Fluid catalytic cracking (FCC) was the focus of extensive discussion at the 2016 American Fuel and Petrochemical Manufacturers Q&A and Technology Forum, Sept. 25-28, in Baltimore, where refiners addressed issues related both to using FCC units to maximize gasoline octane and maintaining their operational reliability and safety during catalyst changeovers.
This annual meeting addresses real problems and issues refiners face at their plants and provides an opportunity for members to sort through potential solutions in a discussion with panelists and other attendees.
Part 1 in the series highlighted discussion surrounding gasoline processes (OGJ, Aug. 7, 2017, p. 54). The second installment addressed hydroprocessing operations (OGJ, Sept. 5, 2017, p. 76).
The forum included five industry-expert panelists from refining companies and other technology specialists responding to selected questions and then engaging attendees in discussion of the relevant issues (see accompanying box).
Processing
As distillate demand has decreased, current economics favor maximizing gasoline and octane. What operating and catalyst changes do you recommend for increasing octane barrels?
Bougrat As the question suggests, there's a shift in the current market towards greater gasoline octane. The topic of octane barrels has been studied for years, but Tier 3 regulations have added another layer of complexity to this issue since FCC naphtha contributes a large portion of the overall cetane pool for the refinery. About 90% of the refinery gasoline-pool sulfur is associated with FCC naphtha. The overall gasoline-sulfur removal process, generally balanced between post-treatment and pretreatment, tends to negatively impact octane. From a process standpoint, the different handles available within an FCC unit tend to pull the gasoline yield and gasoline octane in opposite directions. One process variable that affects gasoline octane is the reactor-outlet temperature. What we've observed is that for roughly every +20° F. in reactor temperature, there will be a research octane number (RON) boost of about 0.75-1.00. The heavier gasoline cut will experience the greatest octane reactions to the change in reactor temperature due to the corresponding increase in aromaticity. Therefore, the heavier portion of the heavy-cracked naphtha (HCN) cut will generally experience a much stronger octane response than the lighter light-cracked naphtha (LCN) cut. The reactor temperature is important within the range of 950-1000° F. Once you get above 1000° F., the gasoline octane response is mostly flat.
As expected, feed quality also generally plays a huge role in gasoline octane. Regarding the UOP K, or the overall paraffinicity or aromaticity of the feed, a decrease in the UOP K of just 0.1 will typically lead to an increase of 0.5 in gasoline RON. Straight-run feed will generally lead to greater gasoline octane than hydrotreated feed without accounting for gasoline hydrotreating.
In terms of the feed properties, you also want to look at the naphthene-to-paraffin ratio. This composition ratio can be estimated through a correlation involving the refractive index, molecular weight, and density of the feed. For every 0.1 increase in the ratio of naphthenes to paraffins, we generally anticipate a 1.0 increase in RON. Regarding light ends in the feed, the 650-minus components of the feed generally just go along for the ride. Furthermore, any naphtha present in the feed will go largely unconverted within the riser to yield lower octane properties than that of conventional cracked naphtha. Lastly, the hydrogen content of the feed, which should track directly with the UOP K, will also play a large role in the resulting gasoline octane and gasoline yield. An increase of 10% in the overall hydrogen content of the feed tends to increase gasoline yields by about 1 vol %, based on historical data.
Regarding the issue from a catalyst perspective, I will let the other, more experienced panelists comment on this topic. It's well documented that the use of Zeolite Socony Mobil 5 (ZSM-5) can help increase gasoline octane at the expense of gasoline yields. The shape-selective additives, in general, help selectively isomerize and crack some of the light, low-octane gasoline. Generated out of the gasoline-range material, LPG olefins usually help concentrate the remaining aromatics within the gasoline cut and help increase the gasoline octane. Other considerations include high catalyst activity, a greater focus on isomerization within your catalyst formulation, and reduced overall hydrogen transfer activity.
All the factors and process variables previously referenced, whether you're looking at the process or catalyst side, will have a much more pronounced effect in low-severity operation. If you haven't already done so, part of your overall octane-barrel strategy for the FCC complex should include upgrading the cracked-gasoline hydrotreating catalyst to industry-leading octane retention. Gasoline-sulfur reduction additives can also be pursued to shift some of the severity requirements away from the gasoline hydrotreater. Process configurations can also play a key role, often involving a delicate balance-here in the US market-between alky utilization and maximizing liquid yields. Adjusting the HCN cutpoints to balance out sulfur content with octane properties can also impact the bottom line. Some customers even have looked at isolating the low-octane mild-cracked naphtha (MCN) cut and routing it to an isomerization process to help boost the overall gasoline-pool octane.
Bezon I broke my answers into operational moves and catalytic moves. To be very broad, an increased catalyst-oil ratio should directionally push up both gasoline and octane. That can be done relatively easily with just a change in preheat temperature, if your regenerator can handle it. And like Luis said, an increased reactor temperature will help boost both gasoline octane and yield, assuming you're not overcracking and already are producing more dry gas.
Another operational change to consider is any change in API gravity. Again, the higher the API gravity, the higher your conversion will be. In this case, you're not sending as many liquid volume barrels to the cat, however, so it's probably not economical. In terms of catalytic moves, increased microactivity testing (MAT), smaller unit cell size, or increased matrix activity will help upgrade the bottoms. I'm sure the catalyst vendors will help reformulate catalyst if you need to make changes. I'll leave that part of the discussion to George.
Yaluris Octane-barrels can be increased by increasing either gasoline yield (without decreasing octane) or by increasing octane (without a large decrease in gasoline yield). The latter approach is, in most cases, the most effective; I'll focus on how to increase gasoline octane.
Operating options for raising gasoline octane include:
• Increasing riser temperature. This option will increase gasoline olefinicity and has the largest effect on increasing gasoline RON. If implemented, however, it will also increase conversion and dry gas. If the unit is operating in distillate-maximization mode, raising the riser temperature may not be a good option for you.
• Minimizing the amount of straight-run distillate doing to the FCC. If processing tight oils (crudes from hydraulic fracturing operations), straight-run distillate is quite paraffinic and can have a large negative impact on octane.
• Separating the light cat naphtha. This fraction is the most olefinic and highest in RON. If the capability exists (e.g., gasoline splitter), separating this fraction before the hydrotreater and processing it separately decreases the octane loss across the hydrotreater.
Catalytic options for increasing gasoline octane include decreasing the hydrogen transfer activity of the catalyst by lowering rare earth content and-or increasing catalyst accessibility and matrix activity. Implementing this option will increase gasoline octane and gasoline olefinicity and will affect the amount of gasoline and LPG production. Unless the catalyst is designed with sufficient matrix to maintain overall activity, decreasing rare-earth content can make the catalyst less active, requiring an increase in catalyst additives to maintain e-cat activity.
ZSM-5 additives are well known to increase octane. Indeed, they were originally commercialized as octane additives. They work by removing low-octane straight-chain olefins and paraffins from gasoline, concentrating the higher-octane components such as aromatics and multi-branched paraffins. Traditional ZSM-5 additives, however, will lower gasoline yields and raise LPG output.
Catalyst suppliers are also marketing catalysts and additives that, among other things, promise to help increase gasoline octane. Albemarle presented its technology for maximizing gasoline octane and C4 olefins at the 2014 Annual AFPM Meeting.1
Shackleford I just want to add a couple of comments to this discussion. George talked about riser temperature, catalyst-activity changes, and ZSM-5. It really comes down to what your unit is limited on. If it's dry gas, raising reactor operating temperature (ROT) isn't going to be the best way for you to increase octane because you'll hit your dry-gas limit. If you're at an LPG-constraint, then ZSM-5 won't be a good solution since it produces high LPG. You need to look at your unit's constraints and balance them with the best way to maximize octane barrels. Again, as you raise your ROT, watch out for overcracking. You don't want to crack more gasoline than is created. Also, look at how much you're sending to the alkylation unit vs. what you're getting back as high-octane alkylate. In other words, look at the overall refinery balance.
An FCC unit contributes to the gasoline pool through both cracked-gasoline production and C4 production, which is used to generate high-octane alkylate for blending. For optimizing octane barrels, there are four main operation variables: ROT, catalyst activity, catalyst selectivity (lower rare-earth content), and use of a light-olefins additive (ZSM-5). Due to differences in operating constraints, a best practice is to use a kinetic model such as FCC-SIM to determine the best operating variables and catalyst properties for the unit. The optimized profit for the unit is likely to be a combination of these variables up to the unit's constraints.
Jiushun I'd like to share our experience regarding how we eliminate light-cycle oil (LCO) to produce more gasoline in FCC units. In China, we're also facing this kind of market change, because the Chinese transportation system-our express train industry-has grown rapidly. The trains use electricity rather than internal diesel engines. Since the market demand for diesel has dropped sharply, we must convert diesel into gasoline. Because we want to increase the octane number for our gasoline pool, however, we have to operate our FCC units in a more severe mode, leaving our LCO quality poorer and poorer.
To address this, we've started mild hydrotreating LCO and then returning it to the cat cracker. The ejection point should be carefully designed to prevent direct blending with the common or conventional olefin stock. To accomplish this, we've redesigned our resid reactors to enable recycling of hydrotreated LCO at the cat cracker, which has allowed us to covert 100% of our LCO into gasoline. The cat yield, which is about 85% of recycled LCO, also can be converted into gasoline. At the same time, the octane number has improved to between 0.6-0.7. We currently have about 27 units working like this.
Yang My comment is about our experience with a client's naphtha splitter that splits naphtha into light, middle, and heavy cuts. The middle naphtha normally has a low octane number after blending the light and heavy naphtha cuts together. The splitter then sends the middle naphtha stream to the naphtha reformer. Using this process, in some cases, you can increase RON by 1.0, while in other cases, you can achieve an increase of about 0.7 in RON.
Fisher I believe George mentioned that ZSM-5 normally will shrink your FCC-gasoline yield, so your octane barrels will be reduced by using ZSM-5. To alleviate this phenomenon, you may want to consider using a ZSM-5 with a high silica-to-alumina ratio to increase the FCC gasoline octane while minimizing the FCC-gasoline yield loss if you're looking for an increase in FCC-octane barrels and not just an increase in FCC-gasoline octane.
Safety
Recent drone technology advancements have enabled refiners and contractors to improve the efficiency of maintenance and inspection activities. With this, how are your hot-work permits and general safety policies evolving to sustain adequate asset and personnel protection at all times? For instance, what additional safety permits or considerations would apply for drone use and aerial inspections?
Niccum In June 2016, the Federal Aviation Administration (FAA) announced rules for routine commercial use of small, unmanned aircraft systems (UAS), including drones. Summarizing the key points from the announcement, the new rule offers safety regulations for drones weighing less than 55 lb that are conducting nonhobbyist operations.
Importantly, flying a drone for commercial purposes no longer requires a pilot's license. The person flying a drone must be 16 years old and have a remote pilot certificate or be directly supervised by someone who has such a certificate. The regulations require pilots to keep the unmanned aircraft within the visual line of sight. Operations are allowed during daylight and during twilight if the drone has anti-collision lights. The new regulations also address altitude and speed restrictions as well as other operational limits. External load operations are allowed if the object carried is securely attached and doesn't adversely affect the controllability of the aircraft. The FAA is offering a process to waive some of these restrictions if an operator proves that the proposed flight will be conducted safely under a waiver.
C.H. Fenstermaker & Associates is active in the use of drones. The company interprets these rules to mean that the following checklist of safety items must be specifically applied prior to operation in refinery applications:
• The lead operator of the unit must be notified of drone operations, and those employees and contractors within the unit shall be notified of the intended drone operations at least 30 min before flight.
• As the drone is not intrinsically safe, hot work permits will be required. Of course, proper PPE (personal protective equipment) is required as well.
• Never fly the drone over active workers unaware of elevated drone inspection activities. Drones may not operate over other persons not directly participating in the operation who are not under a covered structure or inside a covered stationary vehicle.
I'll share an actual story of a plant's drone-safety incident. While looking at the flight screen, a drone operator heard the drone engines revving. He immediately looked up and witnessed the drone bank hard to one side and fall about 100 ft straight down. The drone barely missed a series of pipes and valves, crash-landing just a few feet away from an active laser-scanning crew. Both the refinery representative and drone operator looked at each other in disbelief. What happened? The refinery representative, who just happened to be watching the drone when the incident occurred, said that he witnessed a large bird attack the drone and knock it out of the sky. The moral of the story, then, is that safety is the first priority. You must assume the drone will crash, and you need to prepare the environment within the crash zone to minimize potential impacts.
Kasle One of the discussions we had during an emergency shutdown was around the possibility of sending a drone into the reactor to do cyclone inspections to see if the diplegs were plugged. So, it is something that people are discussing. We did not actually do it, but it is a topic that comes up more and more.
Niccum I would like to just address that point. The FAA is predicting a huge increase in the number of active drones, so their presence in the sky is going to become ubiquitous. You may remember when the government removed constraints on the use of global positioning systems (GPS). Today, low-cost drones the size of my hand can carry a camera and transmit data. The increase in the technology is so rapid that I wouldn't be surprised if lifting of constraints on drone usage happens, too, if not now, within the next few years.
Catalyst
What are your best practices for mitigating operational or performance risks during a catalyst changeover?
Yaluris Catalyst changes are common in FCCs because they can be implemented as soon as an operator decides it can achieve performance improvements by using a different catalyst. While the key issue surrounding a catalyst change typically involves the post-change determination of whether the new catalyst delivered on its promise of improved performance, there are potential operating and performance problems to consider that can occur during the change itself. Examples of operating risks to consider include:
• Circulation problems for certain unit designs with long and curved standpipes.
• Increased carbon on e-cat for units operating in partial burn.
• Increased formation of coke deposits in units processing resid.
• Increased catalyst losses, particularly for units without a wet-gas scrubber (WGS) or an electrostatic precipitator (ESP).
• Increased catalyst deposits, especially for units that have an expander.
• Poor catalyst activity, stability, and yields that can affect the unit economics, increase catalyst use to maintain activity, or push a unit towards an operating limit (e.g., wet-gas compressor limitation, high slurry flow, or an unexpected change in delta coke causing the regenerator to heat up, cool down, or run into an air limit).
The risk-mitigation plan during a catalyst changeover, by necessity, must be developed for the specific FCC unit. Each unit has its own constraints and operating limitations, leading to a unique list of risks to be mitigated. When considering a catalyst change, the refiner should consider taking the following steps:
• Carefully examine the most common constraints and operating problems the unit can encounter. A list of these should be created and shared with the catalyst supplier. Ask the catalyst supplier how the new catalyst will help avoid them.
• Confirm that the new catalyst is designed with the physical properties needed for problem-free operation of the unit. It isn't necessary for the new catalyst to have the same physical properties as the incumbent catalyst if its physical properties are sufficient for what the unit needs. Discuss any concerns with the catalyst supplier and ensure that they are addressed. Ask for the supplier's experience in units with similar design and-or downstream units (i.e., ESP, WGS, TSS, expanders). Consider asking for examples and references of catalysts successfully used in similar units to address both operating concerns and questions about the ability of the new catalyst to deliver on the promised performance. Be aware, however, that confidentiality concerns may limit how much detail regarding similar applications the catalyst supplier can disclose.
There has been considerable debate on how to select a catalyst so that it works in the unit as promised. It's beyond the scope of answering this question to address the issue of how to select a catalyst for a unit, but ample literature is available to consult on this matter.2-3
Regardless of how the catalyst is selected, it's important to analyze its in-unit performance carefully, using rigorous data analysis tools to confirm that the promised performance is achieved.
• Discuss with your supplier how the catalyst change will be monitored, what technical support is available, how performance and operating problems will be addressed, and how a reformulation can be implemented, if needed. Have a clear plan for how frequently the unit performance will be reviewed, including which data to send for analysis and how frequently that data should be sent.
Because of the availability of catalyst plants by all three major catalyst suppliers, the North American supply-chain length typically takes from a few days to as long as 3-4 weeks, depending on the transportation mode selected. While it's usually unnecessary to keep a safety stock of the incumbent catalyst on site, keeping a stock of the old catalyst is an option to consider, particularly when there's an environmental limitation that can't be violated for any length of time.
Thraen There are two situations to consider here: gradual FCC catalyst changeover and sudden changes, such as those that might occur during startup or following an upset if an outside supply of equilibrium FCC catalyst is brought into the refinery. In either case, a formal change-management process with appropriate reviews and signoffs is required before making any catalyst changes.
For gradual changes like those that would occur during a planned catalyst change, we rely on our catalyst-screening process and our ongoing FCC-performance monitoring process. Proposed new catalysts are reviewed, including both yield performance and physical properties. Catalyst yields for proposed new catalysts are generally confirmed using third-party, offsite pilot testing. Physical properties are included in the fresh catalyst data sheets that accompany each shipment. For sudden FCC catalyst changes, such as those that might occur if we need to replace a large percentage of our e-cat inventory, we use a similar but expedited process. The e-cat properties are reviewed, and the supply is chosen to match as best as possible the properties of our normal catalyst. The catalyst chosen will have, as close as possible, similar particle size, activity, metals, etc., as our base-catalyst inventory.
Bougrat UOP encourages establishing a solid baseline for incumbent operation. It's critical to establish a representative base case before initiating the catalyst turnover because, in the event something does go wrong, you'll have a representative case for comparison purposes. We'd also recommend carrying out the catalyst turnover at the usual catalyst-addition rates. Some refiners have run into unexpected issues when trying to expedite the turnover rate, which can lead to wet-gas compressor limitations or other adverse effects across the unit. It's also important to identify the potential impacts of the new catalyst formulation on the process and establish a contingency plan for each of the potential impacts. Also, capitalize on the expertise of your catalyst vendors because all of them provide good services and support. Finally, make sure to frequently request and consult certificates of analysis (COA) for the fresh-catalyst shipments to help identify any potential impacts on the process as early as possible.
Kasle I think those are all great points: knowing where your unit starts, having a plan to monitor the unit, and having a baseline. I'd also add the following to that list: making sure you know what the slurry particle-size distribution looks like over time; knowing the ESP particle-size distribution; and having a reasonable catalyst balance.
We might also do third-party pilot plant testing in the lab to mitigate risk or to help understand exactly what to expect for the yield performance. If we have some uncertainty about the physical properties of a new catalyst, we might do attrition testing at a third-party lab. There are a few labs where you can do that testing. If you're concerned about fluidization or if you have challenges in your standpipes, you can also do fluidization testing on the different catalyst samples. Monitor the unit performance as the catalyst changes out as well.
Have a plan to test the catalyst in the lab (either the vendor's or a third party's) on a frequent basis to gradually monitor the catalyst's performance during the changeout vs. waiting until the turnover is completed. Make sure you understand exactly how long your unit is going to take to turnover. If you are doing a staged trial, you will want to know how much time you'll need to turn over the unit to an effective changeover of the new catalyst. Collect data at that level so you can really compare the different trial periods or catalyst-additive concentrations. All units have some level of variability of performance, so you want to make those comparisons on an equivalent basis. The feed isn't going to be the same, so you need to either have a process model, or some other mechanism, through normal testing to allow you to normalize and make sense of those variations.
You also want to ensure that you work with the new supplier before you transition all the analytical work (i.e., e-cat, ESP fines, and slurry samples) to that new vendor. Maybe submit e-cat samples so you can compare the new supplier's e-cat analysis to your current e-cat analysis, just in case there is some sort of offset. The vendors don't use identical calibrations for their chemical analysis, so there might be an offset you weren't expecting. You want to get a nice baseline leading into your trial, so you'll know how to track your catalyst turnover without any surprises, such as thinking the rare-earth content has decreased when it hasn't. It's all just analytical differences.
References
1. Yaluris, G., and Kramer, A., "Take ACTION to Maximize Distillate and Alky Feed from Your FCC Unit," 2014 AFPM Annual Meeting, Mar. 23-25, 2014, Orlando, Fla.
2. McLean, J., Alvarez, W., and Chaugule, S., "Catalyst Selection Techniques," FCC session, AFPM Q&A and Technology Forum, Oct. 7-9, 2013, Dallas.
3. McLean, J., "The History of FCC Catalyst Development," 2014 Cat Cracker Seminar, Aug. 19-20, 2014, Houston.
The panel
Zach Bezon, process engineer, United Refining Co.
Luis Bougrat, FCC technology specialist, UOP LLC
Phillip Niccum, senior vice-president of process engineering, KP Engineering LP
Eric Thraen, system technical lead, Flint Hills Resources LP
George Yaluris, North American technical services manager, Albemarle Corp.
The respondents
Rick Fisher, Johnson Matthey PLC
Zhang Jiushun, Sinopec Research Institute of Petroleum Processing
Adam Kasle, BP PLC
Alexis Shackleford, BASF Corp.
Yorklin Yang, BASF
