May-2014

Shale crudes and FCC: a mismatch from heaven?

Weighing the effects that the processing of shale oils has on FCC unit operation, such as iron poisoning of FCC catalysts.

Bart de Graaf, Yali Tang, Jeff Oberlin and Paul Diddams
Johnson Matthey Process Technologies, IntercatJM

Article Summary

Shale oils and gas have had a major impact on the US refining landscape. The US overtook Russia as the world’s largest oil and gas producer in the summer of 2013.1 The oil and gas infrastructure has been turned upside down. Pipelines designed to provide imported or offshore crude to inland refineries are now tasked with transporting shale oil crudes to Gulf Coast refineries, where some refineries are now exclusively processing shale oil rather than conventional crudes. LNG import terminals are now being evaluated for conversion to export.

Light feeds
With increasingly tighter air emission specifications, a substantial segment of US refiners are pre-hydrotreating feedstocks to remove contaminant species such as sulphur and nitrogen. This facilitates the FCC operating window with respect to environmental requirements. Processing very light feeds is a well-known problem in FCC operations. The main challenges in processing such light feeds concern making sufficient coke to properly manage the heat balance. The usual response is to use high activity catalysts and relatively high catalyst addition rates to provide sufficient delta coke, which is crucial in this operation to maintain a reasonable minimum regenerator temperature. As such light feeds contain higher levels of hydrogen, hydrogen transfer reactions play a more pronounced role than in FCC units processing standard VGO feeds. This boosts gasoline selectivity, but is detrimental for octane. Lower rare earth catalysts can help in this instance (with some increase in catalyst addition rate) if there is sufficient wet gas compressor capacity. Another effective alternative is octane selective additives that do not boost LPG yields beyond the wet gas compressor or gas plant limits.

High iron content feed
Many shale oils contain much higher levels of iron (Fe) than conventional crudes. In addition, other contaminant metals that promote the detrimental effects of iron (such as calcium, sodium and potassium) are often present at elevated levels. Feeds with high iron content create a number of problems for refiners, but, again, these are well known but challenging to deal with. Iron poisoning of FCC catalysts becomes readily apparent: slurry yield increases, SOx often increases, apparent bulk (ABD) decreases and sometimes fluidisation becomes challenging. Frequently, however, catalyst activity does not suffer, nor does its surface area (SA).

In the case of iron poisoning, activity and SA tests are subject to artefacts that do not reveal that the catalyst is nearly rendered inactive in the FCC unit when the contact time with feed is short (contrary to the activity test unit). In the FCC unit, the feed can no longer penetrate into the catalyst effectively because the surface is smothered by an iron-rich contaminated layer. In extreme cases, the catalyst becomes covered with nodules (Figures 1-4).

The iron does not penetrate the entire catalyst, but forms a low melting point eutectic with silica, calcium, sodium and so on, and forms a barrier layer over the external surface of the catalyst particles. The inner core of the catalyst remains very active and retains a high SA (as testing confirms) but is closed for catalytic conversion of feed. Fresh catalysts contain some Fe associated with their kaolin filler component (typically around 0.3 wt% as Fe2O3).

Iron poisoning can occur at as low as 0.3 wt% added iron (Fe from the feed and not that already present on the fresh catalyst), and pretty much all catalysts clearly exhibit iron poisoning symptoms at levels over 1 wt% total iron (0.7 wt% added Fe). It is worthwhile to note that the damaging iron is highly dispersed feed iron (organic Fe) deposited on FCC catalyst rather than rust particles, which have minimal effect on FCC catalyst performance.
 
Current solutions for iron poisoning
The most common solution for iron poisoning is to increase the fresh catalyst make-up rate and, in more extreme cases, add substantial quantities of equilibrium catalyst (e-cat), 50 to 100% of the base catalyst, in order to dilute the iron by flushing it out of the FCC unit. This is a fairly effective solution to lower the average iron level on catalyst particles and improve activity in the FCC unit. A major drawback with this solution is that the e-cat has very different properties compared with the catalyst that has been carefully tailored for optimal selectivities in this particular unit and application.

Clean e-cat (in other words, low in metals) generally does not contain metal traps (as it was designed for clean feed applications), is different in the zeolite-to-matrix ratio, has a different rare earth concentration on zeolite, and so on. However, these drawbacks are often a small price to pay when compared to the detrimental effects of iron poisoning. When a unit suffers from iron poisoning, or if high iron feeds are anticipated, catalyst suppliers usually offer catalysts with a higher mesoporous alumina-rich matrix content. Appropriately structured mesopores are not so easily smothered as standard FCC catalyst mesopores. This solution can postpone the effects of iron poisoning, but cannot completely prevent it and dilution (high catalyst addition rate and/or use of e-cat) will still be required.

Iron poisoning mechanisms revisited
Metal poisoning of FCC catalyst occurs via a number of mechanisms:
Vanadium is known to form a low melting point eutectic with silica, thereby ‘dissolving’ the zeolite component. Increasing the amount of alumina matrix, use of high rare earth stabilised Y-sieve or use of a (separate) metal trap are the most effective solutions to counter vanadium poisoning. The metal trap can be inter- or intra-particle, as vanadium is highly mobile in the FCC regenerator. Nickel poisoning works via a different mechanism.

Nickel (and vanadium) promotes dehydrogenation reactions in the riser/reactor, forming hydrogen, dry gas and coke. The introduction of high crystalline boehmite aluminas by Katalystiks in the late 1980s was a major step in the alleviation of this problem.

Nickel behaves differently from vanadium; it is virtually immobile in the FCC unit. Metal traps therefore need to be situated where the poisoning occurs: in the FCC catalyst particle itself. Antimony (and bismuth in the distant past) has been used to counter the dehydrogenation promoting effects of nickel. Antimony forms an alloy with nickel, changing its catalytic characteristics dramatically. Antimony is highly mobile in the FCC unit until it sticks to fresh nickel, effectively poisoning/passivating it (nickel mobility is not a requirement for nickel passivation to take effect).