC feed hydrotreater is economically being done today to produce middle distillates at several locations using a catalyst marketed by Acreon. These are in both IFP licensed and non-IFP licensed units. The incremental severity to achieve low sulfur diesel is possible as long as feedstock and operating conditions will allow meeting product properties. Normally cetane is a primary consideration in blending mild hydrocrackate.
LEEN A. GERRITSEN (AKZO NOBEL CHEMICALS B.V.):
Akzo Nobel introduced amorphous mild hydrocracking (MHC) catalyst in the mid '80s. Zeolite-containing MHC catalysts were introduced a few years ago. In addition to extra conversion, the zeolite MHC catalysts give a better hydrodenitrogenation and a higher density reduction, resulting in improved FCC feedstock properties.
Zeolitic hydrocracking catalysts have been loaded in among two other MAK-MPHC units, a moderate pressure single-stage hydroprocessing process developed by Mobil Research & Development Corp. and licensed jointly by Mobil, Akzo Nobel, and the M.W. Kellogg Co.
I refer to presentations given during the Akzo Nobel Symposium in the Netherlands in June of 1994.
Copyright 1995 Oil & Gas Journal. All Rights Reserved.
