[Paul R Orlowski]

I would classify an FCCU or an RFCCU as under-utilized with two very different and distinct definitions. The first definition of an under-utilized FCCU/RFCCU is based upon the unit as a stand-alone entity. I do not necessarily consider the unit under-utilized if it is not at maximum feed throughput/rate or at maximum conversion. The true test of the definition for an under-utilized unit is if the unit is constrained? Is the unit operating or being allowed to operate within parameters within the control of the FCCU/RFCCU operations and/or technical team? Is the unit up against one or more major constraints? Is the unit limited by: main air blower (MAB); wet gas compressor (WGC); hydraulically on catalyst circulation or slide valve; main fractionator bottoms (MFB) heat removal capacity; main fractionator overhead (MFO) cooling/condensing capacity; hydraulically in the vapor recovery unit (VRU) area? Is the unit constrained singularly or with multiple constraints?

If the unit is not constrained, then the unit can be considered under-utilized unless it comes into conflict with the second definition. The second definition is if the refinery is constrained in such a manner that the FCCU/RFCCU is not operating at one or more of its constraints. If the refinery is operating in such a manner that it is uneconomical or unable to push the FCCU/RFCCU to a constraint, then I do not “punish” the FCCU/RFCCU, as this condition is beyond the control on the FCCU/RFCCU operations and/or technical team. There have been many circumstances where the refinery has cut crude runs either for economical reasons or due to one or more processing unit problems or issues. Anything that is “beyond the control” of the FCCU/RFCCU team should not count against them.

It is the duty of the FCCU/RFCCU operations and technical team to always look for ideas/solutions in order to maximize the profitability of their unit; which usually means push the unit to multiple constraints. However, the team also needs to understand the place that their unit has in the over-all scheme of the refinery. There may be times when the FCCU/RFCCU may need to be under-utilized for the greater economic good of the entire refining complex.

Posted for K. P.

[Paul R Orlowski]

In many respects, what defines under-utilization applies to when FCC feedrates are down or when operations are normal; the difference is a matter of degree. In general, under-utilization can be defined whenever:

– Pumps & compressors are unloaded (feed, liquid & vapor products, combustion air)
– Catalyst loading equipment has unused capacity
– Emissions control equipment is unloaded
– Regenerated/spent slide or plug valve DPs are exceptionally high, or
– The unit pressure balance has wider rangeability.
Where to look for profit improvement? Three places should be evaluated:

1) The sources and types of feed components that are cracked
2) The technology, formulation and blend of catalyst and additives cracking the feed
3) The operation of the cracking, combustion, flue gas and product recovery equipment.
Let’s discuss the opportunities that come with each of these as well as potential obstacles hindering your success in pursuing them.

Feed Opportunities

Crudes or purchased feedstocks with lower qualities than what are normally cracked offer the opportunity to improve profitability. Higher sulfur, acid number, carbon, and/or metal feeds can be economically attractive while being processed at levels that minimize your risk. Conversely, some refiners introduce feedstocks with better than normal quality in order to fill out their recovery sections while cracking less feed. Internally to the refinery there are incremental feedstocks, like atmospheric or vacuum resid, that can be cracked to raise overall volume gain for the facility. Also, FCC products can be recycled to obtain more valuable yield distributions, like gasoline re-cracked to generate more olefins or heavy cycle oil re-cracked to more valuable liquid products.

Be mindful that an opportunity feedstock may be less crackable than what its reported bulk properties suggest. Excessive yields of fuel gas and coke could result, negatively impacting the heat balance and reducing the feedstock profitability. If the hydrothermal stability of the catalyst is insufficient, then the catalyst may not be able to tolerance additional metals. Additional or new process chemicals for mitigating fouling or corrosion may be needed when cracking more difficult feedstocks.

Catalyst Opportunities

Optimization of catalyst selectivities, activity or both is worth reviewing with your supplier. Through reformulation or blending ratio changes, the yield structure or metals tolerance can be shifted in a favorable direction. Introduce cracking additive technology (e.g. ZSM-5, bottoms reducer) to shift the yield structure quicker or respond to feed quality changes better. Consider trialing new catalyst or catalyst additive technologies to pursue further product distribution improvement. Lowering catalyst additions would reduce your expenses. Raising catalyst additions may improve cracking selectivities. Trial each to determine what fits your operation better. Also consider adding equilibrium catalyst on a regular basis to reduce your expenses.

Adjusting catalyst technology or formulation could result in physical property changes of the circulating inventory. Monitor the fluidization parameters since fines retention or catalyst attrition may be less. If catalyst additions are reduced, monitor the particle size distribution (PSD) of the inventory…you may have to adjust the fresh catalyst size grade. Standpipes may be over- or under-aerated depending on the operating conditions…adjust aeration rates as necessary for proper pressure build. Closely review the properties of any equilibrium catalyst under consideration for purchase…incompatibility with the fresh catalyst technology could lead to under-performance.

Equipment Operation Opportunities

It is likely that the reduced rate operation will free up combustion air. This opens up the opportunity to operate differently and consume the unused coke burn capacity. Pertaining to the riser operation, lowering or removing feed preheat will increase catalyst circulation, conversion and volume gain. Feed dispersion steam can be conserved to reduce sour water production, or increasing the steam/feed ratio may actually improve feed/catalyst contacting. Being too aggressive with dispersion steam conservation can lead to poor feedstock atomization, feed nozzle plugging, or “wetted” spent catalyst going to the regenerator. Conduct steam/feed trials to determine what works better. Also consider taking feed nozzles out-of-service to improve overall atomization – review the design and procedures before trying this.

By introducing or increasing riser lift gas you may reduce the net dry gas production by “conditioning” the metals on the regenerated catalyst. You can also introduce or increase cracked naphtha injection to promote light olefin production. Under reduced feed rate conditions there will be more riser contact time, which should promote bottoms cracking. You may realize higher slurry ash content as a result, which may impact product blending or increase the erosion rate of system piping. Also, the riser velocity may get too low, which promotes backmixing, resulting in higher fuel gas and coke yields, gasoline overcracking to LPG and other selectivity shifts.

In the reactor section, with the air available, reactor temperature can be raised for more conversion and volume gain. Lowering the reactor pressure will result in less hydrogen transfer (better octane, olefin yield), better stripping, and reduced rotating equipment cost. Longer reactor residence time could also lead to product mix deterioration and vessel coking. Potentially higher butadiene yield may impact the alkylation plant. Slurry fouling rate may accelerate from higher conversion operations – you may need an antifoulant program. LCO properties will shift and may impact distillate blending for cetane, sulfur, and gravity.

In the stripper section, longer stripper residence time (from lower catalyst circulation) can lead to better stripping efficiency, product recovery, and lower coke hydrogen content, especially if reactor temperature is increased. Conserving stripping steam may help with sour water management but be careful of going too low. Like the feed nozzles, getting too aggressive with stripping steam conservation can result in steam distributor nozzle plugging and lower actual stripper bed level, which may uncovering cyclone diplegs. Operating at lower feed rate can also result in higher density spent catalyst, leading to higher slide/plug valve differentials – this presents the opportunity to shift the pressure balance in a positive way by lowering the reactor pressure.

In the product recovery section, generally the lower feed rate condition opens up wet gas capacity. This means you can handle lower suction pressure or lower molecular weight gas streams – this flexibility may expand your feedstock and catalyst options. Downstream recovery sections are likely to be under-loaded, resulting in better liquid/vapor product separations, product purities and improved treating conditions.

When it comes to the regenerator, combustion air will likely be available. You could continue to conserve in the interest of compressor energy savings or apply the excess for higher coke burns. You could also reduce or eliminate the cost of oxygen enrichment if it is part of your base operation. If the refinery needs steam and you have a catalyst cooler increase the coke burn to produce extra steam. If superficial velocities are down and there is less regenerator afterburn then your feedstock and catalyst options grow. You can also save on combustion promoter which may also result in lower emissions. Too low of a combustion air rate can lead to air distributor nozzle erosion and lower regenerator bed penetration. The burn will become uneven with the radial temperature differences growing. The catalyst regeneration will be less uniform resulting in a “salt & pepper” appearance that could impact the catalyst cracking selectivities. The fluidized density will also vary throughout the bed, potentially impacting the stability of the catalyst circulation returning to the riser. Consider plugging off “extra” air distributor nozzles during the next outage if the operation is expected to last a very long time.

In the flue gas system, turboexpander vibrations from fines deposits could be reduced or eliminated. Less or no walnut shelling would be required. However, turboexpander power generation may be lost if the regenerator pressure is lower than the normal operation. Fuel to fired boilers could be reduced or burners modified to reduce emissions. Lower fines accumulation on boiler internals may mean less opacity spiking during sootblowing cycles. Lower chemical cost would be expected for NOx and SOx reduction equipment technologies.

Realize that lower vapor and flue gas rates may result in the loss of cyclone efficiency and less reactor/regenerator catalyst retention. Larger particle sizes may be preferentially lost. Consider sealing off “unnecessary” cyclone pairs during an outage if efficiencies are expected to be low for a very long time. Lowering operating pressure should help in this situation. High slide/plug valve differentials may also lead to high valve erosion rates. See if the pressure balance could be adjusted to address this. Process control valves may perform with less stability since they may be operating near the low end of their range. Retuning may be necessary.

Additional Opportunities

Consider riser, reactor, stripper, recovery, regenerator and flue gas system modifications. Alter or upgrade equipment to address plant constraints and reliability issues. Add new equipment/technology to expand processing capabilities and maximize profitability under reduced rate conditions.

Lastly, take the opportunity to improve your LP vectors for better representations of feed quality and operating condition changes. Also develop various business scenarios of interest and assemble the model projections for each. Conduct plant trials that simulate the cases and provide the necessary data. Update the LP as necessary.

Posted for B. L.

[Paul R Orlowski]

There have been circumstances where catalyst losses have been very severe and a shutdown was imminent. However, it is very unusual for the cause of an immediate/emergency shutdown to be cyclone damage due to normal “wear-and-tear.” The normal “wear-and-tear” damage can usually be observed, monitored, and anticipated well in advance by operations and technical personnel, thus avoiding an emergency outage; a good monitoring program will allow operations and technical personnel to work with planning and scheduling to plan an outage as the damage and losses become unsustainable.

The rapid and imminent causes are usually the result of a significant event or a catastrophic failure. The usual causes are refractory failure/loss following an upset or a thermal cycle which plugs or restricts one or more cyclone diplegs leading to massive catalyst carry-over. A second example is the use of “smear-coating” or “butter-coating” the refractory in the cyclones. “Smear-coats” or “butter-coats” do not adhere to the base refractory and spall off almost immediately upon start-up. Since the secondary cyclones are relatively small in diameter to most units (usually 8-10 inches), a major spall of this type can easily plug (and has) the dipleg. An example unique for the reactor side, would be coke spalling from reactor cyclone gas outlet tubes following a thermal cycle and again plugging or restricting the cyclone dipleg. These situations can be observed in several manners:

– A rapid loss in catalyst level or inventory, usually in the regenerator as the reactor/stripper level is typically in level control and the regenerator level/inventory “takes the swing”
– Significant increase in the catalyst content in the main fractionator bottoms (MFB) slurry loop sample (if it is a reactor cyclone)
– Significant increase in the catalyst content in the flue gas scrubber (FGS) effluent loop sample (if it is a regenerator cyclone)
– Significant increase in the catalyst being dumped to the electrostatic precipitator (ESP) bins (if it is a regenerator cyclone)
– Significant increase in flue gas stack opacity if there is an electrostatic precipitator (ESP) in service (if it is a regenerator cyclone).

The failure mentioned previously may not be easily observed upon the restart / dry-circulation period when catalyst circulation rates are generally very low and the catalyst loadings to the cyclones are extremely low (generally less than 10% of operating catalyst circulation rates). The situation will manifest itself once feed has been introduced into the unit and catalyst circulation rates, and thus catalyst loadings to the cyclones, are increased.

A more unusual, but still possible (possible because it has occurred) circumstance is the loss of a cyclone. A loss in this case means the cyclone failed or dropped from its supports. This can happen following an extreme thermal excursion (usually in the regenerator) or following some seismic activity for either or both the reactor and/or the regenerator sides (usually discovered upon a re-start as the seismic activity most likely took the unit off-line). Again, the same five previously mentioned items will be key indicators of a problem or issue. An FCCU/RFCCU process engineer should know the following critical pieces of information for their unit:

– At what rate would catalyst be lost from the unit if a regenerator primary cyclone were to plug?
– At normal catalyst circulation rates.
– At minimum feed rate catalyst circulation rates (minimum feed rate is usually the point required for main fractionator operational and product yield stability).
– At what rate would catalyst be lost from the unit if a regenerator secondary cyclone were to plug?
– At normal catalyst circulation rates.
– At minimum feed rate catalyst circulation rates.
– At what rate would catalyst be lost from the unit if a reactor primary/rough-cut cyclone were to plug?
– At normal catalyst circulation rates.
– At minimum feed rate catalyst circulation rates (minimum feed rate is usually the point required for main fractionator operational and product yield stability).
– At what rate would catalyst be lost from the unit if a reactor secondary cyclone were to plug?
– At normal catalyst circulation rates?
– At minimum feed rate catalyst circulation rates.

The ability of process personnel to quickly analyze and respond to any of these is the difference between a short five-to-eight day disruption or two-to-three week outage.

Posted for K.P.

[Paul R Orlowski]

There are a number of scenarios that can lead to a FCC unit shutdown due to excessive catalyst losses. Generally, inability to maintain regenerator bed level leads to an imminent unit shutdown. Since the stripper bed level is controlled with the spent catalyst valve any catalyst loss from the reactor or the regenerator is reflected in a loss of regenerator level. Refiners will face the decision to shut down whenever the catalyst loss is so high that:

– Not enough make-up catalyst (fresh or equilibrium) is available to restore the catalyst lost
– Catalyst loading via loader or pressuring from the hopper falls short of the required make-up rate
– Cost of replacing the lost catalyst becomes prohibitive
– Slurry oil ash or BS&W specifications cannot be met, leading to severe product discounts and loss of sales
– Wet gas scrubber (WGS) purge solids separation/containment inventory is inadequate
– Electrostatic precipitator fines collection rate exceeds the number of rolloff bins available
– Stack opacity/particulate emission limit compliance is difficult to achieve
– Slurry oil pump reliability and mechanical availability is unacceptable leading to significant feedrate reductions.

There may be opportunities to mitigate the catalyst loss and delaying the need for shutdown. The delay would allow the refiner to troubleshoot and develop corrective action plans for staying on-line and keeping operating costs down.

The refiner could consider:

– Pressure bumping to possibly dislodge an obstruction or aerate a de-fluidized zone
– Adjusting the fresh catalyst attrition resistance or fines content
– Making up with equilibrium catalyst versus fresh catalyst
– Adjusting air/steam rates, as well as the distribution if multiple grids/rings are available
– Lowering/raising bed levels to check impact on cyclone operation
– Lowering/raising operating temperature to lower/raise velocities
– Lowering/raising operating pressure to raise/lower velocities
– Conducting diagnostic studies to help identify what equipment needs repair.
posted for B.L.

[Paul R Orlowski]

Salvatore Mannello says
I totally agree with Michael about regen operation. I can add that if the RCSP is over fluidized, try to decrease fluidization medium; depending from the unit type, e.g., you can increase regenerator cat level, or if the combustion air is not proper distributed, if you can, try to even distribute. Moreover, once all the CO2 is stripped from the deetanizer, try to not absorb it again in the recontacting 2°stag high pressure accumulator and/or primary absorber. E.g., if the propane-propylene recovery from the fuel gas is well accepted try to increase the temperature around the 2° stg HP drum and primary absorber, or decrease pressure. In amine absorber system the grade of the CO2 and COS absorption from the LPG depends from the type of the solvent used (MEA, DEA, MDEA…); in any case can help: increasing the amine circulation and/or the concentration; decreasing the amine lean loading and the temperature; specialty amines can help COS absorption; the last option is molecular sieves.

[Paul R Orlowski]

Michael Edwards says:
The CO2 does not come from the reactor itself. The regen catalyst is bathed in flue gas components as it enters the regen standpipe, and even full combustion regen operation generates more CO2 (AND CO) in the few seconds the catalyst flows to the riser. Note, most regenerators fluidize the catalyst with air next to the standpipe, so flue gas and air enter the pipe with the regen catalyst. You will see both O2 and CO in reactor fuel gas, though there should be no CO in full regen, and O2 in fuel gas, though there should be no O2 in partial combustion. Some few units do fluidise the regen cat near the standpipe entry or in separate steam stripped withdrawal well next to regen. There, you would increase the stripping steam. More stripping steam in reactor does not affect amount of CO2 (essentially all flue gas components are already stripped), but can increase amount of LPG recovered, so reduce % flue gas in LPG and fuel gas.

[Paul R Orlowski]

Solly Ismail
Salvatore, While you make an excellent point, UOP K also has its limitations. The 90% point is very important. It only shows up indirectly in the UOP K factor. In fact most heavy feed stocks do not vaporise and that s why all the Ni is deposited on the outside edge of the catalyst.

Catalyst companies use smoke and mirrors techniques by telling refineries they have large pore which allows the heavy portion to diffuse into the interior. Then why is all the nickel deposit on the outside edges. The liquid portion of heavy is drawn into only a short distance into the catalyst by capillary action.

I agree the UOP is important but so is the 90% point.

The ultimate driver of catalytic cracking in FCC is the hydrogen content in the feed. High UOP K means higher higher hydrogen. Higher hydrogen is found in paraffinic stocks. These are easily crackable. The worst is aromatics. Aromatics stocks are very stable and do not crack under FCC conditions. They need high H2 partial pressure.

———————-
Salvatore Mannello
For straight run Vacuum Gasoil or HDT feed, density and UOP K tell just enough for the first approach to a FCC feed quality. If in the feed mix there is also some Residue, the simulated distillation is more suitable for the VABP (and CABP) calculation in the UOP K: refineries now have this approach, being D-86 and D-1160 distillation not always applicable, as Solly said. Aniline point analysis is good, fast and reliable, for feed characterization, but for dark feeds, sample needs dilution and operator high attention; nowadays some refineries don’t run any more. Refractive index is among the most used for FCC feed characterizations: needs four decimal digits. There are also other indirect indications: hydrogen content is one, and it is excellent; it is very good to trend in an every single refinery, unless we are in agreement on the calculation method (Total?).
Talking about FCC unit optimization first we need at least this information: feed quality, objectives and unit type hardware.

———————–
Solly Ismail
Excellent Salvatore! You have nailed it. The most important by far for optimizing the FCC unit is unit is the Feed Quality. Then comes operation conditions and lastly catalyst selection.

While catalyst selection is important, make no mistake, it is more important what is going into the unit, than from whom it is supplied. Catalyst quality supplied to refineries varies greatly and quality supplied is not always what was initially agreed on. For example, there was long standing anecdote, that FCC catalyst supplied by the Chinese was inferior. Then during the Rare Earth Crises during 2010 to 2012 , several refineries bought catalyst from the Chinese. When we analyzed the FCC catalyst from the Chinese was excellent! It had low attrition, comparable selectivity, and much lower cost.

Catalyst suppliers often say each FCC is unique and then go on to say how good their catalyst in other refineries, meaning that it will just as well (FCC unique!) in yours.

Does anyone know of objective technical analysis provided by catalyst company A which states that FCC catalyst from company B is better suited for the given unique operation?

————————
Paul Orlowski
Hoekstra Trading offers independent catalyst testing according to this presentation: http://refiningcommunity.local/presentation/pushing-the-limits-of-fcc-gasoline-desulfurization/. For more FCCU presentations on catalyst go here http://refiningcommunity.local/past-presentations/ and type in catalyst

[Paul R Orlowski]

Salvatore Mannello:
We need more info: type of feedstock, product targets, type of unit. If the feed has a low UOP K Factor (very hard to crack) and unit targets are max gasoline/max octane barrel and LPG, and the reactor cyclones system is a short contact time, the multi-feed atomizing system is very well performing (properly steam/feed ratio, good pressure drop, nozzles tips in good shape…), the cat/oil presumably high (high mix zone temp/high feed percent vaporization=avoid coke formation) and the main fractionator bottom residence time short, the said parameters appears adequate. But if the feedstock has a high UOP Kfactor (very crackable), operating the unit with the above parameters, likely, everything is converted: bottom product probably disappear, gasoline decrease dramatically/LPG increase consequently, coke and gas increase significantly.

Maybe your situation is similar to not equal to the first, so I agree with your consultant: run tests to optimize your unit, according to your objectives.

[Paul R Orlowski]

Solly Ismail:
I am retired and have no skin in this game. So what I am writing is I hope to help improve the area of FCC knowledge. It is cynical but unfortunately true.

Most refineries buy or change catalyst based on marketing hype and simplistic technical evaluations provided by suppliers. Yet, suppliers tell FCC operators that each FCC is unique.

Simple questions regarding asymmetry of information are often not raised.

An ISO 9000 certificate of catalyst quality supplied to the refinery is generally not asked. Nor are third party evaluations carried out.

Many refineries are charged highly specialized catalyst prices for formulations which are in fact commodities. Catalyst performances should not be based on lunches/ superficial technical presentations in fancy hotels but hard technical evaluations.

A more discerning critical supply chain/technical expertise can increase FCC profitability and throw light of areas which really need focus.

[Paul R Orlowski]

Nicolas Lambert:
And I concur with Mike too. Directionally, we can understand and explain the benefits of increasing the ROT, decreasing the Feed Temperature and so on. Does that make sense? Impossible to tell as there are as many situations as there are FCC units…

In case a 555°C ROT makes sense (hum… unless you are targeting real high amounts of light olefins, I doubt it), we, licensors, have process solutions to mitigate the formation of diolefins. Similarly, in order to boost the C/O, there are other solutions which do not come at the expense of feed vaportisation (beware of too low temperatures!).

[Paul R Orlowski]

Michael Edwards:
No offense to anyone. Without any information about the current limits on unit, the catalyst (or additives, if used), the goals (max olefins, max conversion, max gasoline and/or octane) this is nonsense. That aside, if they are at wet gas limit, of course reduce RTT (RxT), and allow either more conversion or more feed (the latter being generally more profit). If at coke limit, push up feed preheat to allow more feed rate. Rule of thumb, more feed is more money. If at high RTT because gasoline octane, use additives and reduce RxT. Simple steps. But without details of the situation, rest is pure BS.

[Paul R Orlowski]

Neal Cammy:
With a 180C preheat temperature, you are also in danger of condensing the feed atomization steam, potentially causing poor atomization (in addition to the comments regarding viscosity).

[Paul R Orlowski]

Nashika Maharaj:
Reactor temperature is indeed very high so if there is capacity on gascon section that’s a positive. We operate at 515degC at low rates and are typically constrained on butadiene quality (max of 0.5 for Alkylation unit) and Dry gas make(285tpd). Feed temperature is very low so monitor riser coking and mix zone temperature to ensure feed is fully vaporized. Suggest you also review catalyst formulation or use of additives to maximize margin on FCC.

Also monitor tray loading on Gascon columns(do you have a simsci model for this?) in order to simulate how the operate will operate at these conditions suggest you run the specific feed quality at these conditions on an FCC model and check what constraint you reach based on your unit configuration and downstream unit capacities

[Paul R Orlowski]

Erwin Rommel:
The riser temp does seem a bit high, increase riser increase dry gas production. Bottom temp could increase coke in the bottom of the tower by the time you are at the end of run. The preheat that low, I’ve seen lower, could cause you to reach max cat circulation.

[Paul R Orlowski]

Zachary Ballard
Great points Alan, we walnut shell nightly now and vary the size of this medium. We also complete an expander spall usually every 6 to 8 months.

Douglas Stieb, PMP
I’ve seen walnut shelling as often as every shift. Might also consider the activities involved to bypass the expander during an FCC run. If you envision bypassing fairly often, to service the expander, could install additional equipment to isolate and blank more efficiently – and safer to personnel installing the blanks.

Shobhit Awasthi
As per my little experience with turbo expander , TSS efficiency in view of fines to be taken care, try to operate expander on a constant load and periodic thermal cycling also helps
Regards

Paul Orlowski
Makes sense Douglas Stieb, PMP to install additional isolation valves if expander is frequently bypassed for service.

Mauro Natalini
if interessed on TE Isolation valves to reduce downtime , IMI Remosa has already installed the right systems (Not only special valves) and we have the technology and experiences to support FCC Customers around the World.
Special patented automated Isolation valve can be used to reduce time to blink improving the safety of operators. more information available on demand.

Mohammed Abdulrahman
Are you using walnut shell injection? Do you monitor efficiency drop? Do you take frequent photos from the port to see how deposits looks like? Any use of SOx abatement additives?

Naveen Dimri
Review the internal flow path by doing CFD. Also you need to know if deposition is on blade or on shroud for proper root cause analysis.

[Author]

Posts